US 8,509,495 B2Granted

Subcutaneous vein pattern detection via multi-spectral IR imaging in an identity verification system

AssigneeXerox Corporation

InventorsBeilei Xu, Lalit Keshav Mestha

Life of this patent

Estimating the term…

Loading patent…

Abstract

What is disclosed is a novel system and method for identifying an individual in an IR image involves the following. Intensity values are collected at N wavelengths for each pixel in an IR video-based image. The intensity values are collected using an IR imaging system having an IR detector and an IR Illuminator. The intensity values are then used to identify pixels of human skin in the IR image. If human skin is identified in the IR image then, the human hand is identified in the IR image from the human skin to distinguish the hand from the background. Vein patterns in the hand are then located and extracted. A reference vein pattern is retrieved from a database of known vein patterns for individuals, and a comparison is made to determine a match. If a match is determined, then the individual in the captured IR image can be identified.

Claims24

Dependent claims

A method for identifying an individual in an IR image or IR video, the method comprising:

collecting intensity values at N wavelength bands for each pixel in any of:
an IR image and IR video, said intensity values being collected using a IR imaging system comprising an IR detector and an IR Illuminator, wherein said IR detector is a single IR detector, said intensity value comprising:
Ic(j)=α∫λ1λ2Is(λ)[TG2(λ)Ro(λ)+ηRG(λ)]TL(λ)D(λ)dλ+Ib,
where j=1, . . . , N, such that j is the jth IR band from the illuminator that is illuminating, α is a constant that depends on an angle and distance from the illumination source, an attenuation of an IR wave in the air, and an integration time of the detecting device, Ib is an intensity from a background light source such as that from the IR component of sunlight, Is(λ) is a calculated correspondence of a measured intensity, Im, such that, if the calculated intensity with a particular reflectance agrees with the object in the measured intensity, the correlation will be high (close to 1), otherwise the correlation will be small (close to 0 or negative), TL(λ) is a transmittance of the filter, Ro(λ) is a reflectance of an object detected by the IR detection device, Ro(λ) and TG(λ) are a reflectance and a transmittance of glass or other types of material between the target object and the illumination source and camera, otherwise Ro(λ)=0 and TG(λ)=1, constant η is a measure of the percentage of light from the IR illuminator reflected from glass and received by the detector, otherwise η=0, and where TL(λ)=1, and D(λ) is a responsivity of the detection device;
using said intensity values to identify pixels of human tissue in said IR image or said IR video;
extracting a subcutaneous vein pattern from said identified human tissue;
retrieving a reference vein pattern from a database;
and
comparing said extracted subcutaneous vein pattern with said reference vein pattern to identify the individual in said IR image.

6 more elements

A method for identifying an individual in an IR image or IR video, the method comprising:

collecting intensity values at N wavelength bands for each pixel in any of:
an IR image and IR video, said intensity values being collected using a IR imaging system comprising an IR detector and an IR Illuminator, wherein said IR detector has N band pass filters (N≧1), and said IR Illuminator comprises a single IR illuminator covering a wavelength range of those filters, said intensity value comprising:
Ic(i)=α∫λ1λ2Is(λ)[TG2(λ)Ro(λ)+ηRG(λ)]TLi(λ)D(λ)dλ+Ib,
where i=1, . . . , N, such that i is the ith IR band pass filter, α is a constant that depends on an angle and distance from the illumination source, an attenuation of an IR wave in the air, and an integration time of the detecting device, Ib is a background intensity, Is(λ) is a calculated correspondence of a measured intensity, Im, such that, if the calculated intensity with a particular reflectance agrees with the object in the measured intensity, the correlation will be high (close to 1), otherwise the correlation will be small (close to 0 or negative), Ro(λ) is a reflectance of an object detected by the IR detection device, RG(λ) and TG(λ) are a reflectance and a transmittance of glass, otherwise RG(λ)=0 and TG(λ)=1, constant η is a measure of the percentage of light from the illuminator reflected from glass and received by the detector, otherwise η=0, TLi(λ) is a transmittance of the ith (filter, and D(λ) is a responsivity of the detecting device;
using said intensity values to identify pixels of human tissue in said IR image or said IR video;
extracting a subcutaneous vein pattern from said identified human tissue;
retrieving a reference vein pattern from a database;
and
comparing said extracted subcutaneous vein pattern with said reference vein pattern to identify the individual in said IR image.

