Method and system to detect colour from image

FIELD: measurement equipment.

SUBSTANCE: invention relates to the method to detect colour of colour sample from an image, for instance, when selecting dye colour. A colour sample capture card is proposed with printed colour samples of known colour (for instance, XYZ-coordinates of colour), the image of the test colour sample is captured using domestic equipment with a digital camera. The image is then sent to the remote service of colour detection to detect colour of the colour sample, and regression analysis is used, which uses samples of RGB-colour in the image and its available XYZ-colours for characterisation of response of colour capture of the colour capture device with account of changes of spatial brightness by the image. On the basis of characterisation of the image capture device the XYZ-colour of unknown colour sample may be determined from its RGB-colour in the image.

EFFECT: provision of accurate identification of colour of an unknown colour sample and reliable calibration under conditions of varied illumination.

13 cl, 21 dwg

 

The technical field to which the invention relates

Embodiments of the invention relate to a method and system for determining the color of color sample from an image of the color sample.

Art

When choosing paint colors for the redecoration of the room it often happens that the customer wishes to pick up the paint color to the color of a particular item contained in the room, such as a piece of furniture, or decorative fabric, such as sofa cushions, sofas, curtains, etc., paint Manufacturers usually provide a large color palette, and detailed colour images are provided by the retailers of paint, that customers have choice of color. Card color swatches are available for the user took them home and picked up the object that needs to be color matched. However, usually this requires a visit to a customer in a retail store paints, collecting cards colors, capture cards flowers home and then attempt to compare the color samples colors on the maps with the select object's color. The customer must then return to the store, usually buy color samples of paint, to return home, to use the color samples of paint and then finally make a purchase decision. In addition, such conventional techniques are based on individual perception W�of Cascina, what is the most matching paint color However, it is well known that the perception of colors varies considerably from person to person, so the color combined with the sample, selected by one person, does not seem to be matching color for the other person.

One potential solution to this problem is to try color matching electronically, using a digital image. In this respect at the present time, home users generally have at their disposal numerous digital devices capture (shooting) of the image in a digital camera or a camera-equipped mobile phones.

However, the characteristics of the capture color common household devices capture images, such as digital cameras, mobile phones, etc., vary considerably from device to device and is therefore generally impossible full-time capturing of color. Available special devices spectrophotometers that can accurately measure the color, but they are not affordable for most home customers. Common household device taking the picture shoots image and represent a color using RGB values (red, green, blue) pixels. Typically use 16 bit or 24-bit RGB. When values are used with 16 bits, each of the red and blue channels are usually�about has five bits, associated with them, whereas the green channel has six bits associated with it. In this respect, the human eye is more sensitive to green color than red and blue colors, and therefore it is found more green. When using 24-bit color, then it is equal to eight bits, or 256 colors per color channel.

However, due to the aforementioned differences in the devices taking the picture for accurate capture of colors and also in playback devices, such as monitors, etc., when playing a RGB-color values are not considered as reference values. Instead, there are fixed standards that define color by the International Commission on illumination (CIE), such as the coordinates X, Y, Z CIE color or so-called CIELAB values (L*, a*, b*). The CIELAB values associated with coordinates X, Y, Z color, using a known mathematical formula. XYZ-coordinates of the colors associated with the wavelengths present in a specific color.

Prior art

The issue of calibrating the device for shooting an image by associating RGB values obtained when shooting, with reference values, such as XYZ color coordinates, or CIELAB values, was previously proposed in US 5150199 and WO 01/25737.

More specifically, US 5150199 (Megatronics, Inc. describes how to transform or correlation numeric RGB values, created by different instruments, in a reference color coordinates. In this respect, the iterative regression analysis to determine the functions that convert the RGB values generated by the video camera, from the original colors in the reference XYZ color coordinates. Then used regression analysis to determine additional functions that convert the RGB values generated by the video camera representing additional colors different from original colors in the reference XYZ values. The functions generated for the camera, then used to transform the RGB values generated by the video camera when the image formation of a color object in the reference XYZ values.

More specifically, in the US 5150199 as RGB values and XYZ values are determined from a set of color swatches. RGB values are determined using a standard video camera and ocifrovivaem equipment that can detect and record the numerical values for the RGB components of each color. The XYZ values of the color swatches are defined by using a conventional colorimeter or spectrophotometer.p>

By capturing this data, as a first stage in the analysis, iterative regression analysis is performed to find X as a function of R, Y as a function of G and Z as a function of B. e�from the regression analysis uses the so-called values "gray scale" in the color swatches, where R-, G - and B-values are approximately equal. The resulting functions are power functions. Then at stage 2 is performed multivariate analysis of power functions, defining functions that associate each of X, Y and Z individually with all of R, G and B. In US 5150199 also describes another method which adapts the function Y as a function of the saturation of the red color, although it is not appropriate in this document.

Thus, US 5150199 describes a basic method for the characterization of the transfer function of the capture color capture device image to RGB values captured by the device, to convert to XYZ color coordinates. However, as noted, to use the device for US 5150199 to determine the characteristics of the captured image, the user must have access to a colorimeter or spectrophotometer for color measurement of color swatches, which are also characterized by the capture device image. Usually when you use scenarios presented in General terms above in the prior art, the user does not have access to special equipment, such as colorimeter or spectrophotometer. Consequently, the method according to US 5150199 is mainly experimental.

However, WO 01/25737 partially overcomes these disadvantages US 5150199. WO 01/25737 t�the realm of human agreement describes the captured RGB values with the reference colorimetric data and, in particular, coordination with the CIELAB values. The mathematical analysis described in WO 01/25737, essentially, is the same as that described in US 5150199, although WO 01/25737 introduces the principle of the calibration pattern of known colors, colorimetry data for which are known. Then formed the image of the measured unknown color simultaneously with the calibration pattern. The calibration pattern includes in one example 65 known to colors and in another example 37 known colors distributed in the color space. By capturing the RGB color values of the calibration it is possible to calculate a mathematical model is needed to convert the measured signals are known colors in colorimetric data (for example, CIELab values). If the model is obtained, then the color (CIELab color space) of any unknown colors in the image can be determined from its RGB values.

WO 01/25737 describes that the image of the color sample to be detected is captured simultaneously with the image of the calibration pattern using, for example, a flatbed scanner or digital camera. The captured image is then processed to determine the unknown colors in the image. The device is described as particularly useful in the industry of car repairs. In this respect, the color of the car, unserviceable, the measured�Xia using electronic devices form the image. Before or simultaneously recording the panel, which were caused by different color calibration. Colorimetric data of color of the car then calculated, and is then color formula, which gives a color identical to the color of the car to be repaired. The color formula is prepared in a dosing device and then applied.

So WO 01/25737 describes the device to be used in professional situations, such as in automobile repairing, or painting workshops. Essentially, WO 01/25737 not consider issues relating to matters where the lighting changes on the captured image, where the image is not in the correct orientation or where the color sample actually contains other colors, spatial mixed sample. Conversely, in a home situation can be all of these anomalous situations.

Another prior art for the invention includes WO 02/13136, WO 2008/108763 and WO 2004/028144.

A summary of examples of the invention

Embodiments of the invention aim to address some of the aforementioned issues and relate to the definition of the color of color sample from an image of a color sample, and the image is usually (although not exclusively) performed by an unskilled user, and�using non-specialist equipment. In one embodiment, the implementation is provided by capture card color sample having printed on it the color samples of known color (e.g., XYZ color coordinates). The image of the test colour samples are then captured using available home equipment, such as a household digital camera or a camera-equipped mobile phone, and the image also contains a map of the capture of the color sample. In one embodiment, the implementation of the image is then transmitted to the remote service determine the color to determine the color of the color sample. Regression analysis is performed using samples of the RGB colors in the image and is known for its XYZ color for characterization of the response of the capture color capture device image. Based on the characterization of the capture device image XYZ-sample color unknown color can be determined from an RGB color in the image. In one embodiment, the implementation, knowing the XYZ color, the color can then be matched accurately with a palette of paint colors to determine the color of paint to match the unknown color. In addition, we can identify additional colors to the palette of paint.

When performing the above in one embodiment, the implementation can take into account differences in spatial brightness across the image. In another embodiment, the error �of asemenea card in the image is also adjusted before processing, using the elimination of image skew and rotational transformations. In another embodiment of the XYZ color is computed in two passes, using the information from the first pass to inform the second pass. In yet another variant implementation, where the color sample actually contains more than one colors, individual colors are defined using clustering techniques to identify dominant colors in the sample.

With the above first aspect of the invention provides a method containing the first image data relating to an unknown colour samples, colorimetry data for which should be determined; and receiving the second image data belonging to the set of known calibration color samples, colorimetry data for which are already known. Next, define a set of characteristics to the color calibration of linking the measurement of known color calibration color samples from the second image data with the corresponding known colorimetry data of the calibration color samples; and colorimetric data of the color of the unknown sample are calculated depending on their color measurement of the first image data and certain characteristics color calibration. In addition, can also be compensated brightness differences on a set of known�samples on the Internet color calibration. It allows you to capture image data in an uncontrollable lighting, which can be uneven lighting across the image. This allows easy use for the end user.

In this embodiment, the implementation of compensation may include determining the first set of one or more functions having a first set of calibration coefficients, and one or more functions associated measured colours of the known calibration color samples from the second image data to the known colorimetry data of the calibration color samples and the known position of each known sample in the image. Certain functions can then be analyzed to locate the second set of functions having a second set of calibration coefficients. The first and second sets of features, and calibration coefficients are then used when calculating colorimetry data of the unknown sample color.

In this embodiment, the implementation of the analysis may include calculating an intermediate color values for, essentially, each known sample color calibration and then using the calculated intermediate color values to determine a second set of functions having a second set of calibration coefficients.

