Patents by Inventor Herve Chabanne

Herve Chabanne has filed for patents to protect the following inventions. This listing includes patent applications that are pending as well as patents that have already been granted by the United States Patent and Trademark Office (USPTO).

  • Publication number: 20230401280
    Abstract: A distributed-computing method for computing a metric f(X,Y) representative of a difference between a datum X comprising n bits (x1, . . . , xn) and a datum Y, the metric f(X,Y) taking the form f(X,Y)=fX(X)+?i=1nfi(xi,Y)+fY(Y), the method being implemented by a system comprising m+1 devices and including for any j ranging from 1 to m, computing, by way of the device of index j, at least one intermediate datum depending on the datum X and on the datum Y, and transmitting to the device of index m+1 at least one result comprising or depending on each intermediate datum, and determining, by way of the device of index m+1, the metric f(X,Y), the determining comprising summing each result to obtain a value equal to ?i=1nfi(xi,Y) or a value equal to ?i=1nfi(xi,Y)+fY(Y).
    Type: Application
    Filed: June 14, 2023
    Publication date: December 14, 2023
    Applicant: IDEMIA IDENTITY & SECURITY FRANCE
    Inventors: Hervé CHABANNE, Vincent DESPIEGEL
  • Patent number: 11790222
    Abstract: The present invention concerns a method for learning the parameters of a convolutional neural network, CNN, for data classification, the method comprising the implementation of steps by data processing means (11a, 11 b, 11c) of at least one server (1a, 1b, 1c), of: (a1) Learning, from a base of already-classified confidential learning data, the parameters of a first CNN; (a2) Learning, from a base of already-classified public learning data, the parameters of a last fully-connected layer (FC) of a second CNN corresponding to the first CNN to which said fully-connected layer (FC) has been added. The present invention also concerns a method for classifying an input datum.
    Type: Grant
    Filed: March 18, 2019
    Date of Patent: October 17, 2023
    Assignee: IDEMIA IDENTITY & SECURITY FRANCE
    Inventors: Herve Chabanne, Vincent Despiegel, Anouar Mellakh
  • Publication number: 20230289565
    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.
    Type: Application
    Filed: March 6, 2023
    Publication date: September 14, 2023
    Applicant: IDEMIA IDENTITY & SECURITY FRANCE
    Inventors: Hervé CHABANNE, Linda GUIGA, Vincent DESPIEGEL, Stephane GENTRIC
  • Publication number: 20230196073
    Abstract: A method for secure use of a first neural network on an input datum, the method comprising implementing, by data processing circuitry of a terminal: (a) constructing a second neural network which corresponds to the first neural network and receives at least one convolutional neural network approximating the identity function, (b) using the second neural network on the input datum. Further including a method for training parameters of the second neural network.
    Type: Application
    Filed: May 14, 2021
    Publication date: June 22, 2023
    Applicant: IDEMIA IDENTITY & SECURITY FRANCE
    Inventors: Hervé CHABANNE, Linda GUIGA
  • Publication number: 20230123760
    Abstract: A method for checking the identity of a reference individual, the method comprising the following steps, implemented by a checking device: selecting terminals respectively associated with individuals forming part of a set of individuals whose identities are intended to be checked by the checking device, the individual forming part of the set of individuals; sending, to each of the selected terminals, an input datum associated with the reference individual and a request asking the terminal to implement a first cryptographic processing operation producing an output datum from the input datum and from a private key specific to the individual associated with the terminal; receiving each output datum; and implementing a second cryptographic processing operation producing a check result relating to the reference individual from each output datum.
    Type: Application
    Filed: October 14, 2022
    Publication date: April 20, 2023
    Inventors: Hervé CHABANNE, Vincent DESPIEGEL
  • Patent number: 11574180
    Abstract: The present invention relates to a method for learning parameters of a convolutional neural network, CNN, for data classification, the method comprising the implementation, by means for processing data (11) of a server (1), of steps consisting of: (a1) Learning, from an already classified learning database, the parameters of a CNN, called quantized CNN, such that said parameters are valued in a discrete space; (a2) Generating a white-box implementation of at least one layer of said quantized CNN, said white-box implementation being predetermined based on at least one of said learned parameters. The present invention also relates to a method for classifying an input datum.
    Type: Grant
    Filed: January 15, 2019
    Date of Patent: February 7, 2023
    Assignee: IDEMIA IDENTITY & SECURITY FRANCE
    Inventors: Herve Chabanne, Vincent Despiegel, Anouar Mellakh
  • Publication number: 20230033479
    Abstract: A method for processing personal data, comprising the steps of: (a) For each reference personal data of a reference personal database, calculating in the encrypted domain a similarity rate of the reference personal data with a candidate personal data; said reference personal database being associated with a first partition into a plurality of first sets of reference personal data, and with a second partition into a plurality of second sets of reference personal data, such that each reference personal data of a reference personal database belongs to a single first set and a single second set; (b) For each first set and each second set, calculating an overall similarity rate of said set based on the similarity rates of the reference personal data of said set; (c) Comparing each overall similarity rate of a first and second set with a first and second predetermined threshold, respectively.