6 more elements

A method for identifying an individual in an IR image or IR video, the method comprising:

collecting intensity values at N wavelength bands for each pixel in any of:
an IR image and IR video, said intensity values being collected using a IR imaging system comprising an IR detector and an IR Illuminator, wherein said IR detector comprises a total of N detectors (N≧1) having a total of I filters (I≧1), and the IR Illuminator comprises a total of J illuminators (J≧1) such that the combination of I filters and J illuminators covers a desired wavelength range, said intensity value comprising:
Ic(i)=α∫λ1λ2Isj(λ)[TG2(λ)Ro(λ)+ηRG(λ)]TL(λ)D(λ)dλ+Ib,
where i=1, . . . , I, such that i is the ith IR band from the jth illuminator that is illuminating, α is a constant that depends on an angle and distance from the illumination source, an attenuation of an IR wave in the air, and an integration time of the detecting device, Ib is an intensity from a background light source such as that from the IR component of sunlight, Isj(λ) is a calculated correspondence of a measured intensity, Imj, from the jth illuminator such that, if the calculated intensity with a particular reflectance agrees with the object in the measured intensity, the correlation will be high (close to 1), otherwise the correlation will be small (close to 0 or negative), Ro(λ) is a reflectance of an object detected by the IR detection device, TL(λ) is a transmittance of the filter, RG(λ) and TG(λ) are a reflectance and a transmittance of glass, otherwise RG(λ)=0 and TG(λ)=1, constant η is a measure of the percentage of light from the IR illuminator reflected from glass and received by the detector, otherwise η=0, and where TL(λ)=1, and D(λ) is a responsivity of the detection device;
using said intensity values to identify pixels of human tissue in said IR image or said IR video;
extracting a subcutaneous vein pattern from said identified human tissue;
retrieving a reference vein pattern from a database;
and
comparing said extracted subcutaneous vein pattern with said reference vein pattern to identify the individual in said IR image.

6 more elements

A method for identifying an individual in an IR image or IR video, the method comprising:

collecting intensity values at N wavelength bands for each pixel in any of:
an IR image and IR video, said intensity values being collected using a IR imaging system comprising an IR detector and an IR Illuminator, wherein said IR detector comprises a total of N narrow band detection devices (N≧1) having a total of I filters (I≧1) and the IR Illuminator comprises a total of J illuminators (J≧1), said intensity value comprising:
Ic(i)=α∫λ1λ2Is(λ)[TG2(λ)Ro(λ)+ηRG(λ)]TL(λ)Di(λ)dλ+Ib,
where i=1, . . . , I, such that i is the ith IR band from the jth illuminator that is illuminating, α is a constant that depends on an angle and distance from the illumination source, an attenuation of an IR wave in the air, and an integration time of the detecting device, Ib is an intensity from a background light source such as that from the IR component of sunlight, Is(λ) is a calculated correspondence of a measured intensity, Im, such that, if the calculated intensity with a particular reflectance agrees with the object in the measured intensity, the correlation will be high (close to 1), otherwise the correlation will be small (close to 0 or negative), Ro(λ) is a reflectance of an object detected by the IR detection device, TL(λ) is a transmittance of the filter, RG(λ) and TG(λ) are a reflectance and a transmittance of glass, otherwise RG(λ)=0 and TG(λ)=1, constant η is a measure of the percentage of light from the IR illuminator reflected from glass and received by the detector, otherwise η=0, and TL(λ)=1, and Di(λ) is a responsivity of the detection device at the ith band;
using said intensity values to identify pixels of human tissue in said IR image or said IR video;
extracting a subcutaneous vein pattern from said identified human tissue;
retrieving a reference vein pattern from a database;
and
comparing said extracted subcutaneous vein pattern with said reference vein pattern to identify the individual in said IR image.