More specifically, the calculated intermediate color values, vertical�proposes a multidimensional fitting to determine a second set of functions, having a second set of calibration coefficients. Preferably, the multivariate fit is of the form:

In addition, more preferably, the correction further comprises prior to determining the first set of functions, the definition of the preceding set of features with the previous set of calibration coefficients that relate the measured colours of the known calibration color samples from the second image data to the known colorimetry data of the calibration color samples without regard to the provisions known color samples. The previous set of calibration coefficients is then used as part of the first set of calibration coefficients when determining the first set of one or more functions. In one embodiment, the implementation preferably find the previous set of calibration coefficients using only samples of the gray scale.

In one embodiment, the implementation of certain colorimetric data of the color of the unknown sample can then be compared with the color palette of paint colors to identify selected (consistent) paint colors, and information relating to a consistent color paints are presented to the user.

In one embodiment, the implementation of first data and second image data and�of obrazenia are accepted from the remote user through the communication network. In addition, information relating to a consistent color ink, may be provided to the user through the communication network. Thus, consistent color paint color of the unknown sample can be provided using remote services.

In a variant implementation, the first image data and second image data are received as any of: (i) email; (ii) messaging service multimedia messaging service (MMS); and/or (iii) of the image data on a web page. In addition, information relating to a consistent color paint also may be provided in the form of any of: (i) email; (ii) an MMS message; (iii) messages the short message service (SMS) and/or (iv) data on a web page. Such communication protocols contribute to the remote provision of services selection of paint, being known to the users and easy to use.

In one embodiment, the implementation of the first image data and second image data created by the user using a capture device (shooting) of the image; wherein the capture device image is preferably any of: (i) a digital camera; ii) a camera-equipped mobile phone; and/or (iii) a digital video camera recorders. A normal user easily drawn with such equipment, and the user�spruce familiar with the operation of such equipment.

In one embodiment, the implementation of certain colorimetric data and/or known colorimetric data represent the XYZ color coordinates. XYZ color coordinates define a fixed and specific reference color.

In one embodiment, the implementation can decide the colors, additional to the selected color, and information relevant to some specific color, is provided to the user. Through the provision of additional colors are easier can be defined color schemes.

In one embodiment, the implementation of at least the second image data oriented in a known orientation, allowing them to recognize the known samples of color calibration. Automatic orientation of image data makes it possible for easy use for the end user, as there is no need to capture the captured second image data in a particular desired orientation.

In this embodiment, the implementation of the orientation preferably contains an implementation of the detection of the edges to identify the location of the set of known samples color calibration in the second image data. In addition, the orientation may further comprise identification of a plurality of predefined points belonging to the set of�frequented samples color calibration in the second image data. If these known points are identified, it may be a perspective transformation applied to the second image data depending on the location of the identified pixels to correct the tilt of an image from a set of known samples color calibration.

In addition, in this embodiment, the implementation of orientation may further contain identification labels a predetermined angular orientation related to a set of known examples of color calibration in the second image data. The second image data then can be rotated depending on the location of the identified angular orientation of labels, so that the known calibration color samples are placed in a known position in the second image data.

In one embodiment, the characteristics of the color calibration are defined using theNknown calibration color samples, whereNless than the total number of known calibration color samples throughout the color space. In some circumstances it can provide more accurate results.

More preferably, in the above embodiment, the implementation ofNknown calibration color samples representNsamples that are closer in color space to evaluate the color of the unknown sample �Veta. This effectively allows you to perform a "zoom-in" color space in determining the characteristics of the color calibration, so more accurately characterized by a portion of the color space, which contains an unknown color sample.

In the above embodiment, the implementation of the evaluated color can be obtained by determining the first set of calibration characteristics using all available known samples color calibration, and calculating the evaluated colors using the first set of calibration characteristics. Then executed the "second pass" processing, usingNthe nearest known samples color calibration to evaluate color. Thus, the approach is used with two-pass processing, which enables the characterization of the common color space, and then to characterize in more detail a portion of the space containing an unknown color sample, providing more accurate results.

Alternatively,Nknown calibration color samples represent thoseNsamples used in limited color space, which is known to represent the second image data. In this regard, it may be that the known calibration color samples, as known, in a limited part of the color space, for example can be in�e red, or blue. I.e. if you try to pick up the red color, the user uses the known calibration color samples, which are predominantly red or close to red, thus limiting the portion of the color space of the capture device that requires characterization.

In another alternative case,Nknown calibration color samples representNsamples that have a measured color values from the second image data that are most similar to the measured color value of the unknown sample from the first image data. For example, can be usedNknown calibration color samples that have RGB values that are closest to the unknown sample color.

In the above embodiments, theNpreferably is in the range from essentially 5 to, essentially, 250, or more preferably from essentially 10 to, essentially, 100, or more preferably from essentially 20 to, in essence, 85, or more preferably from essentially 30 to, essentially, 70, or more preferably, essentially 40 to, essentially, 60, or most preferably equal to or about 50. In other embodiments, the implementation can then use other numbers or ranges ofN.

In one embodiment, algor�TM clustering can be applied to the pixel values in pixels, representing the unknown colour sample in the first image, to determine the number of colors in the image sample, and the color is identified for each identified cluster. With such a device, when an unknown color sample contains more than one color, then or can be identified the main color, and/or can separately identify individual colors.

In this embodiment, the pixel values are first calibrated using the characteristics of the color calibration. This results in guaranteeing that the clustering algorithm works on real colors in the color sample.

The clustering algorithm in use can then work through: (i) calibration of the average value of the pixels in the cluster; (ii) then determine the number of pixels within a predetermined threshold distance from the average value; and then (iii) increasing the number of clusters, if a certain number of pixels is less than a predetermined fraction of the number of pixels in the first image data relating to the unknown sample. Thus, it becomes possible to identify different colors in the sample, each identified cluster refers to the respective individual color.

To ensure that the detected dominant or important CEE�and in the sample, implementation option can also filter the clusters to remove those clusters from consideration that do not contain a threshold number of pixels within the second threshold distance from the average value of the cluster. Consequently, clusters of flowers with only a small number of pixels are not identified as prevalent or important colors in the sample.

In a second aspect, the present invention additionally provides a device comprising at least one processor; and at least one memory including computer program code, and at least one memory and computer program code are performed so that cause the executing device using at least one processor of at least the following: receiving first image data relating to an unknown colour samples, colorimetry data for which is to be determined; receiving second image data relating to the set of known calibration color samples, colorimetry data for which are already known; definition of many of the characteristics of the color calibration linking of measurement known color calibration color samples from the second image data with the corresponding known colorimetry data of the calibration color samples; computing�of colorimetric data of the unknown sample color depending on the color measurement of the first image data and certain characteristics color calibration; wherein the device is characterized in that the at least one memory and computer program code are made so that they cause the payment device using at least one processor of brightness differences on a set of known samples color calibration in determining many characteristics color calibration.

Other aspects and features of the present invention will be apparent from the accompanying claims.

Brief description of the drawings

Other features and advantages of examples of the invention will become apparent from the following description of specific embodiments of the invention presented only as an example and by reference to the attached drawings, in which similar positions refer to similar parts and in which:

Fig.1 is a block diagram of a system according to the embodiment of the invention;

Fig.2 is a drawing card samples the color calibration used in the embodiment of the invention;

Fig.3 is a block diagram of the sequence of operations of a process performed in the embodiment of the invention;

Fig.4 is a block diagram of the sequence of operations and associated drawings, illustrating the orientation of the images used in a variant implementation�ia of the invention;

Fig.5 is a block diagram of the sequence of operations describing the calibration process colors used in the embodiment of the invention;

Fig.6 is a block diagram of the sequence of operations illustrating a multi-pass process used in an embodiment of the invention;

Fig.7 is a block diagram of the sequence of operations illustrating a part of the calibration process of the spatial brightness used in the embodiment of the invention;

Fig.8 is a block diagram of the sequence of operations illustrating the clustering process used in an embodiment of the invention;

Fig.9 is a drawing showing the use of the clustering process used in an embodiment of the invention;

Fig.10 is another drawing illustrating the use of the clustering process used in an embodiment of the invention;

Fig.11 is a photograph of the experimental sample template color calibration used for the test embodiment of the invention;

Fig.12 is a graph depicting the stepwise adjustment of gray scale obtained from the calibration process during the test embodiment of the invented�I;

Fig.13-15 are graphs of the regression fit of the power function for X, Y and Z, based on the power functions shown in Fig.12;

Fig.16 is a graph of the fit gray scale using a second order polynomial;

Fig.17 is a graph of the fit gray scale using the fourth order polynomial, restricted to the intersection at zero; and

Fig.18-20 are graphs of test results obtained from the embodiment of running the second pass of processing.

Description of specific embodiments of

Various examples of the invention are described below in relation to the accompanying drawings.

1. The first variant of implementation - correction spatial brightness

Fig.1 is a block diagram of a system according to the first embodiment of the present invention. The system has elements on the user side and the elements on the side of the internal server. Elements on the user side are used to capture images of the color sample to be detected, together with the image calibration color samples, colorimetry data for which is known. Elements on the server side or inner elements refer to the elements of treatment that take the image data, processed image data�of Azania, determine the color of the unknown sample colors, pick the color to the palette of colors and then return the selected color from the palette to the user.

In this respect, the aim of the first embodiment of the present invention is the provision of a system which enables the home to the customer, or other user, to accurately identify the color of the unknown sample color. To accomplish this, the user receives the card samples color calibration, for example, by mail, or by visiting a retail store selling paints, where they are available. Map samples color calibration has a cut section, which may be placed an object whose color is to be determined. The user then performs image capture card samples color calibration with an object whose color needs to be defined in a cutout area that is easily accessible capture device image, such as a digital camera or a camera-equipped mobile phone. Then the image is transferred by the user, for example, by electronic mail, service of the multimedia messaging service (MMS), or using a web interface to an internal server, where it is processed, determines the color of the unknown sample color, and information pertaining to a selected color of paint is passed obra�but the user. In addition, information regarding additional colors of paint to make a color scheme of colors, can also be transmitted back to the user.

