A novel approach for selecting a rectangular template around periocular region optimally potential for human recognition is proposed. A comparatively larger template of periocular image than the optimal one can be slightly more potent for recognition, but the larger template heavily slows down the biometric system by making feature extraction computationally intensive and increasing the database size. A smaller template, on the contrary, cannot yield desirable recognition though the smaller template performs faster due to low computation for feature extraction. These two contradictory objectives (namely, (a) to minimize the size of periocular template and (b) to maximize the recognition through the template) are aimed to be optimized through the proposed research. This paper proposes four different approaches for dynamic optimal template selection from periocular region. The proposed methods are tested on publicly available unconstrained UBIRISv2 and FERET databases and satisfactory results have been achieved. Thus obtained template can be used for recognition of individuals in an organization and can be generalized to recognize every citizen of a nation.
A biometric system comprises a physical or behavioral trait of a person through which he or she can be recognized uniquely. Computer aided identification of a person through face biometric has grown its importance through the last decade and researchers have attempted to find unique facial nodal points. However, change of facial data with expression and age makes it challenging for recognition through face. A stringent necessity to identify a person on partial facial data has been felt in such scenario. There are forensic applications where antemortem information is a partial face. These motives led researchers to derive auxiliary biometric traits from facial image, namely, iris, ear, lip, and periocular region. Recognizing human through iris captured under near infrared (NIR) illumination and constrained scenario yields satisfactory recognition accuracy while recognition under visual spectrum (VS) and unconstrained scenario is relatively challenging. In particular, VS periocular image has been exploited to examine its uniqueness as there exists many nodal points. Classification and recognition through periocular region show significant accuracy, given the fact that periocular biometric uses only approximately 10% of a complete face data (illustrated in Section
Working model of periocular biometric system.
Periocular (peripheral area of ocular) region refers to the immediate vicinity of the eye, including eyebrow and lower eye fold as depicted in Figure
Comparison of biometric traits present in human face.
Trait | Advantages | Possible challenges |
---|---|---|
Iris | High-dimensional feature can be extracted, difficult to spoof, permanence of iris, secured within eye folds, and can be captured in noninvasive way | Yields accuracy in NIR images than VS images, cost of NIR acquisition device is high, low recognition accuracy in unconstrained scenarios, low recognition accuracy for low resolution, occlusion due to use of lens, eye may close at the time of capture, do not work for keratoconus and keratitis patients |
|
||
Face | Easy to acquire, yields accuracy in VS images, most available in criminal investigations | Not socially acceptable for some religions, full face image makes database large, variation with expression and age |
|
||
Periocular | Can be captured with face/iris region without extra acquisition cost | Can be occluded by spectacle, less features in case of infants |
|
||
Lip | Existence of both global and local features | Difficult to acquire, less acceptable socially, shape changes with human expression |
|
||
Ear | Easy segmentation due to presence of contrast in the vicinity | Difficult to acquire and can be partially occluded by hair |
Important features from a periocular image.
This paper approaches to fit an optimal boundary to the periocular region which is sufficient and necessary for recognition. Unlike other biometric traits, edge information is not the required criteria to exactly localize periocular region. Rather periocular region can be localized where the periphery of eye contains no further information. Researchers have considered a static rectangular boundary around the eye to recognize human and termed the localized rectangle as periocular region. However, this approach is naive as the same static boundary does not work for every face image (e.g., when the face image is captured through different distances from the camera, or when there is a tilt of face or camera during acquisition). So there is a need of deriving a dynamic boundary to describe periocular region. While deciding the periocular boundary, the objective of achieving the highest recognition accuracy also needs to be maintained. The paper specifies few metrics through which periocular region can be optimally localized in scale and rotation invariant manner.
The rest of the paper is organized as follows: Section
Investigations have been made by researchers in the direction of localizing iris from high quality constrained eye images captured in NIR illumination. Table
Performance comparison of some benchmark NIR iris localization approaches.
