Iris recognition system and method using multifocus image sequence

• US 20080037835A1
• Filed: 02/09/2007
• Published: 02/14/2008
• Est. Priority Date: 06/02/2006
• Status: Active Grant

• Current Assignee: Industry Academic Cooperation Foundation
• Original Assignee: Korea Advanced Institute of Science and Technology
• Inventors: Son, Byungjun, Lee, Yillbyung
• Granted Patent: US 7,925,058 B2
• Time in Patent Office: Days
• US Class Current: 382/117
• CPC Class Codes: G06V 40/19 Sensors therefor
• CPC Class Codes: G06V 40/193 Preprocessing; Feature extraction

Abstract

The present invention relates to an iris recognition system and method using a multifocus image sequence, the iris recognition system including: an eye image input unit calculating the distance between an image obtaining apparatus and a user using a distance sensor to receive a multifocus eye image sequence having focus information of the user'"'"'s eye varied at regular time intervals within a predetermined distance using the image obtaining apparatus; an eye image pretreatment unit calculating a fuzzy entropy for the multifocus eye image sequence input from the eye image input unit to select eye images of N number that can effectively express eye image data of various focusing levels, and extracting a multifocus iris image sequence by separating an iris image between pupil and collarette that has various iris patterns, from which an area unnecessary for configuring a feature vector is removed, from the selected eye images; a feature vector extraction unit multi-dividing the multifocus iris image sequence extracted in the eye image pretreatment unit, forming a wavelet feature vector from the multi-divided images using a subband including only a low-frequency area, and generating a wavelet feature vector reduced to a dimension having a high discrimination performance and a dimension-reducing transform function P; a recognition unit discriminating the user'"'"'s authenticity by measuring a dissimilarity between the feature vector extracted from the feature vector extraction unit and a feature vector already registered; a register unit registering the reduced wavelet feature vector extracted from the feature vector extraction unit and the dimension-reducing transform function P in a storage unit; and a storage unit storing the dimension-reducing transform function P and the reduced wavelet feature vector. Accordingly, the iris recognition system and method using a multifocus image sequence can effectively recognize low-resolution images and poorly focused images as well as high-resolution images and well focused images. Moreover, the iris recognition system and method using a multifocus image sequence can extract the feature vector having a high discrimination performance, enhance the efficiency of a storage device by effectively reducing the dimension of the feature vector, and minimize recognition errors even using a simple recognition apparatus using a wavelet analysis and a direct linear discriminant analysis.

Claims

1. An iris recognition system using a multifocus image sequence comprising:

an eye image input unit calculating the distance between an image obtaining apparatus and a user using a distance sensor to receive a multifocus eye image sequence having focus information of the user'"'"'s eye varied at regular time intervals within a predetermined distance using the image obtaining apparatus;

an eye image pretreatment unit calculating a fuzzy entropy for the multifocus eye image sequence input from the eye image input unit to select eye images of N number that can effectively express eye image data of various focusing levels, and extracting a multifocus iris image sequence by separating an iris image between pupil and collarette that has various iris patterns, from which an ar ea unnecessary for configuring a feature vector is removed, from the selected eye images;

a feature vector extraction unit multi-dividing the multifocus iris image sequence extracted in the eye image pretreatment unit, forming a wavelet feature vector from the multi-divided images using a subband including only a low-frequency area, and generating a wavelet feature vector reduced to a dimension having a high discrimination performance and a dimension-reducing transform function P;

a recognition unit discriminating the user'"'"'s authenticity by measuring a dissimilarity between the feature vector extracted from the feature vector extraction unit and a feature vector already registered;

a register unit registering the reduced wavelet feature vector extracted from the feature vector extraction unit and the dimension-reducing transform function P in a storage unit; and

a storage unit storing the dimension-reducing transform function P and the reduced wavelet feature vector.

2. The iris recognition system using a multifocus image sequence as recited in claim 1,

wherein the image obtaining apparatus comprises:

a CCD camera in which an infrared transmission filter and a macro lens are attached in front of a camera lens to ensure a recognizable iris pattern and to avoid losing iris image data due to noise images;

an infrared light unit composed an infrared light emitting diode (LED) that irradiates light toward the front of the CCD camera; and

a frame grabber converting analog signals of an eye image photographed by the CCD camera into digital signals.

