System and method for serial conversion and encoding of digital data

FIELD: technology for processing digital images, namely, encoding and decoding of images.

SUBSTANCE: in the system and the method, serial conversion and encoding of digital images are performed by means of application of transformation with superposition (combination) of several resolutions, ensuring serial visualization and reduction of distortions of image block integrity and image contour when compared to many standard data compression systems. The system contains a converter of color space, block for transformation with superposition of several resolutions, quantizer, scanner and statistical encoder. Transformation by scanning with usage of several resolutions outputs transformation coefficients, for example, first transformation coefficients and second transformation coefficients. Representation with usage of several resolutions may be produced using second transformation coefficients with superposition of several resolutions. The transformer of color space transforms the input image to representation of color space of the input image. Then, the representation of color space of input image is used for transformation with superposition of several resolutions. The quantizer receives first transformation coefficients and/or second transformation coefficients and outputs quantized coefficients for use by scanner and/or statistical encoder. The scanner scans quantized coefficients for creating a one-dimensional vector, which is used by statistical encoder. The statistical encoder encodes quantized coefficients received from quantizer and/or scanner, which results in compression of data.

EFFECT: increased traffic capacity and increased precision of image reconstruction.

27 cl, 19 dwg

 

The technical field

The present invention relates generally to digital imaging, and more particularly to a system and method that facilitate the encoding and/or decoding of the image.

Prior art

With the wide spread of computer systems, the Internet and digital storage devices has significantly increased the amount of information that became available through computers. With the increasing amount of information has emerged the need for the rapid transmission of information and its efficient storage. Data compression is a technology that facilitates efficient transmission and storage of information.

Data compression reduces the amount of free space required for the presentation of information, and can be used for many kinds of information. Continuously increases the need for compression of digital information, including images, text, audio and video. Typically, data compression uses a standard computer systems; however, data compression can be used by other technologies, such as, but not limited to, digital and satellite television, cellular/digital telephones, etc.

With the growing need to manage, transfer and processing of large amounts of information also increases the need to compress these Yes the data. Although significantly increased storage capacity, the need for information outweighs the achievements in the development of capacity. For example, uncompressed digital image may require 5 megabytes of free space, while this image can be compressed without loss, and will require only 2.5 megabytes of free space. Therefore, data compression facilitates the transfer of large amounts of information. Even with the increase in data transmission speeds, such as broadband, digital subscriber line (DSL), Internet, and cable modems, when using uncompressed information easily achieved within the bandwidth. For example, the transmission of uncompressed image through a DSL line can take ten minutes. However, the same image compression can be transferred in approximately one minute, therefore, providing a ten-fold gain in throughput data.

Essentially, there are two methods of compression, lossy compression and lossless compression. Lossless compression allows you to restore the exact original data after compression, while when the lossy compression of the data recovered after compression may differ from the original data. There is a compromise solution by using two compression methods, as lossy compression PR is the best compression ratio, than lossless compression, since it assumes some compromise in the integrity of the data. Lossless compression can be used, for example, when the critical compression (quality) of the text, as the impossibility of exact recovery (compressed) data can irreversibly affect the quality and readability of the text. Lossy compression can be used for images or not critical (as) of the text, where some amount of distortion or noise is valid or invisible to the human perception.

Image compression is an important technical problem, because digital images are a significant part of the increasing volume of information, which was mentioned above. A large part of the Web page currently contains a lot of images, and many official documents also contain some images. Rapidly increasing use of digital cameras, many users there are literally thousands of images taken by such cameras.

One of the most well-known and widely used methods of image compression is the compression standard image developed by the joint group of experts in the field of photography AGAF (JPEG). The JPEG standard operates by displaying a square block of 8×8 pixels in the frequency domain using the discrete to inusnace transform (DCT). The coefficients obtained by DCT, divided by the scale factor and rounded to the nearest whole number (a process known as quantization) and then through a fixed zigzag pattern of the scan are displayed in a one-dimensional vector. This one-dimensional vector is encoded using a combination of Huffman coding and coding on the basis of the lengths of the series.

Although JPEG is a well-known and commonly used method of compression, it has some drawbacks. For example, one disadvantage of JPEG is that in low bit rate DCT generates disturbances and discontinuities in the reconstructed image (known as image distortion when combined in the cells or blocks). Image distortion when combined in blocks result in the restored image visible boundaries between groups of blocks 8×8 pixels. These image distortion when combined in blocks lead to undesirable deterioration in the image quality. Another disadvantage of JPEG is that JPEG is unable to perform the recovery image with consistent accuracy. In other words, if the image is encoded with a certain accuracy, and later required a lower accuracy (for example, due to limited bandwidth or availability of the storage device), then the images should be decoded and re-encoded.

Some of the disadvantages of JPEG mitigated in the new JPEG2000, in which DCT is replaced by wavelet transformation. Although wavelet transformation ensure the restoration of the smoothed signal without distortion of the image when combined into blocks, they can lead to increased erosion and to cyclic distortion of the image. Additionally JPEG2000 uses a relatively complex system of coding coefficients, which leads to the compression method, which can be three times 3× (or more) slower than JPEG.

The invention

Below simplistically described the invention to provide understanding of some aspects of the invention. The invention is not an extensive overview of the invention. The invention is not intended to identify key/critical elements of the invention and not outlines features of the invention. Its purpose is to present some concepts of the invention in a simplified form as an introductory Chapter for a more detailed description, below.

The present invention provides a system for compression of digital images and the method of applying the transform with overlap (overlap, combine) multiple permissions (conversion by blending (combining) multiple approximations of the image to each of the categories has their permission), which accepts input values (for example, from the color space Converter), and provides a consistent visualization. Transform with overlapping permissions uses a hierarchical biorthogonol conversion overlay, which reduce "image distortion when combined in blocks (blocking), typical of many standard systems, image compression, such as JPEG uses the discrete cosine transform (DCT). Additionally, the use of biorthogonol transformations overlay reduces noticeable image distortion due to fringing compared to standard systems, image compression based on DCT.

According to one particular aspect of the invention provides a system for compression of an image Converter color space Converter with overlay (combining) multiple permissions, quantizer, razvedyvatel (raster image in a one-dimensional vector) and/or statistical coder. Transform with overlapping permissions gives conversion factors, for example the first conversion and the second conversion factors. Representation using multiple permissions can be obtained using the second transform coefficients by zalozenie multiple permissions. The color space Converter converts the input image into a representation of the color space of the input image (for example, compression schemes, with separate information about brightness and color YUV or YCoCg). Then the representation of the color space of the input image is supplied to the conversion unit with the overlapping permissions. The quantizer receives the first transform coefficients and/or the second transform coefficients, and outputs the quantized coefficients for use by razvedyvatel (raster image in a one-dimensional vector) and/or statistical coder. Razvedyvatel (scanner) scans the quantized coefficients for the formation of one-dimensional vector, which is used by the statistical coder. Razvedyvatel can use the deployment order, such perovskia. Statistical encoder encodes the quantized coefficients received from the quantizer and/or razvedyvatel, resulting in data compression. Statistical encoder can use adaptive encoder for performing encoding on the basis of the lengths of the series. According to another aspect of the present invention are ensured by the compression of an image Converter color space conversion is lossless and statistical coder. The conversion is lossless is animal input values from the color space Converter and uses the conversion is lossless (for example, hierarchical Hadamard transform).

According to another aspect of the present invention is a system restore (compressed data) of the image with the statistical decoder block inverse transform and reverse the color space Converter. Statistical decoder receives a stream of bits (for example, formed in the appropriate statistical coder) and decodes the bit stream. Statistical decoder may use an adaptive decoder performing decoding on the basis of the lengths of the series.

Unit inverse transformation takes input values from the statistical decoder and uses the inverse transform (e.g., translate, reverse hierarchical biorthogonol conversion overlay or reverse hierarchical Hadamard transform). Unit inverse transformation produces output values on the back of the color space Converter. The reverse color space Converter converts the input values (for example, YUV and/or YCoCgin the output image in a palette of red, green, blue" GLC (RGB).