Year | Authors | Approach | Testing database | Accuracy results |
---|---|---|---|---|
2002 |
Camus and Wildes [ |
Multiresolution coarse-to-fine strategy | Constrained iris images (640 without glasses, 30 with glasses) | Overall 98% (99.5% for subjects without glasses and 66.6% for subjects wearing glasses) |
|
||||
2004 | Sung et al. [ |
Bisection method, canny edge-map detector, and histogram equalization | 3,176 images acquired through a CCD camera | 100% inner boundary and 94.5% for collarette boundary |
|
||||
2004 | Bonney et al. [ |
Least significant bit plane and standard deviations | 108 images from CASIA v1 and 104 images from UNSA | Pupil detection 99.1% and limbic detection 66.5% |
|
||||
2005 | Liu et al. [ |
Modification to Masek’s segmentation algorithm | 317 gallery and 4,249 probe images acquired using Iridian LG 2200 iris imaging system | 97.08% rank-1 recognition |
|
||||
2006 |
Proença and Alexandre [ |
Moment functions dependent on fuzzy clustering | 1,214 good quality, 663 noisy images from 241 subjects in two sessions | 98.02% on good data set and 97.88% on noisy data set |
|
||||
2008 | Pundlik et al. [ |
Markov random field and graph cut | WVU nonideal database | Pixel label error rate 5.9% |
|
||||
2009 | He et al. [ |
Adaboost-cascade iris detector for iris center prediction | NIST Iris Challenge Evaluation (ICE) v 1.0, CASIA-Iris-V3-lamp, UBIRISv1.0 | 0.53% EER for ICEv1.0 and 0.75% EER for CASIA Iris-V3-lamp |
|
||||
2010 |
Liu et al. [ |
|
CASIAv3 and UBIRISv2.0 | 1.9% false positive and 21.3% false negative (on a fresh data set not used to tune the system) |
|
||||
2010 | Tan et al. [ |
Gray distribution features and gray projection | CASIAv1 | 99.14% accuracy (processing time 0.484 s/image) |
|
||||
2011 | Bakshi et al. [ |
Image morphology and connected component analysis | CASIAv3 | 95.76% accuracy with processing (0.396 s/image) |
The task of recognition is more challenging than classification and hence draws more attention. The most commonly used feature extraction techniques in context of periocular recognition are Scale Invariant Feature Transform, Local Binary Pattern. Tables Will the accuracy obtained from this arbitrary boundary increase if a larger region is considered? How much of the considered periocular region is actually contributing to recognition? Is there any portion within this arbitrary considered periocular region which can be removed and still comparable accuracy can be achieved?
Survey on classification through periocular biometric.
Authors | Classification type | Algorithm | Classifier | Testing database | Accuracy (%) |
---|---|---|---|---|---|
Abiantun and Savvides [ |
Left versus right eye | Adaboost, Haar, Gabor features | LDA, SVM | ICE | 89.95% |
Bhat and Savvides [ |
Left versus right eye | ASM | SVM | ICE, LG | Left eye 91%, right eye 89% |
Merkow et al. [ |
Gender | LBP | LDA, SVM, PCA | Downloaded from web | 84.9% |
Lyle et al. [ |
Gender and ethnicity | LBP | SVM | FRGC | Gender 93%, ethnic 91% |
Survey on recognition through periocular biometric.
Year | Authors | Algorithm | Features | Testing database | Performance results | |
---|---|---|---|---|---|---|
2010 | Hollingsworth et al. [ |
Human analysis | Eye region | NIR images of 120 subjects | Accuracy of 92% | |
|
||||||
2010 |
Woodard et al. [ |
LBP fused with iris matching | Skin | MBGC NIR images from 88 subjects | Left eye rank-1 recognition rate: | Iris 13.8% |
Right eye rank-1 recognition rate: | Iris 10.1% | |||||
|
||||||
2010 |
Miller et al. [ |
LBP | Color information, skin texture | FRGC neutral expression, different session | Rank-1 recognition rate: | Periocular 94.10% |
FRGC alternate expression, same session | Rank-1 recognition rate: | Periocular 99.50% | ||||
FRGC alternate expression, a different session | Rank-1 recognition rate: | Periocular 94.90% | ||||
|
||||||
2010 |
Miller et al. [ |
LBP, city block distance | Skin | FRGC VS images from 410 subjects | Rank-1 recognition rate: | Left eye 84.39% |
FERET VS images from 54 subjects | Rank-1 recognition rate: | Left eye 72.22% | ||||
|
||||||
2010 |
Adams et al. [ |
LBP, GE to select features | Skin | FRGC VS images from 410 subjects | Rank-1 recognition rate: | Left eye 86.85% |
FERET VS images from 54 subjects | Rank-1 recognition rate: | Left eye 80.25% | ||||
|
||||||
2011 |
Woodard et al. [ |
LBP, color histograms | Skin | FRGC neutral expression, a different session | Rank-1 recognition rate: | Left eye 87.1% |
FRGC alternate expression, same session | Rank-1 recognition rate: | Left eye 96.8% | ||||
FRGC alternate expression, different session | Rank-1 recognition rate: | Left eye 87.1% |
The derivation of optimal dynamic periocular region gives a simultaneous solution to the aforementioned questions.