Fig.1 illustrates in more detail the elements of such a system. Since the terminal devices of the user, the user receives the card 24 samples color calibration, for example, from a local retailer of paint, or sends a letter to her in the mail. Map 24 samples color calibration has on it a number of samples 242 individual colors, the spatial distribution map, and color samples 242 color also distributed in color space. Map 24 samples color calibration has the cut section 244, shown in Fig.1, located in the middle, but which in other embodiments may be located anywhere on the map, which when used is placed the object to be measured, or the card is placed on the object to be measured, so that a portion of the object subject to measurement, is visible through the cut out section 244. For more details, maps 24 samples color calibration are described below in relation to Fig.2.

When using, as noted, the user places the card 24 samples color calibration on the object whose color you want to define. Then, the user uses the device�during image capture, such as a digital camera or a mobile phone provided with a camera to photograph the image of the map 24 samples color calibration with an unknown color sample to be determined, also located in the image. As shown in Fig.1, can be used the device 12 of the image capturing of the user, such as a digital camera, or mobile device 14 of the user device equipped with image capture, such as built-in camera.

If the user performed image capture, the user must then transmit the image to an internal server 10 for image processing. Can be used many different transmission technology to transmit the image data to the internal server 10, and embodiments of the invention are not limited to those described. For example, the user may upload the captured image from the digital camera 12 to the computer 16, the computer 16 is connected to the Internet 22 via the local area network, such as router 18 Wi-Fi (wireless accuracy). The user can then use the computer 16 to send an e-mail message with the image as an attachment by e-mail, which refers to the internal server 10.

Alternatively, the internal server 10 via the network int�of RFAS can provide a dedicated web page, which can boot the computer 16 and displayed by the browser program, which can be placed the image data to be sent back to the internal server 10.

Provides an alternative route to an internal server, when the user uses the mobile phone to capture images. Some mobile devices, often known as smartphones have WiFi functionality and can be used for sending e-mail messages or web access pages similar to a desktop or laptop computer. In this case, the user's mobile device used as a portable computer, and an image captured by them, can be sent by e-mail, or in the quality of the data entered in the web page back to an internal server. Alternatively, the mobile device user can use his cell phone to send the image data to the internal server 10. In this case, the image data can be sent, for example, as a message service, multimedia messaging service (MMS) over the cellular network 26 to the mobile gateway 20, which then transmits the image data to the internal server 10. In this respect, can be provided with specific contact number, and it can be made known to Paul�ovately (for example, printed on the card 24 samples color calibration), which can be sent as MMS.

Internal server 10 includes a network interface 102 coupled to the network 22, for receiving image data from a user and transferring data color selection, as described below. Internal server 10 further comprises a processor 104, which are executed programs that perform color detection and which mainly control the internal server 10. Working memory 106 is provided for use by the processor, which can temporarily store the data.

Also in the internal server 10 is provided computer readable media 108, which forms long-term storage, which can store data and programs. For example, computer readable media 108 may be a hard disk drive, or, for example, may be a solid state storage device. On machine-readable media 108 are a number of management programmes. In this first variant implementation is provided by the module 104 control the selection color, which controls the entire operation of the system and calls other modules to perform operations as and when required. In the first variant of the implementation of an additional module 118 calibration, which when�Imam, accordingly, the control commands from the module 114 controls color coordinated and executed by the processor 104 to perform the calibration function and, in particular, to perform the necessary regression analyses to describe the characteristics of the capture color capture device image used by the user. Additional details of the module 118 calibration is shown below.

In other embodiments, can be provided with additional modules, such as module 116 orientation of the image or the clustering module 120. The work of these additional modules is described below in respect of the relevant variant of implementation.

In the internal server 10 is provided additionally other machine-readable media data 110, which may also take the form of a hard disk, solid state storage, etc. In this regard, the second machine-readable media 110 data may actually represent the same media as the media 108, and may be, for example, a partition of the same hard disk, which comprises a first machine-readable media data 108. The second computer storage media 110 of the data, however, stores a database of colors that contains colorimetric data relating to the sample� colors on the map 24 samples color calibration. Can store multiple sets of such data relating to 24 different cards samples color calibration, which may be available. For each card 24 color samples the calibration is stored the ID card, and then for each known sample colors on the map are saved known XYZ color coordinates with x, y coordinates of the location of the sample color with these color coordinates on a map. Therefore, there are as many sets of coordinate values and the associated color coordinates as there are elements of the sample colors on the map 24 samples color calibration.

Fig.2 illustrates in more detail a sample card 24 color calibration. In particular, the map of 24 samples color calibration has framing 248 at its outer edge and then has printed on it the elements of the known color sample color. The elements of the color sample are arranged so that elements 250 on the outer edge of the Toolbox sample colors represent the elements of the gray scale, i.e., they have the limit changes from black through different colors of gray to white. They should be captured by the capture device image, such as a digital camera, with essentially the same RGB values. They are useful when performing a spatial correction of brightness, as described in a subsequent version of the implementation.

Elements 242 sample C�ETA, located further inward from the edges of the card 24 calibration color samples represent the color elements, each of which has a known color coordinates. In this respect, the color elements should be printed as accurately as possible to the desired values of the trichromatic coordinates. Alternatively, the color map of the calibration can be printed, and then each element is measured to determine its XYZ values using, for example, a spectrophotometer. Color elements 242 of the color sample is preferably distributed throughout the color space. However, in other embodiments, described below, color can be concentrated in a specific area of color space.

Map 24 is also provided with some form of identification marks 246, which may be a bar code, or some other trust mark, such as the printed name, symbol, or etc. It is used to identify which card is used by the user, so that the correct data of the color map can be selected for use.

Finally, the card 24 calibration has the cut section 244, shown here in the middle. However, the position of the cut footage is not important, and it can be located anywhere on the map and even on the edges. In addition, it is not essential that the included neck�nny section; in this respect, map 24 samples color calibration can simply be placed next to the object or the sample, the color of which must be defined, although this is less desirable.

When using, as noted, the user receives the card 24 samples color calibration, for example, from the retailer of paint, and then puts the card 24 calibration so that the cut section is located over the color to be measuring, for example the color of the sofa cushions, curtains, piece of furniture etc. In this regard, the card 24 is placed on top of or against the object whose color needs to be measured so that its color is visible through the cut out section 244. Using a mobile phone, digital camera or etc., the user then takes a still image of the object subject to measurement, with the capture card colors in the image and sends it to the internal server 10, using various routes of communication, as described previously, such as MMS, email or using web access.

Fig.3 depicts in more detail the process performed on the internal server 10.

First data 32 image sent by the user are received at the network interface 102 of the internal server 10. Internal server 10 is controlled by the module 114 controls the selection of the colour, executed on the processor 104. When receiving data, the image�tion, the module 114 controls the selection of the colour at first, optionally, performs image processing for the location and orientation of the card 24 samples color calibration in the image 32. This is done in block 3.2 and is optional because, depending on the instructions given to the user, this step may not be necessary. For example, the card 24 samples color calibration can proceed with the instructions to the user to capture the image so that the position of the card in the image is not tilted. In addition, the user can be instructed to crop the image so that the image is solely map 24 calibration in a known angular orientation before it will be sent to the internal server 10. If the user is given such instructions, and he performs them, there is no need to perform any procedures on the location or orientation of the map. Therefore, in this case, the taken image 32 will be a picture card exclusively calibration with an unknown sample to a known orientation, i.e. it is the image 34 card for map and sample.

If the image 34 card is received, the module 114 controls the selection of the colour controls the processor 104, and runs the module 118 calibration to perform regression analysis for characterization characteristics� capture color capture device image of the user. The regression analysis used in this version of the implementation is essentially the same as described previously in US 5150199 and WO 01/25737 and shown in more detail in Fig.5. In relation to figs.3 regression analysis for characterization of the device is performed in block 3.4 with reference to scheme 35 the placement of the map calibration, the known data from 112 color map stored in the database 110 data colors.

Iterative regression algorithm includes two individual stage of processing as follows:

Step 1: Determine the 3 dependencies between each of the measured components of R, G and B and the known components X, Y and Z, using the color swatches on the grey scale on the map 24 samples color calibration, i.e.

• X as a function of R (called the function R1).

• Y as the function G (called the function G1).

• Z as a function of B (called the function B1).

Fitting exponential curve can be used on gray scale data to obtain the dependencies of R1, G1B1in step 1 above. You can also use the polynomial fit curve or 2nd, 4th or higher order.

Step 2: Determine the multidimensional linear relationship between each of the known constituents X, Y and Z and three functions, defined in step 1 above, i.e.

• X as a function of R1, G1B1(named function X1).

1, G1B1(named function Y1).

• Z as a function of R1, G1B1(named function Z1).

Step 2 in the algorithm performs a multivariate regression of X, Y and Z relative to the data to fit the exponential curve R1, G1and B1obtained in step 1, i.e.

X=f(R1, G1B1)

Y=f(R1, G1B1)

Z=f(R1, G1B1)

or

X=a+b.R1+c.G1+d.B1

Y=a+b.R1+c.G1+d.B1

Z=a+b.R1+c.G1+d.B1

where a, b, c and d are constant coefficients. Three fit X, Y and Z multivariate regression denoted by X1, Y1and Z1respectively.

Fig.5 depicts the above in more detail. In particular, the process of Fig.5 is executed as a block 3.4 Fig.3.

First, in block 5.2, as described, accepts the image data from the color map of known orientation. Then you must identify the color map used in the block 5.4, which can be done using identification marks 246, located on the map 24 calibration. Ie is the recognition identifying marks 246, which is then used as an index to select the appropriate set of colors from the map data base map data 110 colors.

Then begins the first stage of the aforementioned Sal�rhythm, running in blocks 5.6 to 5.14. I.e. in the block 5.6 begins the cycle of reading out the data from the image in known positions in the image. I.e. in the block 5.6 each sample gray scale in a known position (x, y) on the map 24 calibration has RGB values measured from the image in block 5.8, and then searched the XYZ color coordinates for this sample in the same position (x, y) in the database in step 5.10. This process is repeated for all samples of the gray scale in the image that the card 24 calibration are located on the outer edge of the color samples as samples 250. In alternative embodiments do not limit the phase samples of the gray scale, and other color samples can also be used in addition to or as an alternative.

