Associative recognition device. RU patent 2504837.
 Method and apparatus for performing user video authentication / 2504004Disclosed is a method of performing video authentication of a user, which comprises steps of: receiving a user-provided authentication photograph; capturing a continuous video image of the user in real time over a certain period of time using a video capturing device at a user client; performing real-time decomposition of the video image and obtaining a series of video data frames; selecting from the series of video data frames a specific number of video data frames and generating at least one contrast image for video authentication of the user based on the specific number of video data frames; comparing the authentication photograph with the contrast image and making a decision on the user video authentication result according to the comparison result. |
 Pattern recognition method, pattern recognition device and computer program / 2487408Method of recognising a pattern in an image comprises the following steps: starting search for a pattern in received images, thereby subjecting the received image to: dividing the image into regions (A1-An), calculating region-specific reference values from region image information, wherein samples are taken only from a part of pixels in a region and reference values are calculated for said region based on the samples taken, comparing the calculated reference values (REF) with stored reference values (REF), and displaying the recognised pattern if a part consisting of adjacent regions (A1-An) is found in the received image, where reference values (REF) of regions correspond to the stored reference values (REF) with sufficient accuracy. |
 Method of identifying signals / 2485586Method of identifying signals is realised by applying essentially different transformations of digital sequences in form of loops to obtain unique shortcut characteristics of said sequences, use as addresses of which allows to cut the number of successive comparisons with references. Reducing operations for comparing the identified signal segment with a reference will drop, in the extreme case, from j operations for comparison on the number of reference signals to two operations: calculating the binary address Bi and determining the identifier Cj of the value of belonging to the reference class. |
 Method of identifying mobile device user from unique signature thereof / 2483355Method of identifying a mobile device user from a signature thereof involves booting the device, launching an application, opening an input window, entering user signature data, digitising said data, determining the beginning and the end of fragments of the entered signature, comparing with a decision threshold, making a decision on personal identification, and unlocking the device in case of a positive decision, wherein the digitised data are broken into an identical number of sections, the signature time is analysed, the obtained time is compared with the time of input of the signature of the true owner stored in the device memory, and the device is unlocked in case of a positive decision on signature identification. |
 Generator of number of financial operation and financial operations carried out with its help / 2479032Generator of a number of a financial operation comprises an electronic processing device, a memory unit, a device for data input and a display. It is arranged as capable of generation of a unique number of a financial operation, which models a number of the main account of a regular credit or a debit card of a financial institution, and which includes an actual number of a subscriber account in a coded form. An issuing bank extracts the inbuilt number of the account and the amount, verifies compliance of the inbuilt and the provided amount (and also performs other standard inspections) and in case of their compliance authorizes the operation. |
 System and method of controlling electronic financial operations / 2479030System includes a means of determining designation of an identifier for determining designation of an identifier provided by a potential subscriber in accordance with circumstances thereof, as well as a response means for issuing a predetermined response in accordance with designation of the identifier. The designation can either be "normal" or "forced". If a "normal" identifier is provided, the subscriber is allowed a normal set of operations. If a "forced" identifier is provided, operations are not allowed, the account of the subscriber is frozen and personnel is alerted on the need to assist the subscriber. As an alternative, a limited number of operations can be allowed and personnel can be alerted. |
 Subscriber id verification / 2479029System for performance of a financial operation between a potential subscriber and a receiving side includes a facility for ID and account verification, including the following: a storage facility, where a biometric identifier of a potential subscriber is stored, and account details, a receiving facility for getting the biometric identifier and account details of the potential subscriber from the receiving side; in which the received biometric identifier is verified by comparison with the appropriate biometric identifier stored against account details, a facility of communication with a financial institution, and a facility for response to the receiving side in response to the request on verification of the biometric identifier and account details of the potential subscriber. |
