Iris Recognition by Learning Fragile Bits on Multi-patches using Monogenic Riesz Signals

Shekar, B. H.; Bhat, Sharada S.; Mestetsky, Leonid

doi:10.1007/978-3-030-34872-4_51

Cited by 2 publications

(2 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…During experimentation, we have adopted iris preprocessing steps and recognition framework explained in [21]. have used the subsets of MICHE I that are comprised of eye images captured from three different mobile phones, iPhone5 (IP5), Samsung Galaxy S4 (GS4), and Samsung Galaxy Tab2 (GT2).…”

Section: Iris Recognitionmentioning

confidence: 99%

“…First set of experiments are conducted with respect to NIR and VW iris datasets, IITDelhi, MMU v2, CASIA v-4 Distance dataset and UBIRIS v-2. We have made a comparative study with other descriptors such as, LBP and HOG, and also with Riesz signal based binary pattern (RSBP) descriptor proposed in[21]. The experimental results containing the equal error rates (EER) and d − prime value are shown in the Table1.…”

mentioning

confidence: 99%

See 1 more Smart Citation

Complex Gradient Function Based Descriptor for Iris Biometrics and Action Recognition

Shekar

Shetty

Bhat³

2021

Communications in Computer and Information Science

Self Cite

View full text Add to dashboard Cite

Image gradient has always been a robust characteristic of digital image which possesses localised spatial information of each pixel in all the directions. Exploiting the gradient information at the pixel level is a good old technique applied in various fields of digital image processing. In this paper, the magnitude and direction of image gradient is explored to design a local descriptor. We propose a novel local feature descriptor based on Complex Gradient Function (CGF), which maps each pixel from the spatial plane into its complex extension involving the magnitude and direction of image gradient at that pixel. We exploit the proposed descriptor for human action recognition from depth sequences and human authentication using iris biometrics. The efficiency of descriptor is demonstrated with experimental results on benchmark datasets IITDelhi, MMU-v2, CASIA-Iris, UBIRIS, and MICHE-I for iris authentication and MSR Action 3D dataset for human action recognition (HAR).

show abstract

Section: Iris Recognitionmentioning

confidence: 99%

mentioning

confidence: 99%