According to another aspect of the present invention is a system of image compression for use in a large set of applications graphical representations of documents, vkluchaetsia images segmented by levels, photocopiers, document scanners, optical character recognition, personal digital assistants, Fax machines, digital cameras, digital video cameras and/or video games, etc.

According to other aspects of the present invention are provided methods of coding/data compression, decoding/data recovery, deployment portion of the coefficients, color display, and reverse display colors. Additionally it provides media designed for reading by the computer on which there are commands that can be used by a computer system for image compression, and media designed for reading by the computer on which there are commands that can be used by the computer for the system recovery image. It also provides the data packet, which can be used for communication between two or more computing processes, which contains relevant information, facilitating data compression, this information contains the first transform coefficients based, at least in part, on biorthogonol converting the input values from the overlay, and the second transform coefficients based, at least in part, on bioregion the flax converting overlay at least one of the first conversion factor. Additionally it provides the data packet, which can be used for communication between two or more computer components that facilitate the compression of data that contains the data field containing the first transform coefficients based, at least in part, on the hierarchical Hadamard transform of the input values, and the second transform coefficients based, at least in part, on the hierarchical Hadamard transform at least one of the first conversion factor.

For the full scope of the described aspects and related aspects in the following description, accompanied by drawings, lists some illustrative aspects of the invention. However, these aspects are briefly indicate a few different ways of using the principles of the invention, this is understood to include all such aspects and their equivalents in the present invention. Other advantages and new features of the invention will become apparent from the subsequent detailed description of the invention, illustrated in the drawings.

Brief description of drawings

Figure 1 shows a block diagram of a system for image compression according to the aspect of the present invention.

Figure 2 shows a chart biorthogonol conversion overlay, according to the aspect of this is about inventions.

Figure 3 shows the block diagram of the conversion overlay using multiple permissions, according to the aspect of the present invention.

Figure 4 shows the graph transformation with overlay using multiple permissions, according to the aspect of the present invention.

Figure 5 shows a block diagram of the transform using multiple permissions, according to the aspect of the present invention.

Figure 6 shows a diagram illustrating a data block size of four by four, according to the aspect of the present invention.

7 depicts a diagram illustrating the deployment template, such panovska, the macroblock data, the size sixteen sixteen, according to the aspect of the present invention.

On Fig depicts a diagram illustrating a deployment template for a block of coefficients of the second level, measuring four by four, according to the aspect of the present invention.

Figure 9 shows a block diagram of a system for image compression according to the aspect of the present invention.

Figure 10 shows a diagram of the Hadamard transform of length 4, according to the aspect of the present invention.

Figure 11 shows a block diagram of a system recovery image, according to an aspect of the present invention.

On Fig image is of Agen block diagram, illustrating a method of coding/data compression, according to the aspect of the present invention.

On Fig depicts a block diagram illustrating a method of decoding/data recovery, according to the aspect of the present invention.

On Fig depicts a block diagram illustrating a method of deploying a portion of the coefficients according to the aspect of the present invention.

On Fig depicts a chart illustrating the direct component of the color space Converter is lossless and reversible component of the color space Converter according to the aspect of the present invention.

On Fig depicts a block diagram illustrating a method of displaying a color space, according to the aspect of the present invention.

On Fig depicts a block diagram illustrating the method of the inverse display color space, according to the aspect of the present invention.

On Fig shows a possible variant of the operating environment in which you can operate the present invention.

On Fig schematically shows a block diagram of a possible communication environment, in accordance with the present invention.

Detailed description of embodiments of the invention

The following describes the present invention, in accordance with the drawings, in which used sequential numbering. In the following description to explain the additional purposes set forth numerous specific details to provide a complete understanding of the present invention. However, it is obvious that the present invention may be practiced without these specific details. In other possible embodiments, to simplify the description of the present invention, well-known structures and devices are shown in the form of a structural schema.

Used in this application, the term "computer component" identifies the object associated with the computer hardware, a combination of hardware and software, software, or software in execution. For example, a computer component can be, but is not limited to, a process running on a processor, a processor, an object, an executable program, the flow of tasks, program, and/or computer. As an illustration, a computer component may be an application running on the server, and the server. One or more computer components can reside within a process and/or thread tasks, when this component can reside on a single computer and/or be distributed between two or more computers.

1 shows a system 100 of image compression according to the aspect of the present invention. As noted above, the system 100 of the present invention by implementing conversion 120 with superimposed (combined) carried the channels at permissions provides consistent visibility and less distortion of the image when combined into blocks and distortions outline image compared to many of the standard systems of compression. The system 100 to compress the image contains a Converter 110 color space, the unit 120 transforms with overlapping permissions, the quantizer 130, razvedyvatel 140 and statistical encoder 150.

The Converter 110 color space displays the input image in the representation of the color space of the input image. Then the representation of the color space of the input image is supplied to the unit 120 transforms with overlapping permissions. In one possible embodiment, the Converter 110 color space converts the input image into a YUV representation of the input image GLC (for example, made of red, green and blue components). The view uses the YUV brightness, denoted by Y, the color red, denoted by U, and the color blue, marked V.

In another possible embodiment, the Converter 110 color space converts the input image in view YCoCg. View YCoCguses the brightness signal, denoted by Y, the color orange, denoted by Coand the color green, marked Cg.

Input components GLC appear in YCoCg(for example, as an alternative to the standard YUV described above) using the transformation:

It should be noted that the advantage of displaying YCOCGcolor space is that the conversion of GLC in YCOCGand the reverse transformation from YC0CGin GLC can be performed using integer arithmetic, hence, reducing the computational overhead. Additionally, the inverse transform can be performed without multiplication. View YCoCgcolor space can lead to significantly better compression efficiency than the commonly used representation YUV, because it better approximates a statistically optimal space obtained by analyzing the main components of the data on the enhanced digital image.

It should be noted that according to the invention, many different representations of the color space can contribute to data compression using transform with overlapping permissions. It is assumed that any representation of color space that allows implementation in accordance with the present invention, falls within the scope of the attached claims. Additionally, according to the present invention the Converter 110 color space may be performed by any appropriate will calculate the local process (computing processes) (e.g., integer calculations and/or computation with floating point).

Unit 120 transforms with overlapping permissions accepts input values, for example, from the Converter 110 color space. Unit 120 transforms with overlapping permissions can provide a coherent imaging system 100 of image compression. Unit 120 transforms with overlapping permissions uses a hierarchical biorthogonol conversion overlay. When using transformations with the overlay can be reduced block image distortion occurring in the standard image compression using the discrete cosine transform (DCT), for example, in JPEG. Additionally, the use biorthogonol conversion overlay reduces visible distortion of the outline of the image in comparison with the standard systems of image compression based on DCT.

2 briefly illustrates the block biorthogonol conversion of 200 overlay (BNP), according to the aspect of the present invention. Unit BNP 200 includes block 210 of the first conversion, similar to the DCT (e.g., similar to the DCT, but not identical DCT), which has four input x(0), x(1), x(2) x(3)corresponding to the first data block. Block 200 BNP also contains the second block 220 conversion, podobn the th DCT, having four inputs x(0), x(1), x(2) x(3)corresponding to the second data block. Block 200 BNP has four outputs 230, X(0), X(1), X(2) X(3). According to figure 2 with direct conversion (e.g., coding/data compression) data is processed from left to right, while the reverse transformation (for example, decoding/data recovery) data is processed from right to left. For direct (p), reverse (On) transformations of the scale factors may be different.