 Method of classifying fingerprint pictures during personal authentication / 2473125During training, a random number N with length of n bits is obtained; n artificial neurons are created; neuron inputs are randomly connected to memory cells with data on inclination of dermal ridges of fragments of examples of the classified fingerprint picture; further, the neurons are trained to output a random number N when presenting examples of the classified picture; the random number N is the hashed and the reference hash value is stored, as well as the connection of the neural network and its training parameters; and during authentication, there is repeated breaking up of the scanned picture into fragments, calculation of the average inclination on the fragment and the calculated value is transmitted to inputs of the neural network; the hash result is calculated at the output of the neural network and then compared with the reference value; in case of a match, personal authentication is continued, otherwise the authentication procedure is stopped once the time allocated for an authentication session runs out. |
 Method of detecting pornography on digital images (versions) / 2473124Fragments of a digital image similar top human faces are detected in all regions of the digital image; the indicator of the fraction of pixels of skin colours inside the region of a fragment is determined on said fragments; a face is detected if a given threshold value is exceeded; if the threshold value of the fraction of skin colour inside the region of the found face is exceeded, the initial colour palette of human skin is adjusted; outlines of the object of the colour skin are determined; an indicator of geometric similarity of the found outline with typical body outlines is calculated; fragments similar to typical naked body parts are searched; the image is analysed for pornographic visual content. |
 Method of determining value and direction of displacement of monitored object / 2471202According to the method, an image of the object is displayed on the raster grid of a video camera and superposition of video images from the raster structure is determined. Two dimensional distribution of intensity of radiation reflected from the object is measured in each line of the reference and current raster, followed by digital processing thereof. By selecting the central part of the lines of the reference and current raster, a signal with a short length is generated. Digital signal processing is carried out by determining the value of the discriminator curve as a difference between correlation functions of the measured periodically offset short lines of the reference raster and corresponding lines of the current raster. Further, the slope of the discriminator curve is standardised to a value of the discriminator curve at extreme points of the monotonous operating zone depending on the sign of the calculated value of the discriminator curve. The value and direction of displacement of the monitored object is determined by averaging estimates on all raster lines. |
 Automatic recognition of symbols on structural background by combination of models of symbols and background / 2249251Method includes following stages: tuning, forming symbols models, recognition, recording background model together with background of read image, separating model of registered background from elementary image of background, combining for each position of symbol of model of letters and/or digits with elementary displaying of appropriate background, forming of combined models, comparison of unknown symbols to combined models, recognition of each unknown symbol as appropriate symbol, combined model of which is combined with it best in accordance to "template comparison" technology. |
 Method for identification of crossed symbols during recognition of hand-written text / 2251736Method includes dividing image on areas, finding areas with hand-written symbols, using structural and sign classifiers for recognition of symbols, use of structure classifier as main recognition tool, selecting best suiting symbol of several variants.. recognition of symbol includes recognition of symbol by at least one additional sign classifier of crossed symbols, performing concurrent comparison to crossed symbol and at least one common symbol like the latter, and identification of symbol as crossed one in case of better compliance to signs of crossed symbols. |
 Identification device with protected photograph and also means and method for authentication of such an identification device / 2253148Device has photographic image of a person and microprocessor, which has processor, memory, connected to processor and containing authentication data, and interface means, connected to said processor to organize communication with external device. Said photographic image has specially concealed information, contents of which when combined with said authentication data provides for authentication of said photographic image, and said microprocessor is made with possible realization of at least a portion of said authentication. |
 Portable device for scientific identity identification / 2257613Device has in case in form of small suitcase, a computer, which is compatible to operation systems meant for using programs of scientific identification. Computer is connected to display and keyboard, it can be connected to printer external relatively to case, and presumes presence of remote connection to processing center, responsible for identification. Device additionally has fingerprint reader connected to computer and digital camera connected to computer. |
 Method for facsimile recognition and reproduction of printed product text / 2260208Method includes conversion of recognized and standard images to digital form, their digital processing by determining coordinates, comparison and determining of match of recognized and standard contours. Determining of coordinates of line of characteristic contour of recognized image of symbol is performed using appropriate standard graphic image by finding value of coordinates X, Y, angle β of position of optical center of text symbols by superposition along area of printed area of digital images - in straight contrast of standard on appropriate recognized in reversed contrast. |