3. An iris recognition method using a multifocus image sequence comprising:

an image obtaining step (S200) of receiving a multifocus eye image sequence from an image obtaining apparatus;

a pretreatment step (S300) of calculating a fuzzy entropy from the multifocus eye image sequence obtained from the image obtaining step (S200) to select eye images and extracting a multifocus iris image sequence by separating an iris image between pupil and collarette from the selected eye images;

a feature vector extraction step (S400) of multi-dividing the multifocus iris image sequence extracted in the pretreatment step (S300), forming a wavelet feature vector from a subband including only low-frequency components among the multi-divided subbands, and extracting a reduced wavelet feature vector of an iris area for the personal identification from the wavelet feature vector thus formed using a dimension-reducing transform function P stored in a storage unit; and

a recognition step (S500) of discriminating a user'"'"'s authenticity by measuring a dissimilarity between the wavelet feature vector extracted in the feature vector extraction step (S400) and the feature vector registered in the storage unit.

4. The iris recognition method using a multifocus image sequence as recited in claim 3,

wherein the pretreatment step (S300) comprises:

a step (S310) of calculating a fuzzy entropy from the multifocus eye image sequence to select eye images that can effectively utilize eye image data of various focusing levels;

a step (S320) of selecting eye images of N number having a low fuzzy entropy value from the calculated fuzzy entropy;

a step (S330) of detecting an inner boundary and an outer boundary from the selected eye images;

a step (S340) of detecting a collarette boundary for an iris area between the detected inner boundary and the outer boundary; and

a step (S350) of generating a normalized iris image for the iris area between the detected inner boundary and the collarette boundary.

5. The iris recognition method using a multifocus image sequence as recited in claim 3,

wherein the feature vector extraction step (S400) comprises:

a step (S410) of receiving a multifocus iris image sequence composed of the normalized iris images extracted in the pretreatment step (S300);

a step (S420) of multi-dividing the input multifocus iris image sequence using a biorthogonal wavelet transform;

a step (S430) of extracting an area LL including only low-frequency components from the subbands of the multi-divided iris images;

a step (S440) of determining whether or not the number of the multi-division is greater than a predetermined number;

a step (S450) of selecting the area LL including only low-frequency components extracted in the above step (S430) as object images for performing a new multi-division, if the number of the multi-division is smaller than the predetermined number in the above step (S430);

a step (S460) of forming a wavelet feature vector from the area LL obtained by a final multi-division, if the number of the multi-division is greater than the predetermined number in the above step (S430);

a step (S470) of extracting a dimension-reducing transform function P for generating an iris feature vector reduced to a dimension having a high discrimination performance from the extracted wavelet feature vectors using a direct linear discriminant analysis; and

a step (S480) of storing the dimension-reducing transform function P and the wavelet feature vector reduced by the dimension-reducing transform function P in the storage unit.

6. The iris recognition method using a multifocus image sequence as recited in claim 3 or 5,

wherein the feature vector extraction step (S400) further comprises:

a step of (S460) of generating a wavelet feature vector using a biorthogonal wavelet transform that applies a symmetric filter;

a step (S461) of searching the dimension-reducing transform function P already registered so as to reduce the extracted wavelet feature vector to a dimension having a high discrimination performance; and

a step (S462) of generating a reduced wavelet feature vector applying the wavelet feature vector to the searched dimension-reducing transform function P.

7. The iris recognition method using a multifocus image sequence as recited in claim 3,

wherein the recognition step (S500) comprises:

a step (S510) of receiving the reduced wavelet feature vectors obtained in the feature vector extraction step (S400);

a step (S520) of searching the registered feature vectors to compare the input wavelet feature vectors with the already registered feature vectors;

a step (S530) of specifying a dissimilarity by a weighted Euclidean distance between the input wavelet feature vectors and the already registered in the feature vectors;

a step (S540) of comparing the measured dissimilarity with a predetermined reference value; and

steps (S550 and S560) of rejecting and recognizing the user based on the compared results.