Method for characterizing material by analysis of speckles

Patent metadata

Field	Value
Publication number	US-9830520-B2
Application number	US-201615257810-A
Country	US
Kind code	B2
Filing date	Sep 6, 2016
Priority date	Sep 7, 2015
Publication date	Nov 28, 2017
Grant date	Nov 28, 2017

Abstract

A method for characterizing material in which a surface is formed, by analysis of speckles generated on the surface by a coherent light beam by performing the computation of an autocorrelation function of the light intensity on the image and computation of the value of at least one criterion set up from said function, setting up an autocorrelation matrix of the light intensity of the image, computation of the discrete cosine transform of said matrix, and computation of the value of a characterization criterion such as the average on the image of the matrix trace obtained by the discrete cosine transform of the autocorrelation matrix, and computation of the value of at least one representative criterion of a phase shift distribution of the light beams diffused by the surface, said computation comprising computation of a wavelet transform of the image, and computation of the value of at least one criterion from the following group: average of the phases of the wavelet coefficients on the image, or standard deviation of the phases of the wavelet coefficients on the image.

First claim

The invention claimed is:

1. A method for characterizing material in which a surface is formed, comprising the steps of:
projecting a coherent light beam onto the surface to generate on said surface speckles resulting from interferences of the light beams diffused by said surface,
acquiring an image of said surface on which the speckles appear,
processing said image to compute at least one characterization criterion of the material, and
from criteria, determining the material constituting the surface,
in which the processing step of the image comprises:
computation of an autocorrelation function of the light intensity on the image and computation of the value of at least one criterion set up from said function, said computation comprising:
setting up an autocorrelation matrix of the light intensity of the image,
computation of the discrete cosine transform of said matrix, and
computation of the value of a characterization criterion such as the average on the image of the matrix trace obtained by the discrete cosine transform of the autocorrelation matrix, and
computation of the value of at least one representative criterion of a phase shift distribution of the light beams diffused by the surface, said computation comprising computation of a wavelet transform of the image, and computation of the value of at least one criterion from the following group:
average of the phases of the wavelet coefficients on the image, or
standard deviation of the phases of the wavelet coefficients on the image.

2. The characterization method according to claim 1 , wherein computation of the value of at least one characterization criterion from the autocorrelation function of the light intensity comprises:
computation of the value of at least one criterion set up from the Fourier transform of the autocorrelation function, and
computation of the value of at least one criterion set up from the discrete cosine transform of the autocorrelation function.

3. The characterization method according to claim 2 , wherein the criterion set up from the Fourier transform of the autocorrelation function comprises at least one from the following group:
average of the Fourier transform of the autocorrelation function on the image, or
standard deviation of the Fourier transform of the autocorrelation function on the image.

4. The characterization method according to claim 1 , wherein processing of the image comprises setting up an autocorrelation matrix of the light intensity of the image, and computation of the value of a characterization criterion equal to the square root of the sum of the average eigenvalues of said matrix on the image.

5. The characterization method according to claim 1 , wherein the determination step of the material constituting the surface comprises computation of an overall score from the values of the criteria, and classification of the material by comparison of the overall score with a plurality of predetermined ranges corresponding to respective materials.

6. The characterization method according to claim 5 , further comprising a preliminary step of preparing ranges of scores, said step comprising:
for a plurality of surfaces formed in determined materials, performing steps of:
projecting onto the surface a coherent light beam to generate on said surface speckles resulting from interferences of the light beams diffused by said surface,
acquiring an image of said surface on which the speckles appear,
processing said image to compute at least one characterization criterion of the material, said processing comprising computation of an autocorrelation function of the light intensity on the image and computation of the value of at least one criterion from said function, computation of the value of at least one representative criterion of a phase shift distribution of the light beams diffused by the surface, and setting up an overall score from said values,
implementing a learning mechanism on all the overall scores and all the corresponding materials to obtain the ranges corresponding to each material.

7. A biometric control method, comprising executing the method according to claim 1 on a surface presented by an individual during biometric control, the method further comprising emission of an alert in the event where the characterized material does not correspond to biological tissue.

8. A system for characterizing a material comprising:
a coherent light source,
an image sensor, and
a processing unit, comprising a processor adapted to:
process an image of a surface on which speckles appear, said processing comprising computation of an autocorrelation function of the light intensity on the image,
computing, from said images, the value of at least one characterization criterion of the material forming the surface, comprising at least one criterion computed from the autocorrelation function, said computation comprising:
setting up an autocorrelation matrix of the light intensity of the image,
computation of the discrete cosine transform of said matrix, and
computation of the value of a characterization criterion such as the average on the image of the matrix trace obtained by the discrete cosine transform of the autocorrelation matrix,
and at least one representative criterion of a phase shift distribution of the light beams diffused by the surface, computation of said criterion comprising computation of a wavelet transform of the image, and computation of the value of at least one criterion from the following group:
average of the phases of the wavelet coefficients on the image, or
standard deviation of the phases of the wavelet coefficients on the image, and
from values of criteria, determining the material constituting the surface.

9. A non-transitory computer readable medium of a processor comprising the processing of an image of a surface to compute the value of at least one characterization criterion of the material constituting the surface, said processing comprising:
computation of an autocorrelation function of the light intensity on the image and computation of the value of at least one criterion set up from said function, said computation comprising:
setting up an autocorrelation matrix of the light intensity of the image,
computation of the discrete cosine transform of said matrix, and
computation of the value of a characterization criterion such as the average on the image of the matrix trace obtained by the discrete cosine transform of the autocorrelation matrix, and
computation of the value of at least one representative criterion of a phase shift distribution of the light beams diffused by the surface, comprising computation a wavelet transform of the image, and computation of the value of at least one criterion from the following group:
average of the phases of the wavelet coefficients on the image, or
standard deviation of the phases of the wavelet coefficients on the image, when executed by a processor.

Assignees

• Safran Identity & Security

Inventors

• Maalouf Aldo
• Guillemot Florence
• Chiu Remy

Classifications

• G06V40/45

  Detection of the body part being alive · CPC title

• G06V40/40 Primary

  Spoof detection, e.g. liveness detection · CPC title

• G06V10/431

  Frequency domain transformation; Autocorrelation · CPC title

• G01N21/47

  Scattering, i.e. diffuse reflection (G01N21/25, G01N21/41 take precedence {G01N21/55 takes precedence}) · CPC title

• G01N21/25

  Colour; Spectral properties, i.e. comparison of effect of material on the light at two or more different wavelengths or wavelength bands · CPC title