Unlike other biometric traits, periocular region has no boundary defined by any edge information. Hence periocular region cannot be detected through differential change in pixel value in different directions. Rather the location of boundary is the region which is smooth in terms of pixel intensity, that is, a region with no information. The authors of [
The objective of the paper is to attain a dynamic boundary around the eye that defines periocular region. The region hence derived should have the following properties: (a) should be able to recognize humans uniquely, (b) should be achievable for low-quality VS images, (c) should contain main identifiable features of eye region identifiable by a human being, and (d) no subset of the derived periocular region should be equally potent as the derived region for recognition.
The optimally selected periocular template can be a template to hold identity of an individual. If such template can be generated for the whole nation, it can serve as authorized identity (i.e., biometric passport [
To achieve the above stated properties, four different dynamic models are proposed through which periocular region can be segmented out. These models are based on (a) human anthropometry, (b) demand of the accuracy of biometric system, (c) human expert judgement, and (d) subdivision approach.
In a given face image, face can be extracted out by neural training to the system or by fast color-segmentation methods. The color-segmentation methods detect skin region in the image and find the connected components in such a region. Depending on connected components having skin color, the system labels the component largest in size as face. Algorithm
(1) Convert RGB image (2) Normalize (3) Compute the average luminance value of image (4) Brightness compensated image where, (5) The skin map where (6)
(1) Convert RGB image (2) Normalize (3) (4) (5) For each connected component (6) (7) For each pixel ( (8) if (Removal of the (9)
Once the eye region is detected, the iris center can be obtained using conventional pupil detection and integrodifferential approach for finding the iris boundary and a static boundary can be fitted. As described earlier, the authors of [
Anthropometric analysis [
This information can be used to decide the boundary of periocular region. In (
Further, from (
This method achieves periocular localization without knowledge of iris radius. Hence it is suitable for localization of periocular region for unconstrained images where iris radius is not detectable by machines due to low-quality, partial closure of eye, or luminance of the visible spectrum eye image.
However, to make the system work in more unconstrained environment, periocular boundary can be achieved through sclera detection, for the scenario when iris cannot be properly located due to unconstrained acquisition of eye or when the image captured is a low-quality color face image captured from a distance.
The input RGB iris image The input RGB iris image If
All binary connected components present in If size of the second largest connected component is less than 25% of that of the large one, it is interpreted that the largest component is the single sclera detected and the second largest connected component is removed hence. Else both components are retained as binary map of sclera.
After processing these above specified steps, the binary image would only contain one or two components describing the sclera region, after removing noises.
After a denoised binary map of sclera region within an eye image is obtained, it is necessary to retrieve the information about sclera, whether two parts of sclera on two sides of iris are separately visible, only one of them is detected, or both parts of sclera are detected as a single component.
There can be three exhaustive cases in the binary image found as sclera: (a) the two sides of the sclera is connected and found as a single connected component, (b) two sclera regions are found as two different connected components, and (c) only one side of the sclera is detected due to the pose of eye in the image. If the number of connected components is found to be two, then it is classified as aforementioned Case b (as shown in Figures
Result of nodal point detection through sclera segmentation.
Sample output 1 from UBIRISv2 database
Sample output 2 from UBIRISv2 database
Sample output 3 from UBIRISv2 database
Sample output 4 from UBIRISv2 database
Sample output 5 from UBIRISv2 database
Each sclera is subjected to following processing through which three nodal points are detected from each sclera region, namely (a) center of sclera, (b) center of concave region of sclera, and (c) eye corner. So in general cases where two parts of the sclera are detected, six nodal points will be detected. The method of nodal point extraction is illustrated below.
The result of extracting these nodal points from eye image helps in finding the tilt of eye along with the position of iris in eye. Figure
Beginning with the center of the eye (pupil center), a bounding rectangular box is taken of which only encloses the iris. Figure
Cropped images from an iris image centering at pupil center.
Method of formation of concave region of a binarized sclera component.
Different ratios of portions of face from human anthropometry.