For part of the transformation with the imposition of exit 230 for a block of data, which is input to the second block 220 conversion, like DCT, depends on the sign of the previous block of data, which is input to the first conversion block 210 that is similar to the DCT. In a possible variant, where the input is no previous data block (for example, during initialization and/or at a corner point (corner points) of the image), the input values of the first block 210 conversion, similar to the DCT will not be fully defined. Specifically, if the first conversion block 210 that is similar to the DCT, is the first row or column, then x(0) and x(1) extend beyond the boundaries of the image. In this case, a possible solution consists in using even symmetric extensions, sets x(1)=x(2) and x(0)=x(3). Such symmetrical reflection is used in the last conversion 210, p is such DCT, for a row or column of the image. In both cases it is easy to see that the first and last transformation 210 that is similar to the DCT, for a row or column can be replaced by a simple 2x2 operators (for example, two different entrance, two different output).

In one possible embodiment, essentially all calculations in BNP 200 can be performed using only integer arithmetic, without multiplications. For example, for a given z value the new value of z/2 is implemented by a number obtained when the right-shift: z>>1. Further, the number of 1.25 z can be realized by adding the z number, obtained with a double shift z to the right (for example, z+(z>>2)). Although this implementation may lead to small errors drop generated by shifts (with appropriate scaling of the data), it is noteworthy that this implementation is mostly independent of the processor, as a result, essentially, will be the same, regardless of the processor used to perform the conversion. Accordingly, essentially all implementations of the systems and methods provided by the present invention, can, in essence, similar to compressed files for one source bitmap unlike standard systems, data compression, such as JPEG, a standard for the compression of images and sound systems data compression on what I am recording a moving image, developed by the expert group on the moving image of AGDE (MPEG), and other standards.

3 briefly illustrates the conversion unit 300 with the overlapping permissions, according to the aspect of the present invention. The conversion unit 300 with the overlapping of permissions includes blocks from the first source BNP 3101S-e original BNP 310Swhere S is an integer greater than or equal to one. Blocks from the first source BNP 3101S-e original BNP 310Stogether can be identified as the source of the BNP 310. The unit 300 transforms with overlapping permissions also includes block 320 secondary BNP. The conversion unit 300 with the imposition of multiple permissions can be used, for example, unit 120 transforms with overlapping permissions.

Block 310 source BNP accepts input values (for example, from the Converter 110 color space). Block 310 source BNP processes the input values, and outputs the first transform coefficients based, at least in part, on biorthogonol conversion of the input values with superimposed (combined). Block 310 primary source of the BNP can be used, for example, a possible variant of the BNP 200 described above.

Blocks from the first source BNP 3101S-th block 310Sprimary BNP provide the first conversion (the first conversion factors) as input for the secondary BNP 320. In one possible embodiment, block 310 primary BNP provides low-frequency coefficient (e.g., DC) to block 320 secondary BNP. Block 320 secondary BNP handle the first conversion factor (first conversion), and outputs the second conversion factor (second conversion), based at least in part, on biorthogonol transform with overlapping input of the first conversion factor (input of the first conversion factors). Block 320 secondary BNP can be used, for example, a possible variant of the BNP 200 described above.

Representation using multiple permissions can be obtained using the second transform coefficients from the block secondary biorthogonol conversion 320 overlay. For example, bit map, reconstructed using only the second level of reverse hierarchical BNP must restore the bit map image that represents the version of the original image size is reduced by 4 times in comparison with the image obtained using the standard bicubic filter (filter) sample rate.

4 briefly illustrates the conversion of 400 using multiple permissions, according to the aspect of the present invention. Block 400 transformed the education includes the first block 410 1primary BNP, the second block 4102primary BNP, Treti block 4103primary BNP, the fourth block 4104primary BNP and block 420 secondary BNP. Low-frequency coefficient, all of the first block 4101primary BNP, the second block 4102primary BNP, the third block 4103primary BNP and the fourth block 4104primary BNP, served as input to the secondary BNP 420. Block 400 conversion overlay multiple permissions can be used, for example, unit 120 transforms with overlapping permissions.

Below, according to figure 5, illustrated by block 500 conversion overlay multiple permissions, according to the aspect of the present invention. The conversion unit 500 includes a primary unit of the BNP 510 and the secondary unit of the BNP 520. The output of the low-frequency coefficients of the primary unit BNP 510 consistently serves on the secondary unit of the BNP 520. If the primary unit BNP 510 was received enough low-frequency coefficients, the block secondary BNP 520 provides the derivation of the second level. Block 500 conversion overlay multiple permissions can be used, for example, unit 120 transforms with overlapping permissions.

Image processing is used two-dimensional transformation. To obtain two-dimensional is preobrazovaniya to the rows and columns of the input values (for example, each of the Y, Coand Cgtaken from the Converter 110 color space) can be applied to the above-described conversion of the BNP. In one possible embodiment, to reduce computational overhead does not process the columns as a whole, since each column access covers almost the entire array of the bitmap, which would require access to memory outside the cache. Instead, according to the present invention, can be used approach internal buffer scroll", which is the conversion of the column after processing each set of four rows. Thus, it can be calculated two-dimensional transformation only for one scan of the source bitmap.

According to figure 1, the quantizer 130 receives the first transform coefficients and/or the second transform coefficients, and outputs the quantized coefficients for use by razvedyvatel 140 and/or the statistical coder 150. Mainly the loss of information in the system 100 of the image compression makes the quantizer 130. Losses occur due to the quantization coefficient, as for the transformed Y values its quantized version usually is defined by the formula r=int[(Y+f)/s], where s is the step size of the quantizer 130, module f, |f|, basically equal to s/2, and the sign determined by the formula ign(f)=sign(Y). Therefore, when increasing the step size s corresponding to the dynamic range of r decreases as the probability of r to zero. When restoring (e.g., decoding) the approximation of Y is usually restored according to the formulaAccordingly, the smaller the step size s, the closer the approximationBasically if you increase the size of the step data compression is more efficient, however, made large losses. In one possible embodiment, to reduce computational overhead, the quantizer 130 uses integer arithmetic, for example, scaling the values of the integral coefficient of Z and approximating Z/s integer.

Razvedyvatel 140 scans the quantized coefficients for the formation of one-dimensional vector, which is used by the statistical coder 150. In one possible embodiment, razvedyvatel 140 uses progressive scanning (deployment), while in another possible embodiment, razvedyvatel uses scanning columns. In another possible embodiment, razvedyvatel 140 uses the zigzag pattern, such as in standard systems data compression JPEG.

The fourth possible embodiment, the quantized coefficients are deployed in another template, which remains pixaround the m-dependent data (for example, to prevent accidental access to the data). Figure 6 briefly shows the block coefficients of size four by four, according to the aspect of the present invention. Next, Fig.7 illustrates the deployment template (scan), such panovska, the macroblock data, the size sixteen sixteen (group L blocks, in this case L=4), according to the aspect of the present invention. On Fig depicts a deployment template for a block of coefficients of the second level, the size of four by four (for example, formed by blocks 320, 420 or 520 secondary conversion overlay), according to the aspect of the present invention.

For each macroblock (for example, formed by a hierarchical cascade of 4×4 transformations) the converted value is read into one of six groups of factors. Sequential values of each group are read from the M consecutive macroblocks ("portions"), and six groups together in a single vector of length 256 M, which is passed to the statistical coder. Therefore, each portion may be encoded independently. Independent coding enables you to independently decode each portion, as a consequence, if necessary, providing the ability to decode only part of the image bitmap.

The deployment template, illustrated in Fig.7 and 8, is what ombinarea ordered by frequency and in the crawl space (deployment) DC coefficients (for example, which were two levels of BNP) and panovskogo and ordered by frequency and in the crawl space of the AC coefficients (for example, which passed only the first level of the BNP). Pianowski component (template, shaded arrow, figure 7) is used so that for each group of coefficients are AC coefficients AU, which is adjacent in a particular group, come from neighboring blocks 4×4.

Therefore, Group 0 contains certain DC coefficients of the second level of each macroblock, which have passed through the BNP's second level. Then, for each macroblock can be performed scanning of groups with Group 1 to Group 5, with subsequent scanning of groups with Group 1 to Group 5 for the next macroblock, and so on. Group 1 for the macroblock contains the remaining DC coefficients of the macroblock, passed through the BNP's second level. Group 2 contains illustrated the values of the coefficients for each block of the BNP macroblock. Group 3 contains illustrated the values of the coefficients for each block of the BNP macroblock. Group 4 contains illustrated the values of the coefficients for each block of the BNP macroblock. Group 5 contains illustrated the values of the coefficients for each block of the BNP macroblock.

