Abstract: Methods for augmenting a training image base representing a print on a background, for training parameters of a convolutional neural network, CNN, or for classification of an input image The present invention relates to a method for augmenting a training image base representing a print on a background, characterized in that it comprises the implementation, by data processing means (11) of a server (1), of steps of: (b) For at least a first image of said base, and a ridge map of a second print different from the print represented by said first image, generation by means of at least one generator sub-network (GB, GM, GLT) of a generative adversarial network, GAN, of a synthetic image presenting the background of said first image and representing the second print.

Abstract: A parameter training method for a convolutional neural network (CNN) for classifying image type data representative of a biometric trait. The method includes the implementation, by a data processor of a server, the steps of (a) for at least one data item from an already classified training database, generation of several alternate versions of this data each by application of at least one transformation chosen from a set of reference transformations satisfying a statistical distribution of transformations observed in the training database and (b) training the parameters of the CNN, from the already classified training database augmented with said alternate versions of the data.

Abstract: The present invention concerns a method of classification of an input image representative of a biometric trait by means of a first convolutional neural network, CNN, characterized in that it comprises the implementation by data processing means (21) of a client (2) of steps of: (c) Estimation of a transformation parameter vector of said input image, by means of a second CNN, the parameters of the vector being representative of a geometric transformation enabling the biometric trait represented by the input image to be registered in a common reference frame; (d) Application to said input image of a transformation defined by said estimated transformation parameter vector, so as to obtain a registered input image; (e) Classification of the registered input image by means of the first CNN.

Abstract: The present invention relates to a method for extracting a feature vector from an input image representative of an iris by means of a neural network, characterized in that it can be trained end-to-end and comprises the implementation by data-processing means of a client of steps of: (a) Segmentation of the input image representative of the iris by means of a first subnetwork in order to obtain an iris segmentation map, a pupil segmentation map and an attention map; (b) Extraction by a second subnetwork of the neural network of a feature vector from the normalized image representative of the iris segmented by a normalization operation, characterized in that it is derivable.

Abstract: The present invention relates to a method for extracting features of interest from a fingerprint represented by an input image, the method being characterized in that it comprises the implementation, by data processing means (21) of a client (2), of steps of: (a) Binarization of said input image by means of a convolutional neural network, CNN, in such a way as to generate an image called binary; (b) Processing said binary image so as to extract said features of interest from the fingerprint represented by said input image.

Abstract: Methods for augmenting a training image base representing a print on a background, for training parameters of a convolutional neural network, CNN, or for classification of an input image The present invention relates to a method for augmenting a training image base representing a print on a background, characterized in that it comprises the implementation, by data processing means (11) of a server (1), of steps of: (b) For at least a first image of said base, and a ridge map of a second print different from the print represented by said first image, generation by means of at least one generator sub-network (GB, GM, GLT) of a generative adversarial network, GAN, of a synthetic image presenting the background of said first image and representing the second print.

Abstract: The present invention relates to a method for extracting a feature vector from an input image representative of an iris by means of a neural network, characterized in that it can be trained end-to-end and comprises the implementation by data-processing means of a client of steps of: (a) Segmentation of the input image representative of the iris by means of a first subnetwork in order to obtain an iris segmentation map, a pupil segmentation map and an attention map; (b) Extraction by a second subnetwork of the neural network of a feature vector from the normalized image representative of the iris segmented by a normalization operation, characterized in that it is derivable.

Abstract: The present invention concerns a method of classification of an input image representative of a biometric trait by means of a first convolutional neural network, CNN, characterized in that it comprises the implementation by data processing means (21) of a client (2) of steps of: (c) Estimation of a transformation parameter vector of said input image, by means of a second CNN, the parameters of the vector being representative of a geometric transformation enabling the biometric trait represented by the input image to be registered in a common reference frame; (d) Application to said input image of a transformation defined by said estimated transformation parameter vector, so as to obtain a registered input image; (e) Classification of the registered input image by means of the first CNN.

Abstract: Biometric recognition method using a standard electronic processing unit including a first computer and a first memory and a secure electronic processing unit including a second computer and a second memory, the method including the steps of executing a first part of the recognition steps by the standard processing unit and a second part of the recognition steps by the secure electronic processing unit. Device for the execution of that method.

Abstract: The present invention relates to a method for extracting features of interest from a fingerprint represented by an input image, the method being characterized in that it comprises the implementation, by data processing means (21) of a client (2), of steps of: (a) Binarization of said input image by means of a convolutional neural network, CNN, in such a way as to generate an image called binary; (b) Processing said binary image so as to extract said features of interest from the fingerprint represented by said input image.