 Method for recognition of images in optical-digital correlators / 2277257Method for recognition of images in optical-digital correlators includes procedures for input of amplitude distributions of standard and compared objects into correlator, transformation of these distributions to synthesized phase distributions, receiving correlation between them, registration of received recognition signal and estimation of recognition result, distributions of standard and compared objects, related to arbitrary type objects, are unambiguously matched with phase random distributions Ψst(x,y), Ψ(x,y), synthesized from distributions of standard and compared objects and starting phase distribution Ψo(x,y), utilized further during recognition in optical-digital correlator instead of real objects. |
| Method for recognition of papillary pattern / 2279128 Method includes generation of three passports, including an additional statistical one and determinate one. Successive execution of comparison procedures of given papillary patterns received from papillary pattern indicator with passports makes it possible to shorten total duration of recognition procedure due to taking a decision about recognition of papillary patter at early stages while satisfying recognition clarity criterions. |
| Method for recognition of papillary patterns / 2279129 Method includes stage of generation of papillary pattern passport with further placement of the latter into computer memory, and stage of comparison of given papillary pattern to passport of papillary patterns, which utilization procedure for comparing two sets of values of electric parameters with arbitrary number of characteristic points, which is performed by full search of sets of characteristic parameters of all characteristic points. In process of full search of sets of coordinates and characteristic parameters, values of electric parameters are selected, matching in two sets, on basis of numbers of coincidences a signal is generated about match of compared sets of electric parameter values. |
 Method for preventing unsanctioned access to system being protected / 2282240Method includes determining coordinates of certain specifics of papillary pattern of user and on basis of difference of coordinates of produced image of print and one stored in database, positive or negative decision is produced about access of user to system. |
 Monitoring method / 2282895Method includes making a digital image of front side of document by determining content of one-color and multi-color components of image pixels, content of color component of each pixel is compared to range from upper to lower thresholds for appropriate pixel of acceptable unmarked document and appropriate abnormal pixel is generated, if value of pixel is outside interval between upper and lower thresholds, then presence of mark different from dirt is detected, if abnormal pixels produced as a result satisfy previously set conditions. |
FIELD: information technology.
SUBSTANCE: associative recognition device comprises P units for calculating an activation function and P groups of units for generating membership function values, wherein each of the P groups of units for generating membership function values includes K units for generating membership function values, inputs of each of which are connected to inputs of like units of membership function values of each of the other groups from the P groups of units for generating membership function values and are inputs of the associative recognition device, wherein the device also includes P groups of multipliers by weight coefficients, each having K units of multipliers by weight coefficients, inputs of each of which are connected to outputs of corresponding units for generating membership function values from P groups of units for generating membership function values, and the outputs are connected to corresponding inputs of P units for selecting the maximum signal, and outputs of each of the P units for selecting the maximum signal are connected to the input of the corresponding unit for calculating the activation function from the P units for calculating the activation function.
EFFECT: broader functional capabilities of the device, high accuracy of recognition when identifying objects with separate weakly or partially distorted regions.
1 dwg
The invention relates to the field of data processing for special applications, in particular, for the conversion of signals and images, asked unweighted digital codes, to weighted codes, and can be used for processing and recognition of signals and images.
A device containing generators of the sawtooth, analog-digital and digital-to-analog converters, the elements OR blocks of memory functions facilities, blocks of determining the minimum units of comparison, blocks subtraction of units, registers, counter and items delay with relevant links [SU 1791815, G06F 7/58, 1990].
The disadvantage of this device are relatively narrow features.
It is also known device containing n concurrent adders, inputs and outputs which are, respectively, the group of inputs and a group of outputs of the device, as well as the n block the multiplication of the weighting coefficients, the entrance of the i-th block the multiplication of the weighting factors (i=1...N) is connected to the output of the i-th parallel adder, and each of the outputs of the j-th block the multiplication of the weighting factors (j=1...N) is connected to the corresponding input weighted signal of the i-th adder (i=j) [A.V. Nazarov, A.I. Loskutov, "neural Network forecasting algorithms and optimization systems", St. Petersburg, "Science and Technique", 2003, .231].
A disadvantage of this device are also relatively narrow features.