According to figure 1 statistical encoder 150 encodes the quantized coefficients received from the quantizer 130 and/or the disorder 140. The Converter 110 color space, the unit 120 transforms with overlapping permissions, the quantizer 130 and/or razvedyvatel 140 transformed source data of the picture element vector of integers with reduced dynamic range and long strings of zeros, but not implemented data compression. Statistical encoder 150 encodes these quantized coefficients, resulting in data compression.

In one possible embodiment, the encoder 150 is an adaptive encoder for performing encoding on the basis of the lengths of the series. Each bit-slice input vector is processed properly (in a specific order), starting with the high order bit (PRS) to the low order bit. For each coefficient bits marked as "senior", if there was not coded not one bit different from zero, or as a "clarification", if the most significant bit for this factor has already been encoded. Bits refinement with equal probability can be zero or a unit, therefore, they are without change copied into the bit stream. The high-order bits are more likely to be zero, and therefore, they are encoded by the adaptive encoder implementing efficient coding on the basis of the lengths of the series, in which the symbols in accordance with the rule shown in the Table.

Table: encoding Rule on the basis of the lengths of the series to bits with k parameter.
Code wordThe input sequence of bits
0To perform a series of 2kzeros
1cOIncomplete series with <2kzeros followed by a 1, the sign of the coefficient = ′+′ and (C is a k-bit number)
1clIncomplete series <2kzeros followed by a 1, the sign of the coefficient = ′-′

The parameter k controls the compression efficiency of the encoder performs encoding on the basis of the lengths of the series. The larger the value of k, the longer the string of zero bits that can be represented by a code word consisting of a single bit =0, and, consequently, the greater the compression. The parameter k can be adjusted on the basis of statistical data so that 2kthere were approximately equal to the most probable length of the null string.

In the standard encoding on the basis of the lengths of the series of the parameter k, which is either fixed or is updated regularly and is added to the bit stream (because the decoder needs information about any changes to k). However, both approaches can lead to significant loss in efficiency for two reasons. First, the statistics of the input data, based in the nom, changeable, therefore, to track such changes, k should vary. Secondly, the update value k by copying it into the bit stream significantly increases overhead costs, so as to represent the value of k need a few bits. Consequently, the adaptive encoder performs encoding based on the length of the series used in this possible embodiment, for k rule is the reverse adaptation. Under the reverse is understood that k is corrected based on the coded symbols, and not the input. Therefore, as long as the encoder and decoder use the same rules of adaptation, there is no need to transfer values of k. The basic rule of adaptation is very simple. If the code word is zero, it means that he observed a series of zeros, it is assumed that a series of zeros is more likely and, therefore, k increases. If the code word starts with 1, it means that only what was observed incomplete series, therefore, it is assumed that a series of zeros is less likely and, therefore, k is reduced.

Increase k by an integer may result in too rapid adaptation, leading ultimately to the loss in compression efficiency. Accordingly, k can be adjusted fractional values (for example, by increasing and slimming the scaled version of k).

The character encoding on the basis of the lengths of the series can be terminated at the end of each bit layer, and each layer adds a bit field length encoded data. Accordingly, the bit stream can be syntactically parsed, and if required, the youngest of the bit layer may be removed. This is equivalent to re-encode the data with half the step size. Therefore, repeated compression of these data is done simply by highlighting some bits from the compressed file. Can also be achieved scalability precision.

It should be noted that in connection with the invention covers many other methods of statistical coding (e.g., adaptive arithmetic coding), facilitating data compression using transform with overlapping permissions. Any method of statistical coding, which can be implemented in relation to the present invention falls within the scope of the claims.

Although figure 1 shows a block diagram illustrating components of the system 100 of image compression, it should be noted that the Converter 100 color space, the unit 120 transforms with overlapping permissions, the quantizer 130, razvedyvatel 140 and/or statistical encoder 150 can be implemented as one or more if estvo computer components, as this term is defined here. Therefore, it should be noted that according to the present invention executable by the computer components, functioning to implement the system 100 of image compression, the Converter 110 color space, the unit 120 transforms with overlapping permissions, the quantizer 130, razvedyvatel 140 and/or statistical coder 150 may be stored on media that is designed for reading by the computer, including ISOE (integrated circuit applied orientation ASIC), CD (compact disc (CD), UCD (digital versatile disk-DVD), ROM (read only memory device ROM), a flexible disk, hard disk, EEPROM (electrically erasable programmable a persistent storage device EEPROM) and plug the memory card".

Figure 9 depicts a system 900 image compression without losses, according to the aspect of the present invention. The system 900 compression image contains a Converter 110 color space, block 910 lossless transformations and statistical encoder 150.

Block 910 lossless transformations accepts input values, for example, from the Converter 110 color space. Block 910 lossless transformations uses the conversion is lossless. For lossless encoding does not require the use of conversion by applying (images), as it is Udut be no distortion of the image when combined in blocks (as it does not include quantization). For example, block 910 lossless transformations can use hierarchical Hadamard transform. According to figure 10 may be used with diagram 1010 hierarchical transformation, but with blocks of transform size 4*4 implemented circuit 1020 Hadamard transform is lossless. It should be noted that block 1010 lossless transformations can be implemented as one or more computer components, as that term is defined here.

Figure 11 shows a system 1100 of image reconstruction according to the aspect of the present invention. The system 1100 includes a statistical decoder 1110, vertival 1120 (one-dimensional vector to a bitmap), the component 1130, performing the operation, the inverse quantization component 1140 inverse transformation and inverse Converter 1150 color space.

Statistical decoder 1110 receives a stream of bits (for example, formed in the appropriate statistical coder) and decodes the bit stream. In one possible embodiment, the statistical decoder 1110 uses an adaptive decoder performing decoding on the basis of the lengths of the series, similar in operation to the decoder described above with respect to the encoder 150.

Vertival 1120 (one-dimensional vector to bitmap) performs coagulation (transformation, education is Noah deployment) decoded by the statistical decoder input bit stream, taken from the statistical decoder 1110. Vertival 1120 outputs the first quantized transform coefficients and/or second quantized transform coefficients to the inverse quantizer 1130.

In one possible embodiment, vertival 1120 uses coagulation (transformation) in line, while in another possible embodiment, vertival 1120 uses coagulation in columns. In another possible embodiment, vertival 1120 uses the zigzag pattern, such as is used in standard systems data compression JPEG. The fourth possible embodiment, the coagulation of the input bit stream decoded by the statistical decoder is carried out in accordance with the other, but stays fixed (not data-dependent) template (for example, to avoid accidental access to data), such as reverse deployment template, such perovskia.

The inverse quantizer 1130 performs an operation inverse quantization, the quantized first conversion factors and/or second quantized transform coefficients, taken from the component 1120. The inverse quantizer 1130 displays aquantance factors (for example, the first transform coefficients and/or the second conversion factors).

Component 1140 inverse transform output takes the value of the inverse quantizer 1130. In one possible embodiment, the component 1140 inverse transform uses transform, inverse hierarchical biorthogonol transformations overlay, and produces output values to the inverse Converter 1150 color space. For example, a component 1140 inverse transformation can use a conversion opposite to the conversion of 200 with the overlapping permissions, illustrated in figure 2 (e.g., right to left). In another possible embodiment, the component 1140 inverse transform uses a conversion opposite to the conversion is lossless (e.g., translate, reverse hierarchical Hadamard transform), for example, to decode a bitmap image, which was originally coded system 900 lossless encoding. For example, the inverse transform (e.g., lossless) can essentially pay calculations made in block 1020 lossless (for example, to execute in reverse order).