At the end of processing, component units 5.6-5.12, for each known sample color or gray scale in the image XYZ color coordinates are obtained from relevant data of the color map in the database 10 data colors, and will be measured RGB values of the color sample in the image. The corresponding RGB and XYZ values are stored associated with each other in the memory 106. For example, it is possible to plot the measured RGB values for each known sample relative to the known XYZ values of the sample on the chart, as shown in Fig.12, 16 and 17.

<> If the RGB values were measured, and the corresponding XYZ values obtained from the database of flowers on stage, 5.14, is the above stage 1 of the algorithm for determining X-values as functions measured R-values, Y-values as a function of the measured G-values and Z-values as a function of the measured B-values. This step is performed using a power-law fit or a polynomial fit to obtain the function linking X with R, Y, G and Z with B. Usually sedate the fit gives an equation in the form:

where the coefficients αx, y, zand βx, y, zcharacterize these dependencies.

Fig.12, 16 and 17 illustrate an exemplary curve fitting obtained experimental data for tests performed on captured images of the test array 1102 calibration samples, shown in Fig.11. Fig.11 depicts an array of elements 1102 color Swatch together with elements 1104 gray scale located at the bottom of the array. Elements 1102 color sample contain 256 randomly placed reference colors, including six standards grayscale. Elements 1104 gray scale contains 16 colors gray scale from black to white.

To test the experimental test array of Fig.11 was covered with the use of D65 light and the image captured by � high quality digital camera (Cannon PowerShot Pro 90IS). Data of the XYZ color coordinates of the color elements in the test array were known in advance, indexed by the position of the element in the array. With these data it was possible to plot the measured R-, G - and B-values for each element relative to the known XYZ values for each test item, as shown in Fig.12, 16 and 17. It should be noted that the graphs of the data in each of figs.12, 16 and 17 are identical. Different is to fit a curve that was applied. In particular, Fig.12 was used for the power adjustment, in accordance with the above-described dependence. However, as noted, it is also possible to use a polynomial fit, in addition to a power-law fit; Fig.16 depicts a polynomial fit of second order, and Fig.17 depicts a polynomial fit of fourth order, where the function is limited to the intersection at zero. As described below, using a power fit or a polynomial fit, the results are essentially identical, and, apparently, there is very little advantage, if any, when using a polynomial fit over a power-law fit.

If the curve fit was performed for the above functions, block 5.16 performed a multivariate regression of X, Y and Z relative to this function to get the coefficients linking with R, G, and B, Y, R, G, and B and Z with R, G and B, as marked in step 2 above. Fig.13 illustrates a graph of X with respect to known regression adjusting R1 and X1, whereas Fig.17 depicts a known Y from the regression fits G1 and Y1, and Fig.15 depicts a graph of known Z relative to the regression fits B1 and Z1. It finds constant coefficients (a, b, c and d in step 2 above) that help to characterize each of X, Y and Z as a function of R, G and B as described above. If these factors are found, i.e. the coefficients from stage 1 and stage 2 of the above algorithm, they are saved and then characterize the function of the capture color capture device image used by the user. Using these coefficients, you can then find the color of the unknown sample in the image from its RGB values.

According to Fig.3, in block 3.4 is above the calibration process, which returns a set of coefficients 36 calibration, which can then be used for subsequent determination of color.

First, however, you need to determine whether there is any predominant color in the known color sample that is performed in block 3.6. For example, RGB values of a pixel representing the unknown sample may be examined to determine whether there is a predominant RGB value. Alternatively, if there is no predominant RGB values, if ve�-interface is used in the block 3.10, the user may be prompted to select a color to be calibrated. In box 3.12 the selected color is then calibrated. In another embodiment, the implementation describes the process of clustering, which is able to identify numerous color in the unknown sample and calibrated to return a color for each of them.

In box 3.12 selected color is calibrated using coefficients 36 calibration. I.e. RGB values used in the equations found in the unit 3.4 using the 36 coefficients of the calibration. This process gives the XYZ color coordinates of the selected color.

After finding the XYZ values of the unknown sample colors (or selected dominant values in the color Swatch, if you have more than one color) module 114 control colour selection then acts to find the closest color available in the color palette in box 3.14. In this respect, the data 45 of the color palette available for module 114 controls the selection of the colour and stored in the database 110 of these colors. Finding the closest color is done through the use of color difference metrics and compare XYZ color that was defined, with each color in the palette, using the metric differences, this selects the color with the smallest difference. Can be used several different metrics differences, but in variants of the implementation and�gaining preferable to use metrics CIE Delta E. In particular, can be used the original metric of the color difference CIE Delta E (1976), or in another embodiment, the implementation of the metric CIE Delta E (2000). In another embodiment, the implementation can be used Delta E (2000), but with different weighting factors.

The process of selecting colors in the block returns 3.14 paint color selection, which is the paint color in the palette that is closest to a specific XYZ color of the test sample. This information 42 about the color of paint is then given back to the user via the network interface 102 over the network 22. For example, if the user has transferred the image to an internal server 10 via MMS using a mobile device, the network interface 102 may be a message of short message service (SMS) or multimedia message to send information about the color of paint back to the user's mobile device. Alternatively, if the user sent an email message to the internal server 10, the network interface 102 may be an email in response with information about the color of the paint. If you use the web interface, the web page can be sent to the user for display by a web browser of the user, providing information about the color paint selection.

Finally, in some embodiments, in addition to the return information�AI 42 about the color of the paint, in block 3.16 internal server 10 also serves for finding a color scheme that complements a specific color 42 paint. For example, there are several methodologies for determining color schemes that complement each other. For example, a color that is 120° from the first color on the color wheel CIELAB, is often considered an additional color. In addition, a color that is 180° from the first color on the color wheel CIELAB, is also considered to be additional. Therefore, in block 3.16 uses such methods to define additional colors to determine the information about 44 color scheme, which is also returned to the user.

Therefore, in the first variant of implementation, the user may take a digital photo using your mobile phone or your digital camera, of an object whose color you want to define. The photo is taken by placing a card of 24 samples color calibration on top of, next to or near the object, so as the map of 24 samples color calibration and the object captured in the image. The user then sends the image over the network connection from your house to an internal server. In this regard, contact details such as email address, room MMS or the web address may be provided on the back side of the card 24 samples color calibration. EXT�nny server 10 receives the image processes the image, as described, to determine the actual color of the object to be detected, and then selects the color to the palette of colors to determine consistent paint colors to the object. Information relating to the selected color of paint, then returns the reply to the user through the communication network. The answer, for example, can be done by email, SMS, MMS, or by passing the corresponding web page to display in the browser on your computer or mobile phone user. With such a device significantly improved the lung capacity of selection by the user of the paint color. In particular, no longer required to obtain a user with multiple sets of cards color swatches from your local paint store and try to pick colors based on your own perception. Instead, can be obtained much more mathematically precise and reliable colour matching. Furthermore, no specialized equipment required to capture the image, and the user can use the equipment to capture images, which he usually owns.

To evaluate the results of the above process, the measured RGB data for two formulaic standards (second standard template shown in Fig.11 described earlier; the first template benchmark is�I'm the same but without the elements 1104 gray scale below) was also used as the input data sample. These RGB input is used to calculate the calibrated XYZ-values using methods described above. Certain calibrated XYZ color then compared the numerical method with known XYZ values to provide measures of the effectiveness of regression adjustments in the algorithm. To this end, we used two benchmark metrics differences in perceptions CIE dE and CIE DE2000.

The table below displays average values of dE and also DE2000 obtained for each of the methods described above.

the fitting type gray scaleXYZ-componentsthe average dE (reference)the average DE2000
test 1 template 1; camera PowerShot S30 with natural light
the power functionX1, Y1, Z16,04A 3.78
test 2, template 2 (additional gray tones NCS (natural color system); DigiEye camera when the light D65/10°
the power functionX1, Y1, Z1A 4.642,83
the polynomial of the 2nd orderX1, Y1, Z14,312,68
a polynomial of 4-th orderX1, Y1, Z14,802,83

Data in the table above indicate that the replacement fitting exponential curve to data gray scale using polynomial fits have small effect on the resulting values of X1, Y1, Z1with small or no inuence on the average DE2000. Therefore, replacement fit power-law curve to the data gray scale polynomial adjustments leads to insignificant improvement in calibration. This may be due to the fact that is taken into account any variation in the curve fit gray scale in the process of multivariate regression in step 2.

From the point of view of results, metrics differences dE designed, Thu� minimum noticeable difference for the observer-the person will have the value dE, equal to 1. However, for many people dE, is equal to 1, leads to the absence of noticeable differences in color, especially if the colors are not placed near. In the present case describes the process of determining the color, when used on the template with additional gray scale values used in the iterative regression (test 2, using the template shown in Fig.11), leads to the calculated XYZ values having an average dE2000 less than 3 of the actual XYZ values in each test case.

In addition to the above, embodiments of the invention are also considered compensation different spatial brightness across the image. This is accomplished in the present embodiment, as follows.

More specifically, the present variant embodiment of the invention pays attention to the improvement of the determination of the coefficients of the calibration performed in block 3.4 process of Fig.3, and, in particular, given the differences in brightness and contrast in the image 34 card. I.e. the user can clear the image 32 under non-ideal lighting conditions, so on the map there are 24 differences of illumination, so the brightness and contrast on the map are not uniform. This version of the implementation is therefore additional processing that may be performed in the calibration for EXT�rhenium model calibration given such differences in the spatial coverage. Variant implementation is a method that assumes a linear change of brightness and contrast on the map, although it would be possible to find the coefficients of a higher order that simulate the changes of higher order.