In addition, a device containing a group of multipliers on weights inputs are inputs of the device, parallel adder, the inlets of which are connected to the outputs multipliers on the weights, and the evaluation unit of the activation function, the input of which is connected to the output of a parallel adder, and the output is an output device [red'ko V.G. Evolution, neural networks, intellect: Models and concepts of evolutionary Cybernetics. M: Komkniga, 2006, page .5.1].
The disadvantage of this device are also relatively narrow features, due to the fact that the device can detect signals and images with the use of modeling of neurons in the assumption that the weights for the input signals can be determined exactly, that allows to use them in on the weights. In practice, the weighting coefficients cannot always be determined exactly and in most cases their values are unclear, «blurred», which reduces the accuracy of the device and reduces the functionality of the known device.
The closest to the technical nature of the present device is associative recognition containing the first parallel adder and the first block of calculation of the activation function, the input of which is connected to the output of the first parallel adder, and the output is the first output device associative recognition, P-1 parallel adders from the second to the P-th, P-1 blocks calculation of the activation functions from the second to the P-th, inputs, each of which outputs are connected to the same name of parallel adders, and the outputs are the same outputs of the device associative recognition, and also R groups from the first to the P-th blocks the formation of values of functions toiletries, outputs, each of which are connected with inputs of similar parallel adders each of the R groups of blocks of the formation of the values of membership functions contains K blocks the formation of values of functions toiletries from the first to the K-th inputs each of which are connected with inputs of similar blocks of values of functions belonging to each of the other groups of P groups of blocks of the formation of the values of membership functions and are used as inputs of the device associative recognition [EN 2342702, C2, G06K 9/62, 27/06/2008].
The disadvantage of most similar technical solution is relatively narrow features and relatively low recognition accuracy for the case when the requirements (input, excitatory signals, characterizing the observed image, are weak and/or not fully distorted, i.e. the individual elements and image area practically coincide with the corresponding elements and areas of the image of recognized objects. In this case, the values of membership functions for the elements of the charge against the image may lead to disguise of recognized objects due to the effect relative to other elements and areas of the image, which reduces the reliability of their recognition.
Desired technical result consists in expanding the functional possibilities and increase the accuracy of recognition recognition of objects with separate poorly and/or not fully distorted areas.
Desired technical result is achieved by the fact that, in the device containing P blocks calculation of the activation function and P groups of blocks of the formation of the values of membership functions, each of P groups of blocks of the formation of the values of membership functions contains K blocks the formation of values of functions facilities, inputs, each of which are connected with inputs of similar blocks of values of functions belonging to each of the other groups of P groups of blocks of the formation of the values of membership functions and are used as inputs of the device associative recognition, entered P groups multipliers on the weighting coefficients containing K multipliers on weights inputs each of which outputs are connected to the corresponding block the formation of the membership function and P allocation unit maximum signal inputs each of which outputs are connected to the relevant group of multipliers on the weights, and the outputs of each of P blocks to be allocated the maximum of the signal are connected with the input of the unit concerned, the calculation of the activation function of P blocks calculation of the activation function.
Analysis of scientific-technical and patent literature has shown that prior to the filing date were absent device with the specified set of characteristics. Therefore, it meets the requirement of novelty criterion.
In addition, the required technical result is achieved by the fact that to achieve the desired technical result entered P groups multipliers on the weighting coefficients containing K multipliers on weights inputs each of which outputs are connected to the corresponding block the formation of the membership function and P allocation unit maximum signal inputs each of which are connected with outputs of the respective group of multipliers on the weights, and the outputs of each of P blocks to be allocated the maximum of the signal are connected with the input of the unit concerned, the calculation of the activation function of P blocks calculation of the activation function.
Analysis of scientific-technical and patent literature has shown that prior to the filing date lacked the technical solutions, in which the goal of the project was solved using the above combination of signs. Therefore, meets the criterion of inventive step.
Additionally, we note that, as will be shown below, the proposed technical solution can be implemented with the help of known elements of digital technology. Therefore, it meets the criterion of industrial applicability.
On the drawing presents electrical block diagram of the device associative recognition.
