A Novel Image Enhancement Method for Palm Vein Images

Abstract

Palm vein images usually suffer from low contrast due to skin surface scattering the radiance of NIR light and image sensor limitations, hence require employing various techniques to enhance the contrast of the image prior to feature extraction. This paper presents a novel image enhancement method referred to as Multiple Overlapping Tiles (MOT) which adaptively stretches the local contrast of palm vein images using multiple layers of overlapping image tiles. The experiments conducted on the CASIA palm vein image dataset demonstrate that the MOT method retains the finer subspace details of vein images which allows excellent feature detection and matching with SIFT and RootSIFT features. Results on existing palm vein recognition systems demonstrate that the proposed MOT method delivers lower EER values outperforming other existing palm vein image enhancement methods.

A Keypoint Filtering Method for SIFT-based Palm-Vein Recognition

“…The MOT method from [3] was used as the image enhancement method with 16×16 pixel image tiles. The performance of the MOT method was assessed by implementing several existing SIFT-based palm vein recognition systems.…”

“…Further [8] used KNN + RT + bidirectional-matching with RootSIFT features. In [3] the image enhancement methods of the previous work were replaced with the MOT method to verify its performance. EER values were reported using multiple registration images (1-5) with ED, ED+RANSAC, KN+RT, and KNN+RT+RANSAC.…”

“…EER values were reported using multiple registration images (1-5) with ED, ED+RANSAC, KN+RT, and KNN+RT+RANSAC. To verify the performance of the MMD filter from this research, the benchmarking is based upon the results presented in [3] for SIFT descriptor matching and replaced the false positive filtering method with MMD. Thresholds of 25 and 30 pixels were used for Tμ and TD respectively.…”

“…In [3], a contrast enhancement method has been proposed to address the issues associated with current image enhancement techniques used with palm vein recognition. This method is referred to as Multiple Overlapping Tiles (MOT).…”

“…This filter is referred to as the Mean and Median Distance (MMD) filter. The experiment results has been compared with existing palm vein recognition systems based on SIFT [3], [7]- [10]. In section III the pseudocode of the proposed MMD filter is presented.…”