6 more elements

A system for identifying an individual in an IR image or IR video, the system comprising:

an IR imaging system comprising an IR detection device and an IR Illuminator;
a memory and a storage medium;
and
a processor in communication with and said storage medium and said memory, said processor executing machine readable instructions for performing the method of:
using said IR imaging system to collect intensity values at N wavelength bands for each pixel in any of:
an IR image and IR video, wherein said IR detector is a single IR detector, said intensity value comprising:
Ic(j)=α∫λ1λ2Is(λ)[TG2(λ)Ro(λ)+ηRG(λ)]TL(λ)D(λ)dλ+Ib,
where j=1, . . . , N, such that j is the jth IR band from the illuminator that is illuminating, α is a constant that depends on an angle and distance from the illumination source, an attenuation of an IR wave in the air, and an integration time of the detecting device, Ib is an intensity from a background light source such as that from the IR component of sunlight, Is(λ) is a calculated correspondence of a measured intensity, Im, such that, if the calculated intensity with a particular reflectance agrees with the object in the measured intensity, the correlation will be high (close to 1), otherwise the correlation will be small (close to 0 or negative), TL(λ) is a transmittance of the filter, Ro(λ) is a reflectance of an object detected by the IR detection device, Ro(λ) and TG(λ) are a reflectance and a transmittance of glass or other types of material between the target object and the illumination source and camera, otherwise Ro(λ)=0 and TG(λ)=1, constant η is a measure of the percentage of light from the IR illuminator reflected from glass and received by the detector, otherwise η=0, and where TL(λ)=1, and D(λ) is a responsivity of the detection device;
using said intensity values to identify pixels of human tissue in said IR image or said IR video;
extracting a subcutaneous vein pattern from said identified human tissue;
retrieving a reference vein pattern from a database;
and
comparing said extracted subcutaneous vein pattern with said reference vein pattern to identify the individual in said IR image.

10 more elements

A system for identifying an individual in an IR image or IR video, the system comprising:

an IR imaging system comprising an IR detection device and an IR Illuminator;
a memory and a storage medium;
and
a processor in communication with and said storage medium and said memory, said processor executing machine readable instructions for performing the method of:
using said IR imaging system to collect intensity values at N wavelength bands for each pixel in any of:
an IR image and IR video, wherein said IR detector has N band pass filters (N≧1), and said IR Illuminator comprises a single IR illuminator covering a wavelength range of those filters, said intensity value comprising:
Ic(i)=α∫λ1λ2Is(λ)[TG2(λ)Ro(λ)+ηRG(λ)]TLi(λ)D(λ)dλ+Ib,
where i=1, . . . , N, such that i is the ith IR band pass filter, α is a constant that depends on an angle and distance from the illumination source, an attenuation of an IR wave in the air, and an integration time of the detecting device, Ib is a background intensity, Is(λ) is a calculated correspondence of a measured intensity, Im, such that, if the calculated intensity with a particular reflectance agrees with the object in the measured intensity, the correlation will be high (close to 1), otherwise the correlation will be small (close to 0 or negative), Ro(λ) is a reflectance of an object detected by the IR detection device, RG(λ) and TG(λ) are a reflectance and a transmittance of glass, otherwise RG(λ)=0 and TG(λ)=1, constant η is a measure of the percentage of light from the illuminator reflected from glass and received by the detector, otherwise η=0, TLi(λ) is a transmittance of the ith |filter, and D(λ) is a responsivity of the detecting device;
using said intensity values to identify pixels of human tissue in said IR image or said IR video;
extracting a subcutaneous vein pattern from said identified human tissue;
retrieving a reference vein pattern from a database;
and
comparing said extracted subcutaneous vein pattern with said reference vein pattern to identify the individual in said IR image.

