Iris super‐resolution using CNNs: is photo‐realism important to iris recognition?

Ribeiro, Eraldo; Uhl, Andreas; Alonso‐Fernandez, Fernando

doi:10.1049/iet-bmt.2018.5146

Cited by 29 publications

(23 citation statements)

References 46 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Among the methods making use of very small iris images, our works with PCA [11], [12], [42] are among the most competitive. Recent studies adapting deep-learning frameworks [44], [45] still report an accuracy significantly worse in some cases. There is one reconstruction-based method using PCT enhancement [30] which also stands out for its excellent performance with an iris diameter of only 20 pixels.…”

Section: A Taxonomy Of Iris Super-resolution Algorithmsmentioning

confidence: 99%

“…The reported accuracy corresponds to the best accuracy obtained with the smallest iris size (shown in column 7). In all cases, near-infrared (NIR) data is used, except in the works [25], [34], [42], [45]. All other terms are explained in the text or in referenced papers.…”

Section: A Contributionsmentioning

confidence: 99%

“…The average iris diameter of sub-sampled images in these studies ranged from 11 to 53 pixels. Finally, in our studies with the VSSIRIS database [42], [45], we down-sampled iris images to an average diameter of 13 pixels.…”

Section: A Taxonomy Of Iris Super-resolution Algorithmsmentioning

confidence: 99%

See 2 more Smart Citations

A Survey of Super-Resolution in Iris Biometrics With Evaluation of Dictionary-Learning

et al. 2019

View full text Add to dashboard Cite

The lack of resolution has a negative impact on the performance of image-based biometrics. While many generic super-resolution methods have been proposed to restore low-resolution images, they usually aim to enhance their visual appearance. However, an overall visual enhancement of biometric images does not necessarily correlate with a better recognition performance. Reconstruction approaches thus need to incorporate the specific information from the target biometric modality to effectively improve recognition performance. This paper presents a comprehensive survey of iris super-resolution approaches proposed in the literature. We have also adapted an eigen-patches' reconstruction method based on the principal component analysis eigen-transformation of local image patches. The structure of the iris is exploited by building a patch-position-dependent dictionary. In addition, image patches are restored separately, having their own reconstruction weights. This allows the solution to be locally optimized, helping to preserve local information. To evaluate the algorithm, we degraded the high-resolution images from the CASIA Interval V3 database. Different restorations were considered, with 15 × 15 pixels being the smallest resolution evaluated. To the best of our knowledge, this is the smallest resolutions employed in the literature. The experimental framework is complemented with six publicly available iris comparators that were used to carry out biometric verification and identification experiments. The experimental results show that the proposed method significantly outperforms both the bilinear and bicubic interpolations at a very low resolution. The performance of a number of comparators attains an impressive equal error rate as low as 5% and a Top-1 accuracy of 77%-84% when considering the iris images of only 15 × 15 pixels. These results clearly demonstrate the benefit of using trained super-resolution techniques to improve the quality of iris images prior to matching.

show abstract

Section: A Taxonomy Of Iris Super-resolution Algorithmsmentioning

confidence: 99%

Section: A Contributionsmentioning

confidence: 99%

See 1 more Smart Citation

A Survey of Super-Resolution in Iris Biometrics With Evaluation of Dictionary-Learning

et al. 2019

View full text Add to dashboard Cite

show abstract

“…Recently, several methods like [17][18][19][20][21][22][23][24][25][26][27] have exploited convolution neural networks (CNNs) and deep learning methods to enhance the performance of the iris recognition systems via precise segmentation of the iris image or dense extraction of useful features. However, without optimal selection of the features that maximize the discrimination the high computational complexity and memory requirement of such methods lead to considerable challenges in their implementation.…”

Section: Related Workmentioning

confidence: 99%

“…Recently deep learning methods and convolutional neural networks are considered as powerful tools for segmentation and feature extraction of the biometric images [4,17,18,20,26]. They also accelerate the extraction of many diverse features from segmented biometric images in a small time [1,21].…”

Section: Conclusion and Suggestionsmentioning

confidence: 99%

Improving identification performance in iris recognition systems through combined feature extraction based on binary genetics

Abbasi

2019

SN Appl. Sci.

View full text Add to dashboard Cite

The process of iris recognition is composed of several stages. The most important stage is feature extraction. The majority of systems use a single method of feature extraction. We used a binary genetic algorithm with a new fitness criterion to find an efficient combination of feature extraction methods with the aim of improving the performance of iris recognition systems. The proposed method makes use of a large number of filters and transforms which are widely used to extract iris features. The best combination of these techniques is found over iterations of the proposed algorithm. The final combination, which is argued to be the most efficient method of feature extraction, consists of a number of wavelet transforms, Gabor filter, and Fourier transform. ROC graphs are used to show the experimental superiority of the performance of this method over methods that use a single filter. According to our findings, the proposed method shows better performance in most situations. An FAR value of 0 and an FRR value of 0.092 was obtained by this method.

show abstract

Design and implementation of pixel‐based adjustable ESD protection circuits for capacitive fingerprint biometric sensors

Lin

Hsieh

Chang

2022

Circuit Theory & Apps

View full text Add to dashboard Cite

The trade-off design between ESD endurability and polymer thickness on sensor plate for capacitive fingerprint is emerging in these years since the image quality of fingerprint is vital for biometrics authentication. In this paper, we propose a pixel-based ESD protection circuitry in parallel to GDPMOS and GGNMOS and an adjustable body terminal in proposed pixel-based ESD circuit to meet and exceed air discharge ±15 kV without thicker polymer. With the proposed pixel-based adjustable ESD protection circuit, the maximum air discharge achieved by (18 kV, À17 kV), and 10 out of 10 samples passed +15 kV, and 10 out of 10 samples passed À15 kV. Compared to conventional ground wall fingerprint sensor, the maximum air discharge was stopped by (15 kV, À15 kV), and just merely five out 10 samples passed 15 kV, seven out of 10 samples passed À15 kV. Besides the excellent ESD endurability, the proposed pixel-based ESD protection circuit achieves air-discharge 15.5 kV and grayscale count 230 at merely 50 μm thickness of polymer. Compared to the conventionally used ground wall sensor types, it passes only 13 kV and a grayscale count of 210 at a polymer thickness of 50 μm, but fails 15 KV Level 4 of IEC-61000-4-2. Even with polymer thickness increased to 250 μm, ground wall sensor types barely passed 15 kV, and the grayscale count dropped to 120, which is far from PIV image standard.

show abstract

Iris super‐resolution using CNNs: is photo‐realism important to iris recognition?

Cited by 29 publications

References 46 publications

A Survey of Super-Resolution in Iris Biometrics With Evaluation of Dictionary-Learning

A Survey of Super-Resolution in Iris Biometrics With Evaluation of Dictionary-Learning

Improving identification performance in iris recognition systems through combined feature extraction based on binary genetics

Design and implementation of pixel‐based adjustable ESD protection circuits for capacitive fingerprint biometric sensors

Contact Info

Product

Resources

About