Abstract: A method for the secure use of a first neural network on an input datum, the method including the implementation, by data processing device of a terminal, of the following steps: (a) constructing a second neural network corresponding to the first neural network, into which is inserted, at the input of a target layer of the first neural network, at least one auto-encoder neural network trained to add a parasitic noise to its input; (b) using the second neural network on the input datum.

Abstract: Method for monitoring individuals in a place equipped with cameras, the method including the steps of: detecting representative characteristics of an individual in a first zone of the place; detecting representative characteristics of an individual in at least a second zone and comparing these representative characteristics with representative characteristics previously stored in order to determine whether the individual detected in the second zone has been detected in other zones; confirming the determination from a topography of the place by checking consistency of a path of the individual from one zone to another and of the time model by comparing temporal detection information on the representative characteristics in the zones. Device for implementing this method.

Abstract: The invention relates to a method for securing a document including a visual element, carried out by a processing unit comprising processing means, the method comprising generation, from the visual element, of a reference security datum, and storage of the reference security datum, wherein the reference security datum is generated by means of an algorithm configured so as to generate: for any image acquired from the visual element, a security datum whereof the differences relative to the reference security datum are less than a determined threshold, and for any image acquired on a different visual element, a security datum whereof the differences relative to the reference security datum are greater than said threshold.

Abstract: The invention concerns a method to compare two data obtained from a sensor or interface, carried out by processing means of a processing unit, the method comprising the computing of a similarity function between two feature vectors of the data to be compared, characterized in that each feature vector of a datum is modelled as the summation of Gaussian variables, said variables comprising: a mean of a class to which the vector belongs, an intrinsic deviation, and an observation noise of the vector, each feature vector being associated with a quality vector comprising information on the observation noise of the feature vector, and in that the similarity function is computed from the feature vectors and associated quality vectors.

Abstract: A method for identifying and/or authenticating an individual by a system including a storage device storing at least one binary code of a reference iris and a binary code of an associated reference mask: acquisition of an image of an eye of the individual, segmentation of the acquired image to isolate a region containing the texture of the iris and determine an associated mask, division of the region and the mask into N2 concentric rings, normalization of the region, determination of a binary iris code IC and a binary code of mask IM matching at least one ring of the iris with a ring of the reference iris to minimize a matching distance between the acquired image and the reference image, computation of the minimized matching distance, identification and/or authentication of the individual if the minimized matching distance is less than a predetermined threshold.

Abstract: System for processing images including a main neural network, preferably convolution-based (CNN), and at least one preprocessing neural network, preferably convolution-based, upstream of the main neural network, for carrying out before processing by the main neural network at least one parametric transformation f, differentiable with respect to its parameters, this transformation being applied to at least part of the pixels of the image and being of the form p?=f(V(p),?) where p is a processed pixel of the original image or of a decomposition of this image, p? the pixel of the transformed image or of its decomposition, V(p) is a neighborhood of the pixel p, and ? a vector of parameters, the preprocessing neural network having at least part of its learning which is performed simultaneously with that of the main neural network.

Abstract: The invention relates to a method for securing a document including a visual element, carried out by a processing unit comprising processing means, the method comprising generation, from the visual element, of a reference security datum, and storage of the reference security datum, wherein the reference security datum is generated by means of an algorithm configured so as to generate: for any image acquired from the visual element, a security datum whereof the differences relative to the reference security datum are less than a determined threshold, and for any image acquired on a different visual element, a security datum whereof the differences relative to the reference security datum are greater than said threshold.

Abstract: The present invention relates to a method for identifying and/or authenticating an individual comprising the following steps performed by a system comprising a storage device storing at least one binary code of a reference iris and a binary code of an associated reference mask: acquisition 101 of an image of an eye of the individual, segmentation 102 of the acquired image of the eye so as to isolate a region containing the texture of the iris and determine an associated mask, division 103 of said region containing the iris texture and of said mask into N2 of concentric rings, normalization 104 of said region containing the texture of the iris and of said mask of Cartesian coordinates in polar coordinates, determination 105 of a binary iris code IC of size equal to N2*P*F and a binary code of mask IM associated by application of F Gabor filters to each position of P positions of each ring, matching 106 at least one ring of the iris of the acquired image with a ring of the reference iris so as to minimize a ma

Abstract: Method for monitoring individuals in a place equipped with cameras, the method including the steps of: detecting representative characteristics of an individual in a first zone of the place; detecting representative characteristics of an individual in at least a second zone and comparing these representative characteristics with representative characteristics previously stored in order to determine whether the individual detected in the second zone has been detected in other zones; confirming the determination from a topography of the place by checking consistency of a path of the individual from one zone to another and of the time model by comparing temporal detection information on the representative characteristics in the zones. Device for implementing this method.

Abstract: The invention concerns a method to compare two data obtained from a sensor or interface, carried out by processing means of a processing unit, the method comprising the computing of a similarity function between two feature vectors of the data to be compared, characterized in that each feature vector of a datum is modelled as the summation of Gaussian variables, said variables comprising: a mean of a class to which the vector belongs, an intrinsic deviation, and an observation noise of the vector, each feature vector being associated with a quality vector comprising information on the observation noise of the feature vector, and in that the similarity function is computed from the feature vectors and associated quality vectors.

Abstract: A procedure for identifying a person with a data acquisition device (202) that is configured for capturing the image of their eyelashes (102, 104). The captured image is compared with eyelash images in a database.