Fig.7 illustrates in more detail the process. The process includes two main phases (B. 7.6 and B. 7.10). First, in block 7.2 measured samples Ri, Giand Biwhen (xi, yiin the image, and the corresponding XYZ values of Xi, Yiand Ziderived from data of the color map in the database of flowers. Then are the appropriate dependencies that reflect the known X measured at R, given the position (x, yeach measured R in the image 34 card. It is also known to display Y on measured G, and the known Z measured at B. T. E. considering X-R in more detail, we formulate the dependency that associates X with R, using the stepwise adjustment, but where the coefficient at R depends on the position in the map image. In addition, the displacement member is also entered into the equation, which is also dependent on the situation. I.e., the desired relation between X and R is a positional dependent, dependent on the position of samples on the map. Such dependent from the position of the ratio t�activate are between Y and G, and Z and B. In the present embodiment, the implementation uses the following equation:

where αX, Y, Z, βX, Y, Z, ζX, Y, Z, ηX, Y, Z, YX, Y, Zthat δX, Y, Zand εX, Y, Zare constant coefficients, and (xi, yiis ai-sample on the map, and Ri, Giand Biare the measured RGB valuesi-sample. However, in other embodiments can use different equations that can be used any dependencies that take into account the position of the samples on the map.

The above equations are solved using the method of curve fitting by method of least squares in B. 7.6 to determine the values for αX, Y, Z, βX, Y, Z, ζX, Y, Z, ηX, Y, Z, YX, Y, Zthat δX, Y, Zand εX, Y, Z. However, it may be that without any prior knowledge of these equations will not be easily solved (can be a local maximum value or minimum value). Therefore, optional (unit 7.4) the coefficients αX, Y, Z, βX, Y, Zcan be found in advance, using the samples of the gray scale in the image without depending on the position and performing the fit (least squares) the power curve forXirelativeRiand then� similarly for YirelativeGiandZirelativeBigetting 6 coefficientsαX, Y, Z,βX, Y, Z:

These equations do not account for any spatial distribution of brightness, but are performed to ensure that the initial values of αX, Y, Z, βX, Y, Zthat can then be used in solving dependent equations provisions.

Then in block 7.8 these 21 coefficient (7 per channel - αX, Y, Z, βX, Y, Z, ζX, Y, Z, ηX, Y, Z, YX, Y, Zthat δX, Y, Zand εX, Y, Z) are used to calculate values (X'i,Y'i,Z'i) for all the known samples in the image (Ri,Gi,Bi- samples of the grey scale. They are then used for multivariate fitting in the block 7.10, - in essence, performing a fit using the least squares method of these samples in relation to the measured values (Xi,Yi,Ziusing the equation

Multivariate adjustment then provides an additional 12 coefficients (aX, Y, ZbX, Y, Z, cX, Y, Z, dX, Y, Z). A set of 21 coefficients αX, Y, Z, βX, Y, Z, ζX, Y, Z, ηX, Y, Z, YX, Y, Zthat δX, Y, Zand εX, Y, Zand 12 coefficients aX, Y, Z, bsub> X, Y, Z, cX, Y, Z, dX, Y, Zthen stored as data 36 calibration. These 21+12 coefficients can then be used subsequently (in B. 3.12 of Fig.3) to calculateXYZ-value (Xsamp,Ysamp,Zsamp, representing the interest ofRGB-color (Rsamp,Gsamp,Bsampusing the above equations.

Thus, in the present embodiment, the implementation adapts the calibration process to account for changes in brightness and contrast on the card 24 in the image. This makes the system even more easy to use and imposes less restrictions on the lighting of the scene image, while still enabling you to get good results.

2. The second variant of implementation of the orientation image

Below is described a second variant implementation of the invention. The second variant implementation of the invention as its base takes the first variant of implementation described above, and therefore, the common features between them are not described again.

The second variant of implementation refers to the orientation of the image is performed in block 3.2 process of Fig.3. More specifically, as described earlier in the first variant of implementation, the orientation of the image may be optional because the user can create every�Express card by manually cropping and image rotation card 24 samples color calibration and unknown sample before sending it to an internal server. In this respect the user when removing the image can ensure that the orientation of the map to the image plane is the correct one, without any expectation or inclination.

However, for non-professional users is more preferable not to require the execution of the preprocessing of the image by the user or was not carried out special conditions in the orientation of image when removing the image. Instead, the system should be as easy for non-expert users, requiring only that they could do the photo cards 24 designs color calibration with an unknown color sample card 24 samples color calibration in any orientation. Making the system easy to understand and use by non-professional users and, therefore, will promote the use of the system.

In the second variant of implementation, in order to allow easy use, image 32, adopted an internal server that can hold an image of the card 24 samples color calibration in any orientation. However, to process the data in the image must be known orientation maps 24 samples color calibration and the position of elements of the sample colors on the map in the image. Therefore, in block 3.2, you define the location of the� and orientation of the map image via module 116 orientation of the image.

Fig.4 depicts in more detail the operation module 116 orientation of the image. First, in block 32 4.2 data images are accepted from the network interface 102 (or from the module 114 controls the selection of the colour). To determine the location of the card 24 samples color calibration in the image, in block 4.4 detection is performed on the edge image to detect high contrast edges. In this respect, map 24 samples color calibration has a thick double framing 248 that can be used to determine the location of the card in the image 32, and the frame allows easy identification algorithms edge detection. If such contours in the image are found, then in block 4.6 searches for sequences of nested 4-sided convex contours that have the correct sequence orientation, and where each child represents a significant share of their ancestor. In this respect, the thick bezels looks after edge detection as two nested quadrilateral shape, and therefore, identification of such a sub-form in the image map identifies 24.

Determining the position of the card 24 in the image on the basis of the above, the image may be segmented, leaving data 46 map image, as shown. You then need to identify the Glo�nye signs on the map to be able to perform a perspective transformation to correct the angle of the picture. Therefore, in block 4.8 identified known features maps, such as the corners of the map. You can use any trusting a label to identify a fixed point on the map calibration, but in this embodiment of the inventors is required to identify the 4 points on the map to perform a perspective transformation.

After identifying known points on the map image in the block 4.10 a known point (for example, the corners of the inner framing) used to perform the perspective transformation to correct the angle of the picture. The image 50 cards with fixed inclination shown by example in Fig.4. However, this image of 50 cards with fixed slope can have any angular orientation, therefore, used a priori knowledge of the expected layout of the cards to correctly Orient the map. In this respect, the data 112 color map stored in the database 110 of these colors, and the data card 112 flowers store information relating to the location of the confidence that can be recognized and used for orienteering maps. For example, bar code or trade mark on one edge of the frame has a white area next to it. Therefore m�encouraged to endorse lightest 2 angle and rotate the image, so they were down. Therefore, in block 4.12 recognized a well-known symptom that is related to the angular orientation of the card and the image is 50 cards with fixed inclination then is rotated in the block 4.14, so that the sign is placed with a known angular orientation, thereby rotationally orienting a map. Essentially, the data obtained 34 map images of known orientation.

In other embodiments, it is possible to use any known sign card to achieve the angular orientation. This can also be achieved by performing one of trust signs. Another possibility is the implementation of the scheme of arrangement of samples on the map axisymmetric, so that the angular orientation of the map would be immaterial.

The General result of the above steps is that the user need not intervene and look for the card in the image, and needs no special requirements imposed on the user in regards to how to remove the image or to perform pre-processing before sending to an internal server. Essentially, is considerably more user friendly system and is suitable for use, to a greater extent by non-professional users.

3. The third variant of implementation of the regression EN�Liz, using a reduced color space

The following describes the third embodiment of the invention. The third variant of implementation based on the previously described first or second embodiment of the implementation, so the features that are common between them, again not described.

In the embodiments described so far, the regression analysis for finding the calibration coefficients used maximum number of samples on the map throughout the color space. However, in the present embodiment, the implementation of the if can be obtained by some a priori information about potential color of the unknown sample to be detected, then a regression analysis to determine the calibration factor can be performed with the use of these known samples of colors that are close to the color of the unknown sample. This is similar to "zoom in" that part of the color space of interest, i.e. the part of the characteristics of the capture color capture device image of the user, which, in fact, is of the greatest interest that this part was used to capture the RGB values for an unknown sample. This lower part of the characteristics of the capture color then can be characterized as tightly in an attempt to improve the accuracy�the awn.

In more detail, the conventional calibration process consists of 2 main stages:

1. Regression analysis of the measured samples and their known colors ("standards") to obtain the calibration coefficients that characterize the device used for image acquisition.

2. The use of calibration coefficients for the capture of known RGB color (and position relative to the frame calibration) and obtaining the XYZ color.

In the present embodiment, the implementation of this process is expanded and included the second pass: if the XYZ color first pass is known, a subset of the known samples ("standards") on the map calibration is then used to repeat a step 1. In the present embodiment, the implementation uses theNthe next standards to calibrate the color (from stage 2) and are separate sets of nearest colors for parts of the gamma correction calibration (e.g. B. 5.14 Fig.5) and part of multivariate analysis (e.g., phase B. 5.16 Fig.5). Additional details are shown in Fig.6.

More specifically, in block 6.2 running the first pass of the process of Fig.3, with blocks B. 3.4 to B. 3.12. I.e. the coefficients of the calibration method described in the previous variant implementation, using all known samples of the colors on the map 24. Then is defined by the XYZ color known color sample box 6.4.

This information is then ispolzuetsya identification Nnearest colors of the specimen identified to the XYZ color of the unknown sample in the block 6.6. In this version of the implementation areNGnext gray scale samples andNCnearest color samples, whereNGtypically less thanNC. Details of the tests performed to determine the values forNGandNCbelow. There are a few samples of gray scale and color using the metric differences delta_E, such as delta_E(2000).

Finding the nearest color (grayscale and color) in the block 6.8, calibration is executed again to re-determine the calibration coefficients, but now using only found closest colors. As noted, this is similar to the zoom or focus on a specific area with color space. Theoretically, then, should take into account any local effects that are present in the calibration process.

Re-defining the calibration coefficients in the block 6.10 then re-calibrated XYZ-values of the unknown sample, using the new calibration coefficients and the measured RGB values from an image.

Several trials were conducted to assess the effects of this re-calibration, and the details of them below with reference to Fig.18-21.