The exact method of obtaining the dynamic boundary is as follows. For For each image in database, find approximate iris location in eye image. For each image in database, centering at the iris center, crop a bounding box whose width Find accuracy of the system with this image size. Observe the change in accuracy with
Figure
Change of accuracy of periocular recognition with change in size of periocular template tested on subset of UBIRISv2 and FERET datasets.
To validate this experiment, the same experiment has been carried out once again on full database of UBIRISv2 and FERET. The obtained accuracy values as depicted in Figure
Change of accuracy of periocular recognition with change in size of periocular template tested on full UBIRISv2 and FERET datasets.
Distribution of scores for imposter and genuine matching tested on full UBIRISv2 dataset applying LBP + SIFT on periocular template having width as 300% of the iris diameter.
Distribution of scores for imposter and genuine matching tested on full FERET dataset applying LBP + SIFT on periocular template having width as 300% of the iris diameter.
Change of 1 : 1 matching time with change in size of periocular template tested on full UBIRISv2 and FERET datasets.
Human expertise has been utilized to decide a sorted order of importance of different sections of periocular region towards recognition [
During enrolment phase of a biometric system, a human expert needs to verify manually whether the captured image includes expected region of interest. Through automated labeling different sections of an eye, it can be stated which portion of eye is necessary for identification (from human expert knowledge already discussed) and an automated FTA detection system can be made. Hence there is no need of a human expert for verifying the existence of important portions of human eye in an acquired eye image.
The challenge in incorporating this strategy in localization of periocular region is the automatic detection of portions of human eye like eyelid, eye corner, tear duct, lower-eyefold, and so forth. An attempt to do subdivision detection in eye region can be achieved through color detection and analysis and applying different transformations.
There are four methods explained through which an optimal periocular template can be selected for biometric recognition. The first two methods explained in Sections
Detail of publicly available testing databases.
Database | Developer | Version | Number of images | Number of subjects | Resolution | Color model |
---|---|---|---|---|---|---|
UBIRIS | Soft Computing and Image Analysis (SOCIA) Group, Department of Computer Science, University of Beira Interior, Portugal | v1 [ |
1,877 |
241 |
|
RGB |
|
||||||
FERET [ |
National Institute of Standards and Technology (NIST), Gaithersburg, Maryland | v4 | 14,126 | 1,191 |
|
RGB |
Anthropometry based approach performs accurately along with proper skin detection and sclera detection in eye region. The sample outputs are shown in Figure
Saturation accuracy based approach performs with an accuracy more than 80% with noisy and low-resolution images of UBIRISv2 and FERET, which marks the efficiency of the proposed approach. To analyse the performance more deeply, Receiver Operating Characteristic (ROC) curve is experimented out when the width of the periocular region is 200%, 250%, and 300% of the diameter of iris region, respectively. ROC curve depicts the dependence of false rejection rate (FRR) with false acceptance rate (FAR) for change in the value of threshold. The curve is plotted using linear, logarithmic, or semilogarithmic scales. As plotted in Figures
Change of
Width of periocular region ( |
100 | 150 | 200 | 250 | 300 | 350 | 400 |
Value of |
1.23 | 1.60 | 2.05 | 2.34 | 2.61 | 2.72 | 2.85 |
Value of |
1.19 | 1.55 | 2.01 | 2.29 | 2.53 | 2.66 | 2.69 |
Receiver Operating Characteristic (ROC) curve for different template sizes of periocular region for UBIRISv2.
Receiver Operating Characteristic (ROC) curve for different template sizes of periocular region for FERET.
Cumulative Match Characteristic (CMC) curve for different template sizes of periocular region for UBIRISv2.
Cumulative Match Characteristic (CMC) curve for different template sizes of periocular region for FERET.
Human expert judging is experimented by Hollingsworth et al. [
Subdivision approach needs manual supervision in the process of proper labeling of the different portions of human eye. Once the enrolled templates are labeled by the expert, an optimal part of the template can be selected for recognition. The method is tested on FERET database and yielded proper localization of periocular region.
Recent research signifies why recognition through visual spectrum periocular image has gained so much importance and how the present approaches work. While developing recognition system for a large database, it is a crucial factor to optimize the template size. Existence of any redundant region in template will increase the matching time but will not contribute to increase the accuracy of matching. Hence removal of redundant region of the template should be accomplished before the matching procedure. As recognition time of identification is dependent on database size
Near infrared
Visual spectrum
Local Binary Pattern
Scale Invariant Feature Transform
Receiver Operating Characteristic
Cumulative Match Characteristic
Failure to Acquire
False rejection rate
False acceptance rate.