Inverter 1150 color space converts the input values into output image GLC. In one possible embodiment, inverter 1150 color space displays a view of YUV output data in the form of GLC. In another possible embodiment, inverter 1150 color space displays presents the e YC oCgin the output in the form of GLC. It should be noted that in accordance with the invention covers many different representations of the color space, contributing to the restoration of data, for example, using translate, reverse hierarchical biorthogonol conversion overlay. Any suitable representation of the color space that can be used in connection with the present invention, the scope of the claims. Additionally, according to the present invention, the reverse Converter 1150 color space may be performed by any suitable computing process (computing processes) (e.g., integer calculations and/or computation with floating point).

It should be noted that the statistical decoder 1110, vertival 1120, an inverse quantizer 1130, component 1140 inverse transformation and/or inverter 1150 color space can represent computer components.

With regard to possible systems shown and described above, the methods that can be implemented in accordance with the present invention, will be better understood with reference to the flowchart depicted in Fig, 13, 14, 16, and 17. Although for ease of explanation, the methods depicted and described as a sequence of the locks (stages), it should be noted that the present invention is not limited to the order of the blocks, some blocks (stages), according to the present invention, can be performed in a different order and/or concurrently with other blocks (stages), unlike what is depicted and described herein. Moreover, to perform the methods according to the present invention, may be required, not all of the depicted blocks.

The invention can be described in the General context of commands for execution by the computer, such as program modules, executed by one or more components. Essentially, the software modules contain routines, programs, objects, data structures, etc. that perform particular tasks or implement a separate abstract data types. Typically the functionality of the program modules may be combined or distributed in accordance with the requirements in different variants of implementation.

Fig illustrates a method 1200 of coding/data compression, according to the aspect of the present invention. At step 1210 for each macroblock converts each block. In one possible embodiment, for example, the method is lossy) used biorthogonol conversion overlay. In another possible embodiment (for example, the way lossless) used Hadamard transform be the losses (for example, circuit 1020 Hadamard lossless). At step 1220 converts low-frequency coefficient (low-frequency coefficients) of the block. In one possible embodiment, for example, the method is lossy) used biorthogonol conversion overlay. In the second possible variant (for example, the way lossless) used the Hadamard transform is lossless (for example, circuit 1020 Hadamard lossless). Then, at step 1230, the coefficients quanthouse. At step 1240, the coefficients are deployed (scanned). At step 1250 quantized coefficients are encoded.

Fig illustrates a method 1300 decoding/recovery image, according to an aspect of the present invention. At step 1310, the coefficients are decoded. At step 1320 for each macroblock is the inverse transform of the low-frequency coefficient (low-frequency coefficients of each block. In one possible embodiment, for example, the method is lossy) used translate, reverse biorthogonol conversion overlay. In another possible variant (without losses) used the transformation, the inverse Hadamard transform is lossless. At step 1330 is the inverse transform coefficients of each block. In one possible embodiment, for example, the method is lossy) used translate, reverse biorthogonal the resultant transformation of the overlay. In the second possible variant (for example, the way lossless) used the transformation, the inverse Hadamard transform is lossless.

Next, Fig illustrates a method 1400 scanning portions of factors, according to the aspect of the present invention. At step 1410 for each macroblock in the portion scanned one coefficient of the second level (e.g., DC component). Then, at step 1420, for each macroblock in the portion scanned the rest of the coefficients of the second macroblock level. At step 1430 scans the coefficients of the first level of group 2 (e.g., AC components) of each block in the macroblock. At step 1440 scans the coefficients of the first group 3 of each block in the macroblock. At step 1450 scans the coefficients of the first level of group 4 of each block in the macroblock. At step 1460 scans the coefficients of the first group 5 of each block in the macroblock. If there are other portions of the macroblocks that have not been scanned, the scan continues at step 1420. In the described possible ways of scanning formed six groups of transform coefficients (group 0 to 5). Although it is thought that this deployment model (scanning) and grouping gives good compression results can be used any other suitable deployment template (scan) and grouping, for example, e is for faster processing can be neglected compression efficiency. Any such deployment template/grouping, which can be implemented in connection with the present invention, falls under the scope of the claims.

On Fig depicted component 1510 direct Converter (for example, for use by the Converter 110 color space). Component 1510 direct Converter provides conversion of the input source components in the space YCoCg(for example, through scalable versions of Equation (1)). This scaling requires division by 2 (as indicated by the arrows marked 1/2), which can be implemented by shifts to the right, as described above. At first it may seem that the errors introduced by such changes will be unrecoverable. However, in the component 1520 reverse Converter outputs component 1510 of the direct Converter are used in reverse order, so drop because of shifts (e.g., the same as in the component 1510 direct conversion) happen, but their results are now deducted (as indicated by the arrows marked-1/2), as a consequence allowing you to restore the original data. Thus, the component 1520 reverse Converter can restore component YCoCgthe input source components GLC (e.g., exactly).

Fig illustrates a method 1600 of converting the color ol the student. The method 1600 may, for example, be used by the component 1510 of the direct Converter.

In 1610 accepted input GLC (containing the component, component C and component (C). In 1620 are provided by the output of the channel Y, which contain a representation of the average light intensity (brightness). Y channel can be provided based on the transformation of (1)described above (for example, Y is based, at least partially, TO+2H+). In one possible embodiment, the channel Y can be achieved by using addition and/or changes information corresponding to the input data of the CPS, without multiplications.

At step 1630 is provided the output channel Cothat contain the representation of color information (chrominance) input GLC in the direction close to orange. Channel Coyou can provide based on the transformation of (1)described above (for example, Cobased at least in part, on 2K-2C). In one possible embodiment, the channel Coyou can provide using addition and/or changes information corresponding to the input data of the CPS, without multiplications.

At step 1640 is provided the output channel Cgthat contain the representation of color information (chrominance) input GLC in the direction close to the green. Channel Cgyou can provide based on the transformation of (1)described above (in the example, Cgbased, at least partially, on- +2H-C). In one possible embodiment, the channel Cgyou can provide using addition and/or changes information corresponding to the input data of the CPS, without multiplications.

In another possible embodiment, component, component C and/or component can be restored by inverse transform channels YCoCgsecured according to the method 1600.

Fig illustrates a method 1700 of the inverse conversion of the color space. The method 1700 may, for example, be used by the component 1520 reverse Converter.

In 1710 accepted input YCoCgthat contain the channel Y, representing the average intensity of light channel Corepresenting the color information in the direction close to orange, and channel Cgrepresenting the color information in the direction close to the green. At step 1720 is provided by the component, based at least in part, on input YCoCg. The component may be provided on the basis of conversion (1)described above (For example, based at least in part, on the Y+Co-Cg). In one possible embodiment, the component may be provided using addition and/or shifts, the information corresponding to the input data YCoCgwithout multiplications.

At step 1730 is provided a component C, based at least in part, on input YCoCg. Component C may be provided on the basis of conversion (1)described above (for example, C based, at least in part, on the Y+Cg). In one possible embodiment, component C can be achieved using the addition and/or shifts, the information corresponding to the input data YCoCgwithout multiplications.

At step 1740 is provided a component based, at least in part, on input YCoCg. The component can be provided based on the transformation of (1)described above (for example, based at least in part, on the Y+Co-Cg). In one possible embodiment, can be provided using addition and/or shifts, the information corresponding to the input data YCoCgwithout multiplications.

It should be noted that the system and/or method of the present invention can be used in the overall compression system, contributing to the compression of text, handwriting (handwriting), drawings, images, etc. moreover, for specialists in the field of technology it is obvious that the system and/or method of the present invention can be used in a large set of applications graphical representations of documents, including photocopiers, document scanners, optical races is osnovaniya characters personal digital assistants, Fax machines, digital cameras, digital video cameras and/or video games, etc.