Test 1

As an initial evaluation �in the zonal method of calibrating the measured RGB values for the two previously described test patterns (the second test pattern shown in Fig.11 - the first pattern is identical, but without the range of colors on the grey scale at the bottom) was used as the RGB values of the sample. The range of sizes of the subset (i.e. the values forNGandNC) was tested in the second (zone) the passage as follows. Reported values of dE and DE2000 designed for certain values of X1, Y1, Z1.

test 1: pattern 1 (6 Grays), Canon PowerShot S30; natural daylight
the number of nearest colorsthe number of nearest Graysthe average dE (reference)the average DE2000
1st pass256(all)66,04A 3.78
2nd pass15064,633,01
2nd pass 15034,482,85
2nd pass10064,012,69
2nd pass1003Of 3.872,56
2nd pass5063,302,29
2nd pass5033,302,24
2nd pass2562,801,97
2nd pass2532,851,96

From the table above it is clear that in all cases, the second band pass improves the average values of dE and DE2000 (lower dispersion). Fig.18 summarizes the data by reducing the number of colors (NCtemplate, use�x in the second passage, leading to significant improvement DE2000. Reduce the number of colors (NGtemplate gray scale for use in the 2nd pass also improves DE2000, although the effect is not as significant as that which was obtained by reducing the colors.

Test 2

A similar analysis was performed on the data from the second template (shown in Fig.11). As with the first template, the 2nd passage leads to significant improvement in the average dE and DE2000 (see table below). The results are shown graphically in Fig.19.

td align="center"> 25
test 2: pattern 2 (23 Grays), Canon PowerShot Pro90 IS (DigiEye); light source D65
the number of nearest colorsthe number of nearest Graysthe average dE (reference)the average DE2000
1st pass27223A 4.642,83
2nd pass150233,492,10
2-� pass 150103,422,07
2nd pass15053,261,98
2nd pass100233,001,83
2nd pass100102,921,78
2nd pass1005With 2.811,72
2nd pass50232,321,44
2nd pass5010At 2.261,40
2nd pass5052,241,39
2nd pass231,951,23
2nd pass25101,921,20
2nd pass2551,931,20

Fig.19 demonstrates that reducing the number of colors (NC) used in the subset of the calibration data for the second pass, significantly improves the fidelity of certain XYZ (i.e. lowers DE2000). However, reducing the number of samples (NG) gray scale for use at the stage of fitting the exponential curve in the 2nd pass has a small effect on color fidelity.

Test 3 and test 4

Test 3 and test 4 use standards on the template 2, but additionally have "real" data of the sample in the image, by which to evaluate the zonal calibration method.

Test 3

Test 3 represents the scenario "best case", using high-quality digital camera (DigiEye) under controlled conditions of illumination (D65). The results of 10 test samples are presented in the following table.

3,30
test 3: template 2 (23 Grays), Canon PowerShot Pro90 IS (DigiEye), the light source D65
the number of nearest colorsthe number of nearest Graysthe average dE (reference)the average DE2000
1st pass272237,083,90
2nd pass150236,123,45
2nd pass150105,733,22
2nd pass15055,302,93
2nd pass10023Of 5.363,14
2nd pass100 105.08 mm2,96
2nd pass10054,592,62
2nd pass5023To 4.412,67
2nd pass50104,362,59
2nd pass5053,842,28
2nd pass25233,592,33
2nd pass2510Of 3.642,30
2nd pass2553,312,06
2nd pass10232,07
2nd pass10103,452,09
2nd pass1053,392,03

Again, the 2nd band pass reduces the average values of dE and DE2000, providing an improvement relative to the device with a single pass. Impact on DE2000 shown in Fig.20. In this case, the decrease in bothNGandNCinfluenced the decrease of the average values obtained delta-E.

Test 4

Test 4 is a script "realistic case", using "available" digital camera (Canon PowerShot S30) with good natural daylight. The results of 10 test specimens presented in the following table.

Of 7.03
test 4: pattern 2 (23 Grays), Canon PowerShot S30; natural daylight
the number of nearest colorsthe number of nearest Graysaverage E (reference) the average DE2000
1st pass272237.23 percent3,69
2nd pass15023USD 6.163,32
2nd pass150106,083,25
2nd pass15057,083,43
2nd pass100235,272,95
2nd pass100105,112,85
2nd pass10055,472,84
2nd pass504,97To 2.74
2nd pass50104,802,64
2nd pass5055,312,67
2nd pass2523Of 5.062,80
2nd pass2510Of 4.912,73
2nd pass255Of 5.36To 2.74
2nd pass1023Is 6.513,56
2nd pass10106,383,49
2nd pass1053,55

Impact on DE2000 shown in Fig.21. In this test, however, there are minimum values of the DE2000 values at approximately 50 standards. Reducing the number of standards grayscale for use in the second pass has a small effect on DE2000.

These tests show that reducing the number of flowersNCused in multivariate regression, has a significant impact on color accuracy obtained for the unknown sample. In particular, it is provided that may be obtained by some a priori information about the color of the unknown sample, the restriction to nearestNCcolors, whereNCis in the range of 5-250, or more preferably 10-100, or more preferably 20 to 100, or more preferably 30-70, or most preferably 40-60 for multivariate regression, can improve the accuracy of determination of color. Fig.21 depicts that the most accurate definition of color was obtained when about 50 nearest colors were used for multivariate analysis, although good results with DE2000 less than 3.0 are obtained where the number of colors used within the range of from about 20 about 100 colors to colors. Percentage is equal to from about 8% to about 40% of the number of colors that can be available in 24 colors on the map, assuming n�example, that there are about 250 colors on the map.

From the point of view of how can be obtained a priori information about the color of the sample, as noted above, in the present embodiment, the implementation they are obtained by performing processing of the first pass to determine the color, and then perform a second pass with a reduced number of colors in the calibration. However this is not essential and in other embodiments, a priori knowledge can be obtained in some other manner. For example, in one embodiment, the implementation may be possible to make an assumption about the nature of the characteristics of the device forming the image (for example, assuming that the RGB colors are in the color space sRGB). In another embodiment of the reduced number of colors can be obtained by selecting samples that have measured the RGB values close to RGB color to be measured. In another embodiment of the color in the colormap can be with a reduced range. For example, can be produced different versions of the map colors, and each has a subset of the color space, i.e., the map is in "red color", or other card is "blue". The user then selects a card with the colors that are the closest to the color that she desires to pick up, for example, the user knows that she wants to pick up the red a cushion, and, therefore, uses the card 24 having a predominantly red color on it. In all these cases, a reduced set of sample colors, which, as you know, there are about color to be detected, is used to perform the calibration, and, hence, can be taken into account local changes in the characteristics of the capture device colors in this part of the color space.

5. The fourth variant of implementation - clustering for finding many of the colors in the pattern

The following describes the fourth embodiment of the invention. The fourth variant of implementation takes as its basis any of the already described first, second or third embodiments, and therefore not described again the common elements between them.

The fourth variant embodiment of the invention represents a method that can be used, for example, in block 3.6 process in Fig.3, where there is more than one color in an unknown color sample. For example, a user may place a card 24 is applied to an object, which is patterned and which, although there is a predominant color in the pattern, also has a few secondary colors. In this case, the determination must be made, what color should�should perform the pick. In the first variant of implementation was presented an option identify a single predominant color, or by selecting a user color, or by determining the predominant color, using statistical estimates for the pixels representing the sample. In the fourth embodiment of implementation, however, uses a clustering algorithm to identify each of several colors in an unknown color sample so that individual definition of XYZ and selection can then be performed on each individual color.

In the fourth embodiment of the used clustering algorithm k-means to determine the main colors that are present in the sample picture. The k-means clustering is based on Euclidean distance between pixel values. In RGB space, the differences are not observed because they are equal. This means that two pixels that are very close to each other in the RGB space, can appear very different colors or very similar colors. To overcome this, the pixels are converted to the space L*a*b*, which is more homogeneous with regard to perception, so that the perceived difference between the pixels is relatively uniform throughout color space. This process occurs over and�obrezaniem, if he had eliminated the slope, and, preferably, eliminated if the lighting changes on the map (i.e. it works on a calibrated color from image).

Iterative process to determine how many clusters are present in portions of the image representing the unknown sample, and what is the average color in each cluster. The first iteration is the simplest, since it is assumed that there is only one cluster of pixels in the sample. This means that the k-means algorithm returns a cluster containing all the pixels. Then the average L*a*b* pixels in the image, and then calculates the number of pixels within a certain distance from a given medium. If the number of pixels exceeds the threshold, then it is assumed that there is only one color in the image, however, if the number of pixels below the threshold, then the k-means algorithm is performed on the image, trying to group the pixels into two clusters. Calculates the average value of L*a*b* of each cluster, and calculates the number of pixels present within a certain distance from this pixel value. Performed two calculations to check whether this is a significant first verifies that most of the pixels in this cluster is located within an established distance�States (i.e. that average is a good representation of this cluster, and this cluster is ignored if an insufficient number of pixels is within a prescribed distance. The second calculation is that the number of pixels within a set distance from the mean of all valid clusters must be above the threshold (i.e. to check that there was a sufficient number of pixels to be sure that were identified predominant color). If the number of counted pixels is less than the given threshold, then the k-means algorithm is executed again, but tries to group the pixels into three clusters instead of two, and the analysis is repeated.

The following algorithm is used to find clusters, and it is shown in greater detail in Fig.8. The algorithm has several adjustable parameters:

The maximum radius delta-E (dE_thresh);

The required part of the image (F_img);

The minimum share in the cluster (F_cluster);

The maximum number of clusters to attempt (N_max),

and they are set for a particular implementation in the block 8.2. Experimentation shows appropriate values for the adjustable parameters.

The algorithm is as follows:

1. Start with 1 cluster (i.e., all pixels in the sample) (unit 8.4).

2. If the number of clusters is moreN_maxmove to this� 5 (unit 8.6).

3. Calculate the following statistics for each cluster (block 8.8):

a. The mean pixel value (L*a*b*) (unit 8.10);

b. The number of pixels within dE_thresh average value of pixel (P_thresh) (block 8.12).

