Biometric authentication apparatus, biometric authentication method, and program

• US 8,831,296 B2
• Filed: 01/21/2011
• Issued: 09/09/2014
• Est. Priority Date: 01/29/2010
• Status: Expired due to Fees

• Current Assignee: Sony Corporation (Sony Group Corp.)
• Original Assignee: Sony Corporation (Sony Group Corp.)
• Inventors: Muquit, Mohammad Abdul
• Primary Examiner(s): Patel, Jayesh A
• Application Number:
  US12/931,021
• Publication Number:
  US 20110188710A1
• Time in Patent Office: 1,327 Days
• Field of Search: 382115-188
• US Class Current: 382/124
• CPC Class Codes: G06F 18/00 Pattern recognition
• CPC Class Codes: G06V 40/10 Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
• CPC Class Codes: G06V 40/1347 Preprocessing; Feature extraction
• CPC Class Codes: G06V 40/14 Vascular patterns

Abstract

There is provided a biometric authentication apparatus including a vein image extraction unit for extracting a vein image showing positions of veins from an image including veins in a finger portion, a vein image dividing unit for dividing the extracted vein image into a plurality of partial regions, a vein pixel counting unit for counting the number of pixels corresponding to the veins in each of the divided partial regions, a vector generation unit for arranging, in a predetermined order, count results of the respective partial regions, and generating a vein distribution vector, which is a numerical sequence representing a degree of distribution of the veins in the vein image, and an authentication unit for authenticating the vein distribution vector generated by the vector generation unit, based on a registered vein distribution vector, which is a vein distribution vector registered in advance.

Claims

1. A biometric authentication apparatus comprising:

one or more processors configured to;

control a light source that emit a light having a predetermined wavelength to a finger portion;

extract a vein image showing positions of veins from an image including veins in the finger portion;

divide the extracted vein image into a plurality of partial regions;

count a number of pixels corresponding to the veins in each of the partial regions;

arrange, in a predetermined order, count results of the respective partial regions;

generate a vein distribution vector, which is a numerical sequence representing a degree of distribution of the veins in the vein image;

authenticate the generated vein distribution vector based on a registered vein distribution vector, which is a vein distribution vector registered in advance;

calculate a degree of similarity between the generated vein distribution vector and the registered vein distribution vector, while one of the generated vein distribution vector and the registered vein distribution vector is repeatedly shifted by a predetermined number of elements;

correct the vein image by performing at least one of a parallel moving processing and a rotation processing on the extracted vein image;

perform a principal component analysis on a not-yet-corrected vein image, and rotate the not-yet-corrected vein image such that a direction of a first principal component is in parallel with a side in a longitudinal direction of the vein image; and

calculate a barycenter of a distribution of pixels corresponding to the positions of the veins in the not-yet-corrected vein image, and move the not-yet-corrected vein image in parallel such that the barycenter matches a central position of the vein image.

2. The biometric authentication apparatus according to claim 1,

wherein the one or more processors are configured to divide a predetermined region of the vein image including the central position of the vein image into the plurality of partial regions.

3. The biometric authentication apparatus according to claim 1,

wherein the one or more processors are configured to arrange the count results in order from a partial region located at a corner of the vein image.

4. The biometric authentication apparatus according to claim 1,

wherein the one or more processors are configured to arrange the count results in order from a partial region including a central position of the vein image.

5. The biometric authentication apparatus according to claim 1,

wherein the one or more processors are configured to perform an authentication processing based on an element of the vein distribution vector corresponding to a partial region including a central position of the vein image.

6. A biometric authentication method comprising:

emitting a light having a predetermined wavelength to a finger portion and extracting a vein image showing positions of veins from an image including veins in the finger portion from which the image has been obtained;

dividing the extracted vein image into a plurality of partial regions;

counting a number of pixels corresponding to the veins in each of the generated partial regions;

arranging, in a predetermined order, count results of the respective partial regions;

generating a vein distribution vector, which is a numerical sequence representing a degree of distribution of the veins in the vein image;

authenticating the generated vein distribution vector based on a registered vein distribution vector, which is a vein distribution vector registered in advance;

calculating a degree of similarity between the generated vein distribution vector and the registered vein distribution vector, while one of the generated vein distribution vector and the registered vein distribution vector is shifted by a predetermined number of elements;

correcting the vein image by performing at least one of a parallel moving processing and a rotation processing on the extracted vein image;

performing a principal component analysis on a not-yet-corrected vein image, and rotating the not-yet-corrected vein image such that a direction of a first principal component is in parallel with a side in a longitudinal direction of the vein image; and

calculating a barycenter of a distribution of pixels corresponding to the positions of the veins in the not-yet-corrected vein image, and moving the not-yet-corrected vein image in parallel such that the barycenter matches a central position of the vein image.

7. A non-transitory computer readable medium having stored thereon, a computer program having at least one code section executable by a computer, thereby causing the computer to perform steps comprising:

emitting a light having a predetermined wavelength to a finger portion and extracting a vein image showing positions of veins from an image including veins in the finger portion from which the image has been obtained;

dividing the extracted vein image into a plurality of partial regions;

counting a number of pixels corresponding to the veins in each of the partial regions;

arranging, in a predetermined order, count results of the respective partial regions;

generating a vein distribution vector, which is a numerical sequence representing a degree of distribution of the veins in the vein image;

authenticating the generated vein distribution vector based on a registered vein distribution vector, which is a vein distribution vector registered in advance;

calculating a degree of similarity between the generated vein distribution vector and the registered vein distribution vector, while one of the generated vein distribution vector and the registered vein distribution vector is shifted by a predetermined number of elements;

correcting the vein image by performing at least one of a parallel moving processing and a rotation processing on the extracted vein image;

performing a principal component analysis on a not-yet-corrected vein image, and rotating the not-yet-corrected vein image such that a direction of a first principal component is in parallel with a side in a longitudinal direction of the vein image; and

calculating a barycenter of a distribution of pixels corresponding to the positions of the veins in the not-yet-corrected vein image, and moving the not-yet-corrected vein image in parallel such that the barycenter matches a central position of the vein image.