    Type: Application
    Filed: July 28, 2022
    Publication date: February 2, 2023
    Inventors: Vincent DESPIEGEL, Hervé CHABANNE
  • Patent number: 11568199
    Abstract: A method of secure classification of input data by a convolutional neural network (CNN), including (a) determination, by application of the CNN to the input data, of a first classification vector associating with each of a plurality of potential classes a representative integer score of the probability of the input data belonging to the potential class, the first vector corresponding to one possible vector, each possible vector of the first set associating with each of the plurality of potential classes an integer score; (b) construction, from the first vector, of a second classification vector of the input data, such that the second vector also belongs to the first space of possible vectors and has a distance with the first vector according to a given distance function equal to a non-zero reference distance; and return of the second vector as result of the secure classification.
    Type: Grant
    Filed: October 2, 2019
    Date of Patent: January 31, 2023
    Assignee: IDEMIA IDENTITY & SECURITY FRANCE
    Inventors: Vincent Despiegel, Hervé Chabanne
  • Publication number: 20230008719
    Abstract: A method for comparing a first and a second databases to determine whether an individual is represented by both an element of the first database and an element of the second database, wherein said elements are biometric data, including the implementation of the steps applying a classification model for each element of each database so as to construct a set of first and second bins of the respective first and second databases, each bin bringing together similar elements, each first bin being associated with a second bin; comparing the elements from the first database for at least one pair of an associated first bin and second bin belonging to said first bin with the elements from the second database belonging to said second bin, at least one of the first and the second databases then being encrypted homomorphically.
    Type: Application
    Filed: July 1, 2022
    Publication date: January 12, 2023
    Inventors: Hervé CHABANNE, Vincent DESPIEGEL
  • Publication number: 20220391327
    Abstract: The invention relates to a method for enrolling data in order to verify the authenticity of a security datum, the method comprising implementing by data processing means of a server the steps of: obtaining a reference security datum, generating a first encoded datum by applying to the reference security datum an obfuscated fuzzy Hamming distance encoding procedure, determining from the reference security datum, a plurality of derived data of the reference security datum, generating a first random datum, and determining a second encoded datum such that a variable point comparison predicate parameterized by the second encoded datum and the first random datum is true when said variable point has as coordinates said derived data, storing on a data storage means of the server at least said first and second encoded data. The invention also relates to a verification method and a server for this purpose.
    Type: Application
    Filed: May 27, 2022
    Publication date: December 8, 2022
    Inventors: Hervé CHABANNE, Linda GUIGA, Sébastien BAHLOUL
  • Publication number: 20220376918
    Abstract: The invention proposes a method for processing personal data, having the steps of (a) Functional encryption of candidate personal data using a functional encryption public key, (b) For at least one reference personal data, functional decryption of the encrypted candidate biometric data using a functional decryption private key for the polynomial function of degree 1 or 2 parameterized with said reference personal data.
    Type: Application
    Filed: May 17, 2022
    Publication date: November 24, 2022
    Inventors: Hervé CHABANNE, Vincent DESPIEGEL
  • Patent number: 11507690
    Abstract: The present invention comprises an enrolment method comprising steps of: acquisition (100) of an image showing a photograph itself showing an individual, extraction (102), from the image of a characteristic of the image other than a biometric model, obtaining (104) personal data of the individual other than by the image-processing algorithm, generation (106) of a reference datum (W) from the characteristic of the image and the obtained personal data, calculation (108) of an encoded datum (s) by application of an encoding procedure to the reference datum (W) and to a random datum (c), calculation (110) of a hash (h(c) of the random datum, storage (112) in a database of the datum encoded (s) in association with the hash. The invention also comprises an identity-control method using such stored data.
    Type: Grant
    Filed: March 21, 2019
    Date of Patent: November 22, 2022
    Assignee: IDEMIA IDENTITY & SECURITY FRANCE
    Inventors: Herve Chabanne, Sebastien Bahloul
  • Publication number: 20220335157
    Abstract: The invention proposes a personal data processing system (1) comprising a data storage module (12) storing an encrypted reference personal data database, wherein it further comprises a hardware security module (10) storing a private key for decryption of said reference personal data and configured to implement data filtering preventing any output of personal data. The invention further provides a method for processing personal data.
    Type: Application
    Filed: April 11, 2022
    Publication date: October 20, 2022
    Inventors: Hervé CHABANNE, Jean-Christophe FONDEUR, Yannick COURQUIN
  • Patent number: 11436474
    Abstract: The present invention relates to a parameter training method for a convolutional neural network, CNN, for classifying data, the method comprising the implementation by data processing means (11c) of servers (1a, 1b, 1c) of steps of: (a1) Obtaining parameters of a set of at least one first CNN; (a2) For a first CNN of said set: Training, based on a database of already-classified public training data, parameters of a final representation block (B) of a second CNN corresponding to the first selected CNN to which said representation block (B) has been added; Retraining, based on a database of already-classified confidential training data of a secondary server (1a, 1b), parameters of the second CNN; Transmitting to the main server (1c) parameters of a third CNN corresponding to the second CNN without a final representation block (B); (a3) Replacing a first CNN of said set of first CNNs with the third CNN; (a4) Aggregating said set of at least one first CNN into a fourth CNN.