4. If Sum(P_thresh)/(Number of pixels in the image) is less F_img (block 8.14), to increase the number of clusters by 1 and go to step 2 (block 8.16).

5. To filter the clusters to include only those that have P_thresh/(the number of pixels in the cluster)>F_cluster (block 8.20).

In the above, the inventors refer to color values in Lab space, the algorithm can also be run using the XYZ values as two sets of color data are mathematically related.

Fig.9 and 10 graphically illustrate the operation of the algorithm. Fig.9(a) is identified 92 cluster, but the cluster does not pass the test at the threshold density, as too high a percentage of pixels is outside of the distance dE_thrash from the mean of the cluster. Fig.9(b) attempts to cluster distribution in two clusters, but the cluster 94 is invalid because an insufficient number of pixels is within the radius of the cluster. In addition, the overall sample does not pass the threshold for the entire image of the specimen, as too many pixels is not a valid clusters. Therefore, the number of clusters increases�I to 3, and clustering is performed again.

Fig.10 illustrates the same distribution as that in Fig.9(b), but with three clusters. In part (a) of Fig.10 the number of pixels within a distance average value is not high enough for passage, using two clusters in k-means algorithm, so the analysis is performed again using three clusters. Then the number of pixels within a fixed distance is high enough that the three colors found in the image, represent the average values of each cluster of pixels. In this case, can be identified clusters 1010, 1020 and 1030, as each satisfies applied the threshold tests.

Various modifications may be made in the above described embodiments to provide further embodiments. For example, each of the second to fourth embodiments each have described as based on the first variant implementation. In the first variant of implementation, the image is transmitted over the network connection to an internal server for processing. In the varieties of the first-fourth embodiments, however, this is not necessarily true. Instead, the program can be made available for download to a computer or mobile phone user who can perform the described� machining operations. Thus, the computer or the phone user can calculate the color of the unknown sample from the captured image and to offer optional rebounds paint colors, without any of the image data required for sending over the network.

In addition, in the above embodiments, it was stated that the image is removed, the map contains both 24 and unknown sample. However this is not essential. In other embodiments, may be provided two separate images, separated in time. The first image may be a map image 24, which is used for finding the coefficients of the calibration device for the formation of the image of the user. A separate image then may contain unknown sample, the calibration coefficients determined from the first image, then applied to the RGB values of the unknown sample in the second image. However, this device is less preferred than the above-described device; as for accuracy, it is necessary to maintain essentially identical to the lighting conditions of the first and second images. However, this obstacle is removed, if you shoot a single image containing both map 24 calibration and sample.

Various other modifications, posledstviya, removal or replacement apparent to those skilled in the art and provide additional examples, any and all of which are assumed to be covered by the scope of the attached claims.

1. Method for determining colour from an image that contains:
receiving first image data relating to an unknown colour samples, colorimetry data for which should be determined;
receiving second image data relating to the set of known calibration color samples, colorimetry data for which are already known;
definition of many of the characteristics of the color calibration linking of measurement known color calibration color samples from the second image data with the corresponding known colorimetry data of the calibration color samples; and
calculating colorimetry data of the unknown sample color depending on the color measurement of the first image data and certain characteristics color calibration;
moreover, the method is characterized in that it further comprises a compensation of differences in brightness on a set of known samples color calibration in determining many characteristics color calibration;
the compensation contains the definition of the first set of one or more functions having a first set of coefficients �of alibabki, moreover, one or more functions associated measured colours of the known calibration color samples from the second image data to the known colorimetry data of the calibration color samples and the known position of each known sample in the image; and an analysis of certain functions to locate the second set of functions having a second set of calibration coefficients and the first and second sets of features, and calibration coefficients are used when calculating colorimetry data of the unknown sample color.

2. A method according to claim 1, wherein the analysis includes a calculation of intermediate color values for each known sample color calibration and then using the calculated intermediate color values to determine a second set of functions having a second set of calibration coefficients.

3. A method according to claim 2, wherein the computed intermediate values of color are subjected to a multivariate fit to determine a second set of functions having a second set of calibration coefficients.

4. A method according to claim 3, wherein the multidimensional adjustment is:

5. A method according to any one of claims. 1-4, in which the compensation further comprises, before determining the first set of functions, the definition of the preceding set of functions having the previous set kOe�of pacientov calibration linking the measured colours of the known calibration color samples from the second image data to the known colorimetry data of the calibration color samples without regard to the provisions of the known samples of color with the previous set of calibration coefficients is used as part of the first set of calibration coefficients when determining the first set of one or more functions.

6. A method according to claim 5, in which the previous set of features is determined using samples of only gray scale.

7. Machine-readable media having a computer program stored on it which is designed so that when executing on the computer system executes the computer system of the method according to any one of claims. 1-6.

8. A device for determining colour from an image that contains:
at least one processor; and
at least one memory including computer program code, and at least one memory and computer program code are performed so that cause the executing device, using at least one processor, at least the following:
receiving first image data relating to an unknown colour samples, colorimetry data for which should be determined;
receiving second image data,�wearing to many known samples color calibration colorimetric data for which are already known;
definition of many of the characteristics of the color calibration linking of measurement known color calibration color samples from the second image data with the corresponding known colorimetry data of the calibration color samples; and
calculating colorimetry data of the unknown sample color depending on the color measurement of the first image data and certain characteristics color calibration;
wherein the device is characterized in that the at least one memory and computer program code are made so that they cause the payment device, using at least one processor, the brightness differences on a set of known samples color calibration in determining many characteristics color calibration;
moreover, the compensation contains the definition of the first set of one or more functions having a first set of calibration coefficients, and one or more functions associated measured colours of the known calibration color samples from the second image data to the known colorimetry data of the calibration color samples and the known position of each known sample in the image; and an analysis of certain functions to locate the second set of functions having a second set of coefficients�of antov calibration moreover, the first and second sets of features, and calibration coefficients are used when calculating colorimetry data of the unknown sample color.

9. The device according to claim 8, in which the analysis includes the calculation of intermediate color values for each known sample color calibration and then using the calculated intermediate color values to determine a second set of functions having a second set of calibration coefficients.

10. The device according to claim 9, in which the calculated intermediate color values are subjected to a multivariate fit to determine a second set of functions having a second set of calibration coefficients.

11. The device according to claim 10, wherein the multivariate fit is:

12. Device according to any one of claims. 8-11, in which the compensation further comprises, before determining the first set of functions, the definition of the preceding set of features with the previous set of calibration coefficients that relate the measured colours of the known calibration color samples from the second image data to the known colorimetry data of the calibration color samples without regard to the provisions of the known samples of color with the previous set of calibration coefficients is used as part of the first set of coefficients Cali�toos when determining the first set of one or more functions.

13. The device according to claim 12, in which the previous set of features is determined using only samples of the grey scale.



 

Same patents:

FIELD: physics.

SUBSTANCE: device comprises RGB→NTSC converter units (1), (20), a label mask selection unit (2), multiplier units (3), (18), (19), a breakdown unit (4), a dispersion calculation unit (6), a weight function generating unit (10), a label coordinate determining unit (5), buffer units (7), (8), (12), (21) delay units (11), (13), (14), (15), (16), a sparse matrix generating unit (17), a control unit (9), a clock-pulse generator (22).

EFFECT: reduced colour distortion when colouring monochromatic images.

1 dwg

FIELD: physics.

SUBSTANCE: illumination system (100) comprises: a set (14) of lamps; a controller (115); a user input device (19); memory (120) which determines discrete colour points containing an ID table (121) of hue, an ID tale (122) of saturation, an ID table (123) of brightness and boundary memory (124) which determines the boundary of the colour space. Based on data (x1, x2, x3) received from the user input device and information in the memory, the controller generates colour control signals (ξ1, ξ2, ξ3) for the set of lamps. The controller compares user input data with information in the boundary memory. If the controller detects that the said point lies beyond the boundaries of the colour space, the controller calculates the replacement point on the boundary of the colour space which was determined in the boundary memory (124), and generates is control signals based on the replacement point.

EFFECT: reduced volume of memory space required.

3 cl, 3 dwg

FIELD: information technologies.

SUBSTANCE: colour components are placed into cache memory, when they are properly determined after conversion of colour space. When after conversion components become indeterminate, values placed into cache memory are used instead of using default arbitrary value. Resulting system of colour editing is "stable" in meeting users expectations, protecting them from sudden failures that occur when using arbitrary values for indeterminate colour components.

EFFECT: invention provides for saving of properly determined colour components in conversion from the first into the second colour space.

16 cl, 6 dwg

FIELD: instrument making.

SUBSTANCE: invention is related to calibration of colours in chamber and/or display design. Result is achieved by the fact that digital image generated with the help of chamber and/or display design, is corrected with the help of data base formed from information of reference colours spectrum.

EFFECT: correction of colour defects in digital images.

9 cl, 2 dwg

FIELD: physics, image processing.

SUBSTANCE: invention is related to digital processing of image in process of scanning and copying, and especially to the field of colour and black-and-white text segmentation, when text is automatically extracted from scanned document. According to the present invention, method of text segmentation by colour criterion consists in performance of the following operations: initial image is broken down into non-overlapping units of pixels; new image Z is generated, in which each pixel represents corresponding unit of initial image; in process of scanning serial units are selected from initial image; classification is carried out for current unit by criterion "monochromatic/colour" in space of opposite colours; initial colour unit RGB is broken down into monochromatic units R, G and B; detector of Laplacian-Gaussian edges is applied with specified threshold T to monochromatic unit; number of edges is calculated for each pixel in Z; classification of "text/non-text unit" is performed by comparison of edge number with set threshold C; classified channels are combined, using logical operator OR.

EFFECT: provision of universal approach to preliminary processing of initial document, providing for faster printing of copy and saving of multifunctional device resources.

4 cl, 3 dwg

FIELD: information technologies.