10 more elements

A system for identifying an individual in an IR image or IR video, the system comprising:

an IR imaging system comprising an IR detection device and an IR Illuminator;
a memory and a storage medium;
and
a processor in communication with and said storage medium and said memory, said processor executing machine readable instructions for performing the method of:
using said IR imaging system to collect intensity values at N wavelength bands for each pixel in any of:
an IR image and IR video, wherein said IR detector comprises a total of N detectors (N≧1) having a total of I filters (I≧1), and the IR Illuminator comprises a total of J illuminators (J≧1) such that the combination of I filters and J illuminators covers a desired wavelength range, said intensity value comprising:
Ic(i)=α∫λ1λ2Isj(λ)[TG2(λ)Ro(λ)+ηRG(λ)]TL(λ)D(λ)dλ+Ib,
where i=1, . . . , I, such that i is the ith IR band from the jth illuminator that is illuminating, α is a constant that depends on an angle and distance from the illumination source, an attenuation of an IR wave in the air, and an integration time of the detecting device, Ib is an intensity from a background light source such as that from the IR component of sunlight, Isj(λ) is a calculated correspondence of a measured intensity, Imj, from the jth illuminator such that, if the calculated intensity with a particular reflectance agrees with the object in the measured intensity, the correlation will be high (close to 1), otherwise the correlation will be small (close to 0 or negative), Ro(λ) is a reflectance of an object detected by the IR detection device, TL(λ) is a transmittance of the filter, RG(λ) and TG(λ) are a reflectance and a transmittance of glass, otherwise RG(λ)=0 and TG(λ)=1, constant η is a measure of the percentage of light from the IR illuminator reflected from glass and received by the detector, otherwise η=0, and where TL(λ)=1, and D(λ) is a responsivity of the detection device;
using said intensity values to identify pixels of human tissue in said IR image or said IR video;
extracting a subcutaneous vein pattern from said identified human tissue;
retrieving a reference vein pattern from a database;
and
comparing said extracted subcutaneous vein pattern with said reference vein pattern to identify the individual in said IR image.

10 more elements

A system for identifying an individual in an IR image or IR video, the system comprising:

an IR imaging system comprising an IR detection device and an IR Illuminator;
a memory and a storage medium;
and
a processor in communication with and said storage medium and said memory, said processor executing machine readable instructions for performing the method of:
using said IR imaging system to collect intensity values at N wavelength bands for each pixel in any of:
an IR image and IR video, wherein said IR detector comprises a total of N narrow band detection devices (N≧1) having a total of I filters (I≧1) and the IR Illuminator comprises a total of J illuminators (J≧1), said intensity value comprising:
Ic(i)=α∫λ1λ2Is(λ)[TG2(λ)Ro(λ)+ηRG(λ)]TL(λ)Di(λ)dλ+Ib,
where i=1, . . . , I, such that i is the ith IR band from the jth illuminator that is illuminating, α is a constant that depends on an angle and distance from the illumination source, an attenuation of an IR wave in the air, and an integration time of the detecting device, Ib is an intensity from a background light source such as that from the IR component of sunlight, Is(λ) is a calculated correspondence of a measured intensity, Im, such that, if the calculated intensity with a particular reflectance agrees with the object in the measured intensity, the correlation will be high (close to 1), otherwise the correlation will be small (close to 0 or negative), Ro(λ) is a reflectance of an object detected by the IR detection device, TL(λ) is a transmittance of the filter, RG(λ) and TG(λ) are a reflectance and a transmittance of glass, otherwise RG(λ)=0 and TG(λ)=1, constant η is a measure of the percentage of light from the IR illuminator reflected from glass and received by the detector, otherwise η=0, and TL(λ)=1, and Di(λ) is a responsivity of the detection device at the ith band;
using said intensity values to identify pixels of human tissue in said IR image or said IR video;
extracting a subcutaneous vein pattern from said identified human tissue;
retrieving a reference vein pattern from a database;
and
comparing said extracted subcutaneous vein pattern with said reference vein pattern to identify the individual in said IR image.

10 more elements

Dependencies are inferred from the claim text.

Facts

Application: 13/087,850Filed 15 Apr 2011
Published: US 2012/0263357 A118 Oct 2012
Granted: 13 Aug 2013
Term ends: Not estimatedThe source records don’t support an estimate.

Classification4 codes

G06V 40/13
Fingerprints or palmprints: sensors therefor
G06V 10/145
Optical characteristics of the device performing the acquisition or on the illumination arrangements: illumination specially adapted for pattern…
G06V 40/14
Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands: vascular patterns
G06V
Image and video recognition