Abstract: The present invention relates to a method of displaying images implemented by a data processing module able to be connected to a display system comprising at least two display zones (ZA1, ZA2), said method of displaying images comprising the following steps: —displaying of a first image (11) in a first display zone (ZA1) of said display system, —displaying of at least one second image (12) obtained by applying to the first image (11) at least one image processing algorithm, each second image (12) being displayed in a second zone (ZA2) of said display system, —as long as one of said second zones (ZA2) displaying one of said second images (12) is hovered over by a pointer, displaying, in all or part of the first display zone (ZA1), of at least one part of said first image to which is applied the at least one processing algorithm making it possible to obtain said second hovered-over image (12).

Abstract: The present invention relates to a parameter training method for a convolutional neural network, CNN, for classifying image type data representative of a biometric trait, where the method is characterized in that it comprises the implementation, by data processing means (11) of a server (1), of steps of: (a) For at least one data item from an already classified training database, generation of several alternate versions of this data each by application of at least one transformation chosen from a set of reference transformations satisfying a statistical distribution of transformations observed in the training database. (b) Training the parameters of said CNN, from the already classified training database augmented with said alternate versions of the data.

Abstract: A method of detecting zones liable to contain at least one finger print in an image of at least one portion of hand of an individual includes a step of acquiring the image, a step of determining a plurality of zones of the acquired image and a first selection step. The step of determining a plurality of zones determines that each zone is a continuous zone of points of the acquired image which each exhibit a luminous intensity level greater than or equal to a selected luminous intensity threshold value. The first selection step selects, from among the zones determined, the zones which are liable to represent at least one print of a finger.

Abstract: The invention relates to a method for processing an image comprising a set of pixels, each pixel being associated with a grey level, the method comprising a step of segmentation of the image to generate a modified image only containing regions of the image having an alternation of bright zones and dark zones at a frequency greater than a minimum frequency, said step of segmentation comprising: the allocation, to each pixel of the image, of a frequency response level, corresponding to a frequency of alternations of bright zones and dark zones in the vicinity of the pixel, the definition of regions of the image by grouping together neighboring pixels of same frequency response level, the determination of a threshold frequency response level, and the generation of an image only comprising regions of which the pixels have a frequency response level greater than or equal to the threshold frequency response level.

Abstract: Identification method using data in a database that are distributed in at least a first group of data and a second group of data, a comparison being made using first of all the biometric data in the first group and then, in the event of failure of the comparison, using the biometric data in the second group, the distribution of the data between the groups being modified according to a number of validated identifications stored for each user, the first group comprising the data of the user comprising a number of validated identifications greater than a predetermined threshold over a predetermined period and identification system for implementing this method.

Abstract: The invention relates to a method for processing an image comprising a set of pixels, each pixel being associated with a grey level, the method comprising a step of segmentation of the image to generate a modified image only containing regions of the image having an alternation of bright zones and dark zones at a frequency greater than a minimum frequency, said step of segmentation comprising: the allocation, to each pixel of the image, of a frequency response level, corresponding to a frequency of alternations of bright zones and dark zones in the vicinity of the pixel, the definition of regions of the image by grouping together neighbouring pixels of same frequency response level, the determination of a threshold frequency response level, and the generation of an image only comprising regions of which the pixels have a frequency response level greater than or equal to the threshold frequency response level.

Abstract: (EN)The present invention relates to a method of displaying images implemented by a data processing module able to be connected to a display system comprising at least two display zones (ZA1, ZA2), said method of displaying images comprising the following steps: —displaying of a first image (11) in a first display zone (ZA1) of said display system, —displaying of at least one second image (12) obtained by applying to the first image (11) at least one image processing algorithm, each second image (12) being displayed in a second zone (ZA2) of said display system, —as long as one of said second zones (ZA2) displaying one of said second images (12) is hovered over by a pointer, displaying, in all or part of the first display zone (ZA1), of at least one part of said first image to which is applied the at least one processing algorithm making it possible to obtain said second hovered-over image (12).

Abstract: Identification method using data in a database that are distributed in at least a first group of data and a second group of data, a comparison being made using first of all the biometric data in the first group and then, in the event of failure of the comparison, using the biometric data in the second group, the distribution of the data between the groups being modified according to a number of validated identifications stored for each user, the first group comprising the data of the user comprising a number of validated identifications greater than a predetermined threshold over a predetermined period and identification system for implementing this method.

Abstract: A method for signing up a person for biometric verification purposes is provided, the method including: acquiring an image of a biological attribute of the person, the biological attribute including a set of characteristic elements, defining, within the acquired image, at least one area that includes at least part of the biological attribute, and storing, in a biometric database, at least one piece of information indicating that the number of characteristic elements included in said defined area is less than a respective predetermined integer. A related biometric verification method is also provided.

Abstract: A method of detecting zones liable to contain at least one finger print in an image of at least one portion of hand of an individual includes a step of acquiring the image, a step of determining a plurality of zones of the acquired image and a first selection step. The step of determining a plurality of zones determines that each zone is a continuous zone of points of the acquired image which each exhibit a luminous intensity level greater than or equal to a selected luminous intensity threshold value. The first selection step selects, from among the zones determined, the zones which are liable to represent at least one print of a finger.