Visible Spectrum, Bi-Modal Ocular Biometrics

Tankasala, Sriram Pavan; Doynov, Plamen; Derakhshani, Reza

doi:10.1016/j.protcy.2012.10.068

Cited by 5 publications

(5 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Given the reported selective test procedures summarised under ‘comments’ in Table 2, we note that it may not be straightforward to compare our results with most of the previous studies. More specifically, the EER's reported in [21, 22, 24, 27] were generated after eliminating unspecified poor quality images. In [26, 28], three samples from each user were used to derive a codebook, and remaining two samples from each user were used to generate match results on that user.…”

Section: Resultsmentioning

confidence: 99%

“…Since its introduction by Derakhshani and Ross, other researches have been introducing various segmentation, image enhancement, feature extraction, and classification algorithms for conjunctival vasculature biometrics. So far, eyeprint biometric studies have been performed either by utilising UBIRIS v1 RGB dataset, or researchers' in-house ocular images [12,13,[16][17][18][19][20][21][22][23][24][25][26][27][28][29]. Some studies have also reported using multispectral captures [30,31].…”

Section: Iet Biometricsmentioning

confidence: 99%

“…The performance of gray-level cooccurence matrix (GLCM) [18] and pyramidal directional filter [19] features were investigated using dSLR captures. In related work [21][22][23][24][25][26][27][28][29], several methods were evaluated on UBIRISv1 database [20] compared with in-house datasets. In [30,31], authors examine eyeprint recognition captured by a multispectral camera.…”

Section: Related Workmentioning

confidence: 99%

See 2 more Smart Citations

Method for using visible ocular vasculature for mobile biometrics

Gottemukkula¹,

Saripalle

Tankasala

et al. 2016

IET biom.

Self Cite

View full text Add to dashboard Cite

Securing personal information on handheld devices, especially smartphones, has gained a significant interest in recent years. Yet, most of the popular biometric modalities require additional hardware. To overcome this difficulty, the authors propose utilising the existing visible light cameras in mobile devices. Leveraging visible vascular patterns on whites of the eye, they develop a method for biometric authentication suitable for smartphones. They start their process by imaging and segmenting whites of the eyes, followed by image quality assessment. The authors' stage 1 matcher is a three-step process that entails extracting interest points [Harris-Stephens, features from accelerated segment test, and speeded up robust features (SURF)], building features (SURF and fast retina keypoint) around those points, and match score generation using random sample consensus-based registration. Stage 2 matcher uses registered Gabor phase filtered images to generate orientation of local binary pattern features for its correlation-based match metric. A fusion of stage 1 and stage 2 match scores is calculated for the final decision. Using a dataset of 226 users, the authors' results show equal error rates as low as 0.04% for long-term verification tests. The success of their framework is further validated on UBIRIS v1 database.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Iet Biometricsmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Method for using visible ocular vasculature for mobile biometrics

Gottemukkula¹,

Saripalle

Tankasala

et al. 2016

IET biom.

Self Cite

View full text Add to dashboard Cite

show abstract

“…Thomas et al [39] line descriptors 1.03 800 best quality images were considered for matching Zhou et al [24] line descriptors 3.70 used 1168 images with manual sclera segmentation Zhou et al [24] line descriptors 1.34 800 best quality images were considered for matching Oh et al [23] LBP 0.47 additional user-specific tokens were used for decisions Tankasala et al [40] GLCM 10.22 only 40 high-quality images were used Das et al [41] dense SIFT 0.66 a codebook was derived using three samples from each subject Lin et al [42] Y shape descriptors 3.05 46 images with artefacts were excluded Das et al [43] dense LBP's 0.80 a codebook was derived using three samples from each subject Oh et al [25] LBP's 8.49 all images were used Gottemukkula [12] SURF, fast retina key points, LBP's 2.39 all images from session 1 were used (1205 × 1205)…”

Section: Comparisons Using Cibit-ii Datasetmentioning

confidence: 99%

“…This public dataset has been a popular choice for OSV biometric studies. To facilitate comparisons with other studies [12,[23][24][25][39][40][41][42][43], using UBIRIS V1, we performed 1205 pair-wise comparisons from its session 1 and 670 comparisons from its session 2. However, given that most of the reported studies have chosen to work on their preferred subset of UBIRIS images, it is difficult to directly compare our work with previous research.…”

Section: Comparative Evaluation Ubiris V1 Datasetmentioning

confidence: 99%

Ocular surface vasculature recognition using curvelet transform

et al. 2016

Self Cite

View full text Add to dashboard Cite

The vascular patterns seen on the white of the eye, mainly in conjunctival and episcleral layers, are termed as ocular surface vasculature (OSV). OSV is visible in images captured with commercial RGB cameras, and its unique texture can be used for biometric recognition. This study demonstrates the capabilities of curvelet transform for OSV feature extraction. Nonlinear feature enhancement and feature mapping in curvelet domain are shown to be effective in differentiating OSV texture. Linear discriminant analysis and similarity metrics are used for matching. A match-score level fusion is used across multiple gaze directions for both eyes. Using a multi-distance dataset of 50 volunteers, where eye images were acquired from 30, 150, and 250 cm using a dSLR, a best equal error rate (EER) of 0.2% is obtained. Using a second dataset of 40 volunteers acquired from 150 cm using a dSLR, a best EER of 3.1% is obtained. For a 216-participant dataset of ocular images acquired using cellular phones from close proximity, an EER of 0.9% is obtained. The proposed methodology was also tested on the publicly available UBIRIS V1 dataset, yielding an EER of 0.7%. The experimental results support the theoretically formulated advantages of the curvelet transform and its capability in successful extraction of curved structures when applied to OSV patterns.

show abstract

Multispectral Ocular Biometrics

Crihalmeanu

Ross

2016

Face Recognition Across the Imaging Spectrum

View full text Add to dashboard Cite

Visible Spectrum, Bi-Modal Ocular Biometrics

Cited by 5 publications

References 28 publications

Method for using visible ocular vasculature for mobile biometrics

Method for using visible ocular vasculature for mobile biometrics

Ocular surface vasculature recognition using curvelet transform

Multispectral Ocular Biometrics

Contact Info

Product

Resources

About