To provide additional context for various aspects of the present invention Fig and its subsequent description are intended to provide a brief General description of the corresponding operating environment 1810, which can be implemented in various aspects of the present invention. On Fig depicts additional and/or alternative operating environment, which can operate the present invention. Although the invention is described in the General context of commands for execution by the computer, such as program modules, executed by one or more computers or other devices, for specialists in the field of technology it is obvious that the invention may also be implemented in combination with other program modules and/or as a combination of hardware and software. However, essentially, the software modules include procedures, programs, objects, components, data structures, etc. that perform particular tasks or implement particular types of data. Operating environment 1810 is only one possible option suitable environment and is not intended to impose any limitation on the sphere of the use or functionality of the invention. Other well-known computer systems, environments and/or configurations that may be relevant for use by the invention include personal computers, handheld or laptop devices, multiprocessor systems, microprocessor-based, programmable consumer electronics, network personal computers, mini-computers, General-purpose computers, distributed computing environments that include systems or devices described above, and so on

According Fig possible environment 1810 to implement various aspects of the invention includes a computer 1812. The 1812 computer contains a processor 1814, system memory 1816 and the system bus 1818. The system bus 1818 connects the system components, including the connection system memory 1816 processor 1814, etc. Processor 1814 may be any of various suitable processors. As processor 1814 can also be used dual microprocessors and other multi-processor structure.

System bus 1818 may be any bus of several types of bus structure (bus structures including a memory bus or memory controller, a peripheral bus or external bus, and/or a local bus using any variety of available bus structures, including an 18-bit bus, a standard architecture for promyshlennogo the SAP application (ISA), Microchannel structure ISS (MSA), extended industry-standard architecture for industrial applications RSAP (EISA), Intelligent Electronics IE (IDE), VESA local bus (VLB), a 32-bit system bus expandable to 64 bits, the interaction with which is made without the participation of the Central processor (PCI), Universal Serial Bus upsh (USB), Advanced Graphics Port PHA (AGP)bus, an International Association of manufacturers of memory cards for personal computers IBM PC MEPPPC (PCMCIA), small computer system interface IMX (SCSI), etc.

System memory 1816 includes a volatile memory 1820 and non-volatile memory 1822. In the non-volatile memory 1822 stores the basic input-output BSW (BIOS), containing basic routines to transfer information between elements within the computer 1812, for example, at startup. For the purposes of illustrating non-volatile memory 1822 may include a permanent storage device, a ROM, a programmable ROM (EPROM), electrically programmable ROM (EPROM), electrically erasable programmable permanent memory (EEPROM) or flash memory, etc. Short term memory 1820 includes random access memory (RAM), which acts as external cache memory. As illustrations of RAM is available in many of the x forms such as synchronous RAM (POPS SRAM), dynamic RAM (DOSE, DRAM), synchronous DOSE (SDSU, SDRAM), SDSU with double data rate (SDSU DDS, DDR SDRAM), enhanced SDSU (USDOS, ESDRAM), the DOSE of synchronous communication (DOSES, SLDRAM), and direct Rambus RAM (POSE Rambus, DRRAM) etc.

Computer 1812 also includes removable/non-removable, volatile/non-volatile storage media. On Fig depicted, for example, the storage device 1824 on the disks. Storage device 1824 disks includes devices like a magnetic disk, a flexible disk drive (magnetic) tape, Jazz drive, Zip drive, tape drive, LS-100, Board flash memory, or removable memory card" etc. Optional storage device 1824 drives can include storage media separately or in combination with other storage media including an optical disk such as a compact disk ROM (CD-ROM, CD-ROM), recordable compact disc (H-KD, CD-R), rewritable compact disc (EP-CD, CD-RW), or a digital versatile disk ROM (UCD-ROM, DVD-ROM), etc. To ensure the connection of storage devices 1824, disk-based system bus 1818 commonly used removable or non-removable interface, such as interface 1826.

It should be noted that Fig illustrates a software that acts as an intermediary between users and re is urami host computer, as illustrated in the appropriate operating environment 1810. Such software includes the operating system, 1828. Operating system 1828, which may be stored in storage device 1824 drives, acts to control and allocate resources of the computer system of 1812. System applications 1830 take advantage of the organization's resources, implemented by the operating system 1828, by means of software modules 1832 1834 and data programs stored in system memory 1816 or storage device 1824 on the disks. It should be noted that the present invention may be implemented with various operating systems or combinations of operating systems.

The user enters commands or information into the computer 1812 through the device (s) 1836 input data. Devices 1836 input data include device management position, such as a mouse (pointing device for cursor control, trackball, pen, touch pad, keyboard, microphone, joystick, game pad, satellite dish, scanner, card receiver of a television signal, a digital camera, digital video camera, web camera, etc., These and other input devices are connected with a block 1816 memory port(s) 1838 interface via a system bus 1818. Port(s) 1838 interface includes n is the sample, serial port, parallel port, game port and universal serial bus (upsh, USB). Device(s) 1840 output uses some of the same types of ports that are used by the device(s) 1836 input data. Therefore, the port upsh, for example, may be used to provide input information to the computer of 1812 and to output information from computer 1812 on the device 1840 output data. Adapter 1842 output data is provided to illustrate that among other devices 1840 output data there are some devices 1840 output data, such as monitors, speakers and printers that require special adapters. Adapters 1842 output data include, in the form of illustrations, video and sound cards, etc. which provide a secure connection between the device 1840 output data and system bus 1818. It should be noted that other devices and/or systems are devices that provide input and output data, such as remote computer (remote computers) 1844.

Computer 1812 can operate in an environment with network structure using logical connections to one or more remote computers, such as remote computer (remote computers) 1844. The remote computer (remote computers) 1844 may be a personal computer, behold the ver, a router, a network personal computer, a workstation, a device with a microprocessor, a peer device or other common network node, etc. and typically includes many or all of the elements described in relation to computer 1812. For brevity in the remote computer (remote computers) 1844 shows only storage device 1846. The remote computer (remote computers) 1844 logically connected to the computer 1812 through the network interface 1848 and at the same time physically connected through the connection 1850 communications. Network interface 1848 covers communication network such as a local area network (LAN) and wide area network (WAN). Technology drugs include distributed data interface fiber optic channels (RED), distributed wired data interface (CID), standard organization of local networks, described in the IEEE specifications for Ethernet/IEEE 1502.3), the specification of the local network ring topology Token Ring /IEEE 1502.5 etc. Technology HS include point-to-point communication lines, networks, circuit-switched, such as a Digital network with integrated services (ISDN) and their variants, network with packet switching and Digital subscriber line (DSL), etc.

The compound(I) 1850 communication refers to the hardware/software used to connect network 1848 interface the bus 1818. Though to illustrate the connection 1850 relationships depicted inside the computer, 1812, it can also be external to the computer 1812. Hardware/software necessary for connection with the network interface 1848, include internal and external technologies such as, modems including regular modems for telephone, cable modems and DSL modems, ISDN adapters, and Ethernet cards etc.

On Fig schematically shows a block diagram of a typical computing environment 1900, which can interact with the present invention. The system 1900 includes one or more clients 1910. The client(s) 1910 may represent a hardware and/or software (e.g., threads, processes, computing devices). The system 1900 also includes one or more servers 1930. The server(s) 1930 may also represent a hardware and/or software (e.g., threads, processes, computing devices). For example, the server(s) 1930 may post threads to perform transformations by using the present invention. One possible link between 1910 client and server 1930 may be made in the form of a data packet intended for transmission between two or more computing processes. The system 1900 includes a structure of communication 1950, the cat heaven may be used for communication between the client(s) 1910 and the server(s) 1930. The client(s) 1910 operatively connected with one or more devices 1960 storing customer data, which can be used to store information local to the client(s) 1910. Similarly, the server(s) 1930 operatively connected with one or more devices 1940 data storage server, which can be used to store information local to the server(s) 1930.

The above description includes the possible variants of the present invention. Of course, it is impossible to describe every possible combination of components or methods for describing the present invention, but for specialists in the art it is obvious that many other combinations are possible and modifications of the present invention. Accordingly, it is understood that the present invention includes all such changes, modifications, and variations that do not depart from the essence and fall within the scope of the claims. Additionally, in order to distribute the term "includes" in the detailed description and in the claims, it is understood that this term is used similarly to the term "containing", since the term "containing" in the application interpreted as an unstable word of the claims.