    Type: Grant
    Filed: May 4, 2020
    Date of Patent: September 6, 2022
    Assignee: IDEMIA IDENTITY & SECURITY FRANCE
    Inventors: Hervé Chabanne, Vincent Despiegel
  • Patent number: 11354388
    Abstract: A method for detecting bots in a user network (R), the method comprising the following steps: receiving (102) by a user terminal (2) an identifier (ID) associated with a network user account; transmitting (104), by the user terminal (2), the identifier (ID) to an access control system (3) configured to determine whether or not a mobile terminal owner has the right to access an area or service, the area or service being independent of the user network (R); transmitting (106), by the access control system (3) to the server (1), a representative data element supporting that the identifier (ID) has been received by the access control system (3); and using (108) by the server (1) the representative data element to determine whether the user of the account associated with the identifier (ID) is a bot or not.
    Type: Grant
    Filed: March 23, 2020
    Date of Patent: June 7, 2022
    Assignee: IDEMIA IDENTITY & SECURITY FRANCE
    Inventors: Hervé Chabanne, Vincent Bouatou
  • Publication number: 20220138527
    Abstract: Process and system for processing data by an artificial neural network comprising several pooling or convolutional layers all associated with neural matrices, including for each layer of the several successive layers obtaining a reordered matrix, obtaining a division of the reordered matrix into a plurality of contiguous submatrices having given widths and heights, and grouping execution of the individual operations to be performed for each submatrix.
    Type: Application
    Filed: November 5, 2021
    Publication date: May 5, 2022
    Inventors: Hervé CHABANNE, Linda GUIGA, Jean-Luc DANGER
  • Patent number: 11102012
    Abstract: A method for digital signing of a document using a predetermined secret key. An initial internal state is determined by application to a condensate of the document of a first white box implementation of generation of a main nonce; then a modular sum of the main nonce and of a predetermined constant. The method also determines a first internal state by application to the initial internal state of a first modular arithmetic operation, then of a modular product with exponentiation of the predetermined constant. The method then determines a second internal state by application to said condensate of a second white box implementation of generation of the main nonce; and a second modular arithmetic operation function of the first internal state, of the main signature nonce and of the secret key. It then generates a digital signature of the document from the first internal state and the second internal state.
    Type: Grant
    Filed: May 22, 2018
    Date of Patent: August 24, 2021
    Assignee: IDEMIA IDENTITY & SECURITY FRANCE
    Inventors: Victor Servant, Emmanuel Prouff, Herve Chabanne
  • Patent number: 11032079
    Abstract: A method for processing biometric data performed by a proof entity and a verification entity; the proof entity having a biometric candidate datum, a biometric reference datum, a hash value of the biometric reference datum, a hash value of the biometric candidate datum; the verification entity having only the hash value of the biometric candidate datum; the method including steps of: generation by a data-processing unit of the proof entity of a zero-knowledge proof of the assumption that the biometric candidate datum and the biometric reference datum coincide; transmission to the verification entity of said zero-knowledge proof, the hash value of the biometric candidate datum, and the hash value of the biometric reference datum; verification by a data-processing unit of the verification entity that the zero-knowledge proof is valid, and that the hash value received from the biometric candidate datum corresponds to the one the verification entity has.
    Type: Grant
    Filed: October 3, 2018
    Date of Patent: June 8, 2021
    Assignee: IDEMIA IDENTITY & SECURITY FRANCE
    Inventors: Julien Paul Keuffer, Herve Chabanne
  • Patent number: 11005647
    Abstract: The present invention relates to a method for processing an image executed by a terminal (1), comprising steps of receiving a proof datum previously input by a user of the terminal (1), setting (104, 106) of at least one parameter to a first value when the proof datum is equal to a secret reference datum, and to a second value different to the first value when the proof datum is different to the secret reference datum, and generation (200) of an output datum from an input datum being or dependent on an image previously acquired by a sensor (4), and also from the parameter, the output datum having a value as function of the value the parameter has been set to.
    Type: Grant
    Filed: November 14, 2018
    Date of Patent: May 11, 2021
    Assignee: IDEMIA IDENTITY & SECURITY FRANCE
    Inventors: Herve Chabanne, Julien Bringer
  • Patent number: 11003991
    Abstract: A method for secure learning of parameters of a convolution neural network, CNN, for data classification includes the implementation, by data processing of a first server, including receiving from a second server a base of already classified learning data, the learning data being homomorphically encrypted; learning in the encrypted domain, from the learning database, the parameters of a reference CNN including a non-linear layer (POLYNOMIAL) operating an at least two-degree polynomial function approximating an activation function; a batch normalization layer before each non-linear layer (POLYNOMIAL); and transmitting the learnt parameters to the second server, for decryption and use for classification.
    Type: Grant
    Filed: October 2, 2017
    Date of Patent: May 11, 2021
    Assignee: Idemia Identity & Security France
    Inventors: Herve Chabanne, Jonathan Milgram, Constance Morel, Emmanuel Prouff