Device associative recognition contains P blocks to be allocated the maximum of the signal from the first 1-1 1-P & P blocks calculation of the activation function of the first 2-1 2-P, inputs, each of which outputs are connected to the same name allocation unit maximum signal 1-1...1-P.
Device associative recognition also contains P groups with the first 3-1 on the P-th to 3-P blocks forming the values of membership functions and P groups with the first 4-1 at the P-th to 4-P multipliers on the weights, each P groups multipliers on weights contains K multipliers on weight the coefficients from the first to the K-th (4-1-1, ...4-1-K, ...4-R-1 ...4-P-K), the inputs of each of which are connected to the outputs of similar blocks the formation of the values of the function toiletries P groups formations values of function facilities, and the outputs are connected to the inputs of the same name allocation unit maximum signal 1-1...1-P, each of P groups of blocks of the formation of the values of membership functions contains K blocks the formation of values of functions toiletries from the first to the K-th (3-1-1...3-1-K, 3 ...-P-1...3-P-K), the inputs of each of which are connected with inputs of similar blocks of values of functions the membership of each of the other groups of P groups of blocks of the formation of the values of membership functions and are used as inputs of the device associative recognition.
The isolation block of the maximum of the signal is a standard element of the computing engineering and blocks the formation of values of functions facilities and units of calculation of the activation functions defned on a functional level and in the proposed technical solution uses the examples of their implementation from the device prototype.
The device operates associative recognition as follows.
Previously analyzed P sample groups of signals and images fuzzy, «fuzzy» information, for example, the count of photos. Therefore, the lack of accurate reference images of samples is replaced by their associative, indistinct, «blurred images. As a result, each element of P images formed the membership function of this element in accordance with the most typical of the level of the signal corresponding to the reference image. On the basis of such functions is programming blocks the formation of values of functions facilities (3-1-1...3-1-K, ...3-R-1...3-P-K).
In blocks 4-1-1, ...4-1-K, ...4-P-1 ...4-P-K multipliers on weights generated function facilities are multiplied by the weights reflecting the sense of the corresponding element of the image in the reference image.
In blocks 1-1...1-P allocation maximum signal strength of the generated values of signals with multipliers on the weights corresponding groups of blocks of the multipliers on the weights are allocated the maximum signals that are fed to the inputs of the relevant blocks 2-1...2-P calculate the activation function. In the simplest case, they may be in the form of threshold blocks.
If exceeded their threshold levels is formed at the outputs of the level of logical units, which indicates a recognition of input signal with a fuzzy view on the meaning of its separate elements.
Thus, the proposed device achieved the desired technical result consists in expanding the functional possibilities, because recognition of the image is made using information on the significance of individual picture elements and rank signal processing by allocating the maximum of the signal.
In addition, increases and the recognition accuracy when submitted (input, excitatory signals, characterizing the observed image, are weak and/or not fully distorted, i.e. the individual elements and image area practically coincide with the corresponding elements and areas of the images of the objects being recognized this takes into account information about the significance of individual picture elements. In the proposed technical solution selection of maximum values from the set of values of functions facilities practically excludes recognition error, when applicable (input, excitatory signals, characterizing the observed image, are weak and/or not fully distorted, i.e. the individual elements and image area practically coincide with the corresponding elements and areas of the image of recognized objects, taking into account the significance of data elements.
Device associative recognition device containing P blocks calculation of the activation functions and P groups of blocks of the formation of the values of the functions of the facilities, with each of the P groups of blocks of the formation of the values of membership functions contains K blocks the formation of values of functions facilities, inputs, each of which are connected with inputs of similar blocks of values of functions belonging to each of the other groups of P groups of blocks of the formation of the values of membership functions and are used as inputs of the device associative recognition, wherein entered P groups of blocks of the multipliers on the weights, each of which contains K blocks multipliers on weight the coefficients of inputs, each of which outputs are connected to the blocks form the values of the membership function of P groups of blocks of the formation of the membership function, and the outputs are connected to the corresponding inputs of P blocks to be allocated the maximum of the signal, and the outputs of each of P blocks to be allocated the maximum of the signal are connected with the input of the unit concerned, the calculation of the activation function of P blocks calculation of the activation function.