1. The compression system image containing the first block biorthogonol transformation the overlay, receiving input values, which provides the output data containing the first conversion, the first conversion factor is based at least in part, on biorthogonol transformation with the imposition of the input values and the second block biorthogonol conversion overlay, receiving from the first block biorthogonol conversion overlay at least one of the first transform coefficients, which provides an output that contains a second gain, the second conversion factor is based at least in part, on biorthogonol converting overlay at least one of the first conversion, Converter color space, which converts the input image into a representation of the input image and provides this representation as input values for the first block biorthogonol conversion overlay; a quantizer that performs quantization at least one of the first conversion and the second conversion factor and provides an output of quantized coefficients; razvedyvatel, which deploys quantized coefficients; statistical encoder that encodes kwantowani the coefficients, using a digital statistical encoding.

2. The system of image compression according to claim 1, in which the color space Converter converts the representation YCoCg.

3. The system of image compression according to claim 1, in which the color space Converter converts the YUV representation.

4. The system of image compression according to claim 1, in which razvedyvatel uses, at least partially, the order of deployment, such perovskia.

5. The system of image compression according to claim 1, in which the first block biorthogonol conversion overlay uses integer math when performing biorthogonol conversion of input values to overprint.

6. The system of image compression according to claim 1, in which the first block biorthogonol conversion overlay uses math floating-point while performing biorthogonol conversion of input values to overprint.

7. Photocopier using image compression according to claim 1.

8. Document scanner using image compression according to claim 1.

9. Optical character recognition using image compression according to claim 1.

10. Personal digital assistant using image compression according to claim 1.

11. Facsimile machine, etc which changes the image compression according to claim 1.

12. Digital camera using the image compression according to claim 1.

13. Digital camera using the image compression according to claim 1.

14. The system is segmented layered image using image compression according to claim 1.

15. The video game using the image compression according to claim 1.

16. The compression system images that contain the color space Converter, which displays the input image in the representation of the color space, the conversion unit without loss, receiving input values from the color space Converter that provides an output that contains the converted without loss coefficients, and the coefficients lossless transformations based at least in part, on the hierarchical Hadamard transform of the input values, and the statistical encoder that encodes the converted without loss factors using statistical digital encoding.

17. The compression system image on clause 16, in which the color space Converter converts the input image in view YCoCgthe input image.

18. System recovery image, containing statistical decoder that decodes the input bit stream using a digital statistical decoding, omponent inverse transformation, receiving input values from the statistical decoder component and the inverse transform uses transform, inverse hierarchical biorthogonol transformations overlay, and provides output values, and the inverse color space Converter, which displays the output values of the component reverse transformation into the output image in accordance with color RGB (red, green, blue); zwartewater, which performs the operation opposite to deploy, over a statistically decoded input bit stream, and vertival provides output at least one of the first quantized conversion and second quantized transform coefficients; an inverse quantizer which performs inverse quantization on at least on the quantized first conversion factors and/or second quantized transform coefficients and provides the output aquantance factors.

19. System recovery image on p, in which an inverter that converts the color space displays the output values from the view YCoCg.

20. System recovery image, containing statistical decoder that decodes the input bit stream, ISOE is isua statistical digital decoding, component reverse transformation that takes input values from the statistical decoder component and the inverse transform uses translate, reverse hierarchical Hadamard transform, and provides output values, and the inverse color space Converter, which displays the output values from the component reverse transformation into the output image in accordance with color RGB (red, green, blue); zwartewater, which performs the operation opposite to deploy, over a statistically decoded input bit stream and provides the output at least some of the quantized first transform coefficients and the second quantized transform coefficients; an inverse quantizer which performs the operation inverse quantization, at least one of the first quantized conversion and second quantized transform coefficients, and provides the output aquantance factors.

21. System image recovery in claim 20, in which an inverter that converts the color space displays the output values from the view YCoCg.

22. How to compress/encode the image data, including the provision of coefficients of the first level based, what about at least in part, on biorthogonol conversion of the input values with the overlay, ensuring that the coefficients of the second level based at least in part, on biorthogonol the transform coefficients of the first level overlay, and at least one of the following: the quantization coefficients of the first level, the quantization coefficients of the second level, the deployment of at least one of the coefficients of the first level and/or coefficients of the second level, and encoding at least one of the coefficients of the first level and/or coefficients of the second level.

23. Method of recovering/decoding image data, comprising stages of: decoding coefficients, providing the coefficients of the second level based at least in part, on the conversion, reverse biorthogonol conversion overlay, the decoded coefficients, and providing coefficients of the first level, based at least in part, on the conversion, reverse biorthogonol transform with overlapping coefficients of the second level and the decoded coefficients.

24. How to compress/encode the image data containing the steps: providing coefficients of the first level, based at least in part, on the hierarchical Hadamard transform of the input values, and ensure Uchenie coefficients of the second level, based at least in part, on the hierarchical Hadamard transform coefficients of the first level; and at least one of the following: the quantization coefficients of the first level, the quantization coefficients of the second level, the deployment of at least one of the coefficients of the first level and/or coefficients of the second level, and encoding at least one of the coefficients of the first level and/or coefficients of the second level.

25. Method of recovering/decoding of image data, including decoding coefficients, providing the coefficients of the second level based at least in part, on the conversion, reverse hierarchical Hadamard transform, the decoded coefficients, and providing coefficients of the first level, based at least in part, on the conversion, reverse hierarchical Hadamard transform, the coefficients of the second level and the decoded coefficients.

26. The method of deployment of the quantized transform coefficients with the overlapping permissions of the data portion, comprising: deploying at least one coefficient of the second level for each macroblock in portions, the deployment of the remaining coefficients of the second macroblock level and the deployment of the second group of factors first the level for each block in the macroblock.

27. The method according to p, optionally containing the deployment of the third group of factors of the first level for each block in the macroblock.

28. The method according to item 27, further comprising deploying the fourth group of factors of the first level for each block in the macroblock.

29. The method according to p, optionally including the deployment of the fifth group of the coefficients of the first level for each block in the macroblock.

30. A data packet transmitted between two or more computer components that facilitates data compression, the data packet contains a data field containing the first transform coefficients based, at least in part, on biorthogonol converting the input values from the overlay, and the second transform coefficients based, at least in part, on biorthogonol converting overlay at least one of the first conversion factor.

31. A data packet transmitted between two or more computer components that facilitates data compression, and data packet includes a data field containing the first transform coefficients based, at least in part, on the hierarchical Hadamard transform of the input values, and the second transform coefficients based, at least what actiono, in hierarchical Hadamard transform at least one of the first conversion factor.

32. Media designed for reading by the computer that contains the system components of image compression, designed for execution by a computer, containing the first component biorthogonol conversion overlay, which accepts input and provides output data which contains the first transform coefficients based, at least in part, on biorthogonol converting the input values from the overlay, and the second component biorthogonol conversion overlay, which accepts at least one of the first transform coefficients from the first component biorthogonol conversion overlay and provides output data containing the second conversion factor based, at least partially, biorthogonol converting overlay at least one of the first conversion factor.

33. Media designed for reading by the computer that contains the system components of image compression, designed for execution by a computer, containing the first component of the hierarchical Hadamard transform, which accepts input and provides output the derivative data which contain the first transform coefficients based, at least in part, on the hierarchical Hadamard transform of the input values, and the second component hierarchical Hadamard transform, which takes at least one of the first transform coefficients from the first component of the hierarchical Hadamard transform and provides output data containing the second conversion factor based at least in part, on the hierarchical Hadamard transform at least one of the first conversion factor.