SUBSTANCE: in color correction scheme per each tint any image which is combination of three different chroma signals is splitted into multiple ranges by tint so that colour correction may be performed by range of each tint. To prevent leaving of uncorrected peripheral range area for colour correction, colour correction procedure is performed so that correction ranges are overlapped with each other.

EFFECT: creation of colour correction procedure performed to provide possibility of true correction of all tints in image without leaving any uncorrected range.

5 cl, 4 dwg

FIELD: office equipment, in particular, system and method for correction of image during output to printing device.

SUBSTANCE: computer, with printing device connected to it, is equipped with information carrier, containing file with image, agent module, data storage module, at least one application, graphic device interface and printing device interface. Agent module is made with possible finding of new file with image on data carrier, aforementioned file containing additional information, calculations for this file of numeric index based on image data and recording of computed index together with additional file information to data storage module. Printing device driver is made with possible calculation of numeric index on basis of additional information data received in it, appropriate to calculated numeric index, and correction of image data received by it and transfer of corrected image data to printing device.

EFFECT: possible correction of image in accordance to additional information during output to printing device from various applications, installed on computer.

2 cl, 1 dwg

FIELD: printing ink production systems.

SUBSTANCE: system has first computer which can exchange data with second computer. Second computer transmits data to first computer. Data contains info on required color of ink but it doesn't obligatory has info on other required properties of ink. First computer has data base for predicting color data of ink receptions when using selected set of basic colors of ink, program for selecting receipt of ink based on data for ink required, and program for transmitting info to second computer for representing color which relates to selected receipt of ink. Second monitor has color display where indo is represented. Set of basic colors of ink can be selected on the base of other required properties of ink, for example, low cost, light resistance or chemical resistance.

EFFECT: close match of colors obtained.

15 cl, 3 dwg

FIELD: physics.

SUBSTANCE: method comprises the following steps: determining the standard colour matching the colour of the object to be repaired and determining the best matching version of the standard colour from a given number of version colours, wherein a sample coated with the paint of the standard colour is visually compared under at least two different angles of illumination and/or observation with the colour to be matched, and the visual deviation from the standard colour and the colour of the object to be matched is evaluated based on predetermined deviations for visual properties, wherein the predetermined visual properties include at least one colour property and at least one texture property. Based on the predetermined deviations for the visual properties of the standard colour and the colour of the object to be matched, the best matching version of the standard colour is determined.

EFFECT: reliable colour matching of repair paints without using a data base of samples of all possible colours, and also without using expensive and sensitive equipment such as spectrophotometers and computers.

13 cl

FIELD: measurement equipment.

SUBSTANCE: method to measure coordinates of colour includes illumination of an investigated sample, correction of relative spectral sensitivity of a multi-element photodetector, conversion of reflected or transmitted radiation by the multi-element photodetector into electric signals and their summation with three artificial neurons with appropriate synaptic links. At the same time a new system of colour coordinate measurements is generated, simultaneously coefficients are determined, which establish links between functions of standard system and new system colours summation, structural parameters of correcting light filters and values of synaptic links, and with which the target functions meet the minimum conditions. Then in front of appropriate photodetectors the correcting light filters are installed, and serially F samples are measured with available spectral coefficients of transmission or reflection and colour coordinates in the old system, receiving signals at the same time from photodetectors for each f sample, for the samples they confirm force factors, values and signs of synaptic links, with which target functions meet the minimum conditions, specified values of synaptic links of coefficients, entered into the permanent memory for further measurement and calculation of colour coordinates.

EFFECT: provision of the possibility to increase accuracy of measurement of colour coordinates.

1 dwg

FIELD: measurement equipment.

SUBSTANCE: device comprises a sensor with attached three couples of optic fibres, arranged at the angle of, for example, 45°, relative to the normal line to the controlled surface in the reflection point, with a jacket made of soft rubber, and an electronic unit comprises a clock oscillator, a switchboard, which is connected to three triggers, connected with three measurement and three compensation light diodes, three receivers of optical radiation, connected via Y-shaped optic fibres with a controlled object, and also three additional triggers, connected with additional three measurement and three compensation light diodes.

EFFECT: invention provides for simultaneous control of several process parameters of metal surfaces, such as roughness, grain, colour and imperfection.

2 dwg

Polymer materials // 2463162

FIELD: process engineering.

SUBSTANCE: method of selecting and producing nonvolatile color composition for making parts in selected color from primary or secondary polymer material to be used in first station (4, 6, 8, 10, 12), transmitting color data into centralised system of forecasting coincidence (16) of desirable color and other data as regards polymer material to be produced in desirable color. Said forecasting system (16) comprises data base and software related therewith to receive color data and forecast making and reproducing colors using color distribution system coupled therewith. Device (30) controls distribution units (32, 34, 36, 38 and 40) to be located at Customers' stations, or the like, to distribute dyes for use.

EFFECT: perfect reproduction of part color, particularly, in remolding.

21 cl, 1 dwg

FIELD: physics.

SUBSTANCE: apparatus for testing a hair sample has a measurement region which can hold a hair sample and an image forming unit for receiving light flux reflected from the measurement region. The image forming unit forms an image of the hair sample lying in the measurement region. The apparatus further contains a spectral analysis unit configured to receive light flux. The spectral analysis unit outputs hair sample spectral analysis results in accordance with the received light flux.

EFFECT: invention enables to check whether the required hair sample is located in the measurement region.

30 cl, 5 dwg

FIELD: medicine.

SUBSTANCE: invention refers to medicine, medical equipment, namely optically aided tests. A method for colorimetry of a patient's body surface area involves measuring three separate primary colours of a reference sample and a tested sample, conversion of the measurement results into codes and digital signals. It involves the initial measurement of colour coordinates of the body surface areas and the post-traumatic dynamic measurement ensured by the simultaneous determination of the colour coordinates of an injured and intact (reference) patient's body surface area. The derived colour coordinates of the injured area are corrected for the colour coordinates of the intact area in accordance with colour composition law that is ensured by comparing the colour coordinates of the intact body surface area to the derived results obtained by the initial measurement with dividing the ones by the others. The colour coordinates of the injured body surface area are multiplied by the derived coefficients to derive the corrected colour coordinates of said area with tracking the dynamics of time colour shift.

EFFECT: method provides the accurate colorimetry and colour quantification of greater coloured areas ensured by directly determining the colour coordinates taking into consideration all the factors having an effect on the colour of an injured area, without the need to place the sample inside the colorimetric devices, and enables establishing a prescription of a patient's body area injury.

2 dwg

FIELD: cell microfluorometry.

SUBSTANCE: microfluorometer used to study the fluorescence of single cells comprises a fluorescent microscope with a camera/a microphotometer, a monitor and a light source for fluorescence incitement. The light source is a set of extra bright light-emitting semiconductor diodes with fixed light waves length. The diodes are interconnected with the diode switch and the signal monitor.

EFFECT: measuring sensitivity increase through unwanted emission reduction; measurement accuracy increase.

3 dwg

FIELD: wood treatment.

SUBSTANCE: method of production of outer coating for paper backing in which the outer coating gives paper backing the color which corresponds to the target object color. According to the method the calculations are made to define quantities of at least one group of coloring agents necessary for production of transparent and semi-transparent wood stain from bounding agent. At this when semi-transparent wood stain is applied to paper backing the later will have color corresponding to the target object. Calculations are made using measurements of target object reflectance received with spectrophotometer and preliminary received spectral date of the coloring agents at their application to paper backing. Coloring agents used to create transparent or semi-transparent wood stain does not include white coloring agent or monochrome tone mixture.

EFFECT: simplification of color choice for semi-transparent stains.

8 cl

FIELD: physics.

SUBSTANCE: analyser consists of transducer and electronic unit. Said transducer comprises three optocouplers and is made up of semi-sphere with annular case made from soft resin whereto said three optocouplers secured thereto. Electronic unit comprises master oscillator, switch, three triggers, photoelectric signal processing unit, memory and measuring system.

EFFECT: higher sensitivity and accuracy, simplified design.

3 dwg

FIELD: physics.

SUBSTANCE: analyser for hair mix specimen comprising mix of two colours and common spectrum made up of two colours comprises first colour selector able to select first spectrum representing first colour in said mix from first group made up of at least one spectrum, and iterative spectra combiner connected with first colour selector to repeatedly combine it with second spectrum representing second colour in hair mix from second group of spectra by first spectrum.

EFFECT: optimum combination of first and second spectra with maximum match with first spectrum.

47 cl, 14 dwg

FIELD: measurement technology.

SUBSTANCE: electronic representation former is subject to calibration by means of measuring color signals at least of two calibrating colors. Colorimetric parameters of any calibrating colors are known before. Chosen color is measured by means of electronic representation forming device. Parameters for conversion of measured color signals of calibrating colors to known colorimetric data are calculated by means of mathematical model. Color signals of measured selected color are subject to conversion to colorimetric data and, by using data bank, formula of color is found which has colorimetric data to correspond closely to calculated colorimetric data of chosen measured color.

EFFECT: improved precision of measurement.

14 cl, 2 dwg, 6 tbl

FIELD: printing ink production systems.

SUBSTANCE: system has first computer which can exchange data with second computer. Second computer transmits data to first computer. Data contains info on required color of ink but it doesn't obligatory has info on other required properties of ink. First computer has data base for predicting color data of ink receptions when using selected set of basic colors of ink, program for selecting receipt of ink based on data for ink required, and program for transmitting info to second computer for representing color which relates to selected receipt of ink. Second monitor has color display where indo is represented. Set of basic colors of ink can be selected on the base of other required properties of ink, for example, low cost, light resistance or chemical resistance.

EFFECT: close match of colors obtained.

15 cl, 3 dwg

FIELD: measuring engineering.

SUBSTANCE: method comprises creating databases containing colorimetric characteristics of prime compositions and the base or top layer of the multilayer coating, entering the colorimetric data of the object to be repaired into computer, determining composition of the prime, and choosing the composition of the prime so that the color of the coating applied corresponds to the color of the object to be repaired providing the minimum amount of the coating material.

EFFECT: reduced material consumption.

6 cl, 1 dwg

Up!