Compact multi-scale periocular recognition using SAFE features

Alonso‐Fernandez, Fernando; Mikaelyan, Anna; Bigün, Josef

doi:10.1109/icpr.2016.7899842

Cited by 2 publications

(3 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Their experiments demonstrated that fusing multiple image features improves recognition accuracy. Alonso-Fernandez et al [9] proposed integrating LBP, HOG, SIFT, and SURF at the score level by weighted sum and reported that this method is robust to the distance between the camera and the subject. Park et al [10] proposed integrating gradient orientation, LBP, and SIFT at the score level by weighted sum.…”

Section: Periocular Recognitionmentioning

confidence: 99%

See 1 more Smart Citation

Personal Authentication for Periocular Region in Thermal and Visible Light Images by Using CNN

Nakamura,

Kawamura,

Yoshiura

et al. 2024

Journal of Information Processing

View full text Add to dashboard Cite

Personal authentication based on the periocular region can be performed even when the person's mouth and nose are hidden by a face mask. However, using visible light images for periocular recognition is problematic because recognition accuracy is affected by changes in the lighting conditions. We have developed a method for periocular recognition that overcomes this problem by using thermal images, which are less affected by changes in lighting conditions, in addition to visible light images. In this paper, we propose a method using both thermal and visible light images for periocular recognition based on features obtained by CNN. In addition, our method uses deep metric learning to deal with persons who are not included in the training data. To evaluate the accuracy of the proposed method under unstable conditions, we conducted recognition experiments using images of 83 subjects obtained from the USTC-NVIE database, which contains visible light and thermal images taken simultaneously under various lighting conditions and with various facial expressions. The experimental results show that using both visible light and thermal images achieves higher recognition accuracy than using only visible light images.

show abstract

Section: Periocular Recognitionmentioning

confidence: 99%

“…Many methods using handcrafted features have been proposed for periocular authentication [3], [7], [8], [9], [10]. However, deep learning, which has recently been shown to be effective, has been demonstrated to greatly improve accuracy [11].…”

Section: Periocular Recognitionmentioning

confidence: 99%

Personal Authentication for Periocular Region in Thermal and Visible Light Images by Using CNN

Nakamura,

Kawamura,

Yoshiura

et al. 2024

Journal of Information Processing

View full text Add to dashboard Cite

show abstract

“…Symmetry Patterns based on the Symmetry Assessment by Feature Expansion (SAFE) descriptor [7], which encodes the presence of various symmetric curve families in concentric annular rings around image key-points. We use the sclera center as unique key-point.…”

Section: Periocular Recognition Systemsmentioning

confidence: 99%

Log-likelihood score level fusion for improved cross-sensor smartphone periocular recognition

Alonso‐Fernandez

Raja

Busch

et al. 2017

2017 25th European Signal Processing Conference (EUSIPCO)

Self Cite

View full text Add to dashboard Cite

Abstract-The proliferation of cameras and personal devices results in a wide variability of imaging conditions, producing large intra-class variations and a significant performance drop when images from heterogeneous environments are compared. However, many applications require to deal with data from different sources regularly, thus needing to overcome these interoperability problems. Here, we employ fusion of several comparators to improve periocular performance when images from different smartphones are compared. We use a probabilistic fusion framework based on linear logistic regression, in which fused scores tend to be log-likelihood ratios, obtaining a reduction in cross-sensor EER of up to 40% due to the fusion. Our framework also provides an elegant and simple solution to handle signals from different devices, since same-sensor and crosssensor score distributions are aligned and mapped to a common probabilistic domain. This allows the use of Bayes thresholds for optimal decision making, eliminating the need of sensor-specific thresholds, which is essential in operational conditions because the threshold setting critically determines the accuracy of the authentication process in many applications.

show abstract

Compact multi-scale periocular recognition using SAFE features

Cited by 2 publications

References 15 publications

Personal Authentication for Periocular Region in Thermal and Visible Light Images by Using CNN

Personal Authentication for Periocular Region in Thermal and Visible Light Images by Using CNN

Log-likelihood score level fusion for improved cross-sensor smartphone periocular recognition

Contact Info

Product

Resources

About