34. Media designed for reading by a computer that stores the components of the system recovery image intended for execution by the computer, containing the component of the statistical decoder that decodes the input bit stream using a digital statistical decoding component reverse transformation, which takes input values from the statistical component decoder, and a component of the inverse transform uses transform, inverse hierarchical biorthogonol transformations overlay, and provides output values, and reverse component of the color space Converter that converts the output values of the component reverse PR is education in the output image in accordance with color RGB (red, green, blue).

35. Media designed for reading by a computer that stores the components of the system recovery image intended for execution by the computer, containing the component of the statistical decoder that decodes the input bit stream using a digital statistical decoding component reverse transformation, which retrieves the values from the statistical component of the decoder, while the component of the inverse transform uses transform, inverse hierarchical Hadamard transformation, and provides output values, and reverse component of the color space Converter that converts the output values of the component reverse transformation into the output image in accordance with color RGB (red, green, blue).

36. The compression system image that contains a tool to convert the color space to convert the input image into a representation of the color space, the tool do the conversion with the overlapping permissions on the representation of the color space and provide the first conversion and the second conversion coefficient, the quantization means the first conversion and the second conversion factors the deployment tool (scan) the first conversion and the second conversion factors and statistical tool for digital encoding deployed the first conversion and the second conversion factors.

37. The method of converting the color space, comprising the steps of: receiving input data, is presented in the color RGB (red, green, blue - GLC), providing the output of the channel Y, containing the representation of the average light intensity of the input data of the CPS, providing output channel Withaboutcontaining the representation of color information (chrominance) input GLC in the direction close to orange, and providing output data channel Cgcontaining the representation of color information (chrominance) input GLC in the direction close to the green.

38. The method according to clause 37, additionally comprising at least one of the following, where the input GLC contain component (red)component, G (green) and the component (blue): Y channel based, at least partially, TO+2H+, provision of channel Cobased at least in part, on 2K-2C, and the provision of channel Cgbased, at least partially, on- +2H-C.

39. The method according to clause 37, additionally comprising at least one of the following: channel Y performed using additions and shifts, ensuring channelaboutmade using additions and shifts, and ensuring channel Cg

40. The method according to clause 37, in which component It can be restored by inverse transform channels YCoCg.

41. The method according to clause 37, in which the component 3 can be restored by inverse transform channels YCoCg.

42. The method according to clause 37, in which a component can be restored by inverse transform channels YCoCg.

43. The method of the inverse conversion of the color space that includes receiving input data YCoCgcontaining the channel Y, representing the average intensity of light channel Withaboutrepresenting the color information in the direction close to orange, and channel Cgrepresenting the color information in the direction close to the green, the software component (red), based at least in part, on input of CAaboutCgthe software component C (green), based at least in part, on input YCoCgand the software component (blue), based at least in part, on input YCoCg.

44. The method according to item 43, additionally comprising at least one of the following: software component, based at least in part, on the Y+Co-Cgthe software component C, cos the bath, at least in part, on the Y+Cgand providing a component based, at least in part, on the Y-Co-Cg.

45. The method according to item 43, additionally comprising at least one of the following: providing a component To made using additions and shifts, ensuring component C, is made using additions and shifts, and software component executed by using additions and shifts.



 

Same patents:

The invention relates to communication technology, in particular to circuits noise and quadrature lowering converters

The invention relates to techniques for transmitting messages using a conversion of analogue signals into digital form based on Delta modulation and can be used in a multi-channel telemetry systems seismic acquisition

Delta codec // 2172554
The invention relates to the field of telecommunication and may find application, for example, in a digital telephone devices for the qualitative transformation of rapidly changing analog signals into digital form

The invention relates to measuring technique and can be used in cascades modulation schemes, signal processing

The invention relates to electronic devices for special purposes, particularly to analog-to-digital converters integrating type with Sigma-Delta architecture, designed to work in real time and with high resolution in the elementary cycle conversion

The invention relates to electronic devices for special purposes, particularly to analog-to-digital converters integrating type with Sigma-Delta architecture, designed to work in real time and with high resolution in the elementary cycle conversion

The invention relates to communication technology and computer technology and can be used in communication systems for any type of Delta modulation (DM)

The invention relates to the field of automation and can be used to convert the analog signal into a digital form with high resolution seismic acquisition or research seismic equipment

The invention relates to the field of encoding and transmitting information and can be used in the information electronic devices, recording, radio broadcasting, television

FIELD: image processing systems, in particular, methods and systems for encoding and decoding images.

SUBSTANCE: in accordance to the invention, input image is divided onto several image blocks (600), containing several image elements (610), further image blocks (600) are encoded to form encoded representations (700) of blocks, which contains color code word (710), intensity code word (720) and intensity representations series (730). Color code word (710) is a representation of colors of elements (610) of image block (600). Intensity code word (720) is a representation of a set of several intensity modifiers for modification of intensity of elements (610) in image block (600), and series (730) of representations includes representation of intensity for each element (610) in image block (600), where the series identifies one of intensity modifiers in a set of intensity modifiers. In process of decoding, code words (710, 720) of colors and intensity and intensity representation (730) are used to generate decoded representation of elements (610) in image block (600).

EFFECT: increased efficiency of processing, encoding/decoding of images for adaptation in mobile devices with low volume and productivity of memory.

9 cl, 21 dwg, 3 tbl

FIELD: method for encoding and decoding digital data transferred by prioritized pixel transmission method or stored in memory.

SUBSTANCE: in accordance to the invention, informational content being encoded and decoded consists of separate pixel groups, where each pixel group contains value of position, at least one pixel value and priority value assigned to it, where at least one key is used, with which value of position and/or pixel value/values of pixels of pixel group are selectively encoded or decoded. Depending on used keys and on parts of informational content which are encoded, for example, value of positions and/or values of pixel groups, many various requirements may be taken into consideration during encoding.

EFFECT: ensured scaling capacity of encoding and decoding of digital data.

8 cl, 5 dwg, 3 tbl

FIELD: systems for encoding and decoding video signals.

SUBSTANCE: method and system for statistical encoding are claimed, where parameters which represent the encoded signal are transformed to indexes of code words, so that decoder may restore the encoded signal from aforementioned indexes of code words. When the parameter space is limited in such a way that encoding becomes inefficient and code words are not positioned in ordered or continuous fashion in accordance with parameters, sorting is used to sort parameters into various groups with the goal of transformation of parameters from various groups into indexes of code words in different manner, so that assignment of code word indexes which correspond to parameters is performed in continuous and ordered fashion. Sorting may be based on absolute values of parameters relatively to selected value. In process of decoding, indexes of code words are also sorted into various groups on basis of code word index values relatively to selected value.

EFFECT: increased efficiency of compression, when encoding parameters are within limited range to ensure ordered transformation of code word indexes.

6 cl, 3 dwg

FIELD: technology for encoding and decoding of given three-dimensional objects, consisting of point texture data, voxel data or octet tree data.

SUBSTANCE: method for encoding data pertaining to three-dimensional objects includes following procedures as follows: forming of three-dimensional objects data, having tree-like structure, with marks assigned to nodes pointing out their types; encoding of data nodes of three-dimensional objects; and forming of three-dimensional objects data for objects, nodes of which are encoded into bit stream.

EFFECT: higher compression level for information about image with depth.

12 cl, 29 dwg

The invention relates to the representation of three-dimensional objects on the basis of images with depth

The invention relates to the representation of three-dimensional objects on the basis of images with depth

The invention relates to the representation of three-dimensional objects obtained using photos of real objects

The invention relates to photo - and video system technology

The invention relates to the encoding of the synthesized images

FIELD: technology for encoding and decoding of given three-dimensional objects, consisting of point texture data, voxel data or octet tree data.

SUBSTANCE: method for encoding data pertaining to three-dimensional objects includes following procedures as follows: forming of three-dimensional objects data, having tree-like structure, with marks assigned to nodes pointing out their types; encoding of data nodes of three-dimensional objects; and forming of three-dimensional objects data for objects, nodes of which are encoded into bit stream.

EFFECT: higher compression level for information about image with depth.

12 cl, 29 dwg

Up!