Abstract
Palmprint recognition has received tremendous research interests due to its outstanding user-friendliness such as non-invasive and good hygiene properties. Most recent palmprint recognition studies such as deep-learning methods usually learn discriminative features from palmprint images, which usually require a large number of labeled samples to achieve a reasonable good recognition performance. However, palmprint images are usually limited because it is relative difficult to collect enough palmprint samples, making most existing deep-learning-based methods ineffective. In this paper, we propose a heuristic palmprint recognition method by extracting triple types of palmprint features without requiring any training samples. We first extract the most important inherent features of a palmprint, including the texture, gradient and direction features, and encode them into triple-type feature codes. Then, we use the block-wise histograms of the triple-type feature codes to form the triple feature descriptors for palmprint representation. Finally, we employ a weighted matching-score level fusion to calculate the similarity between two compared palmprint images of triple-type feature descriptors for palmprint recognition. Extensive experimental results on the three widely used palmprint databases clearly show the promising effectiveness of the proposed method.
Highlights
As one of the most important solutions for performing personal authentication our modern society, biometric recognition can effectively and efficiently identify an individual based on one’s physiological or behavioral traits [1,2,3]
There have been many palmprint recognition methods proposed in the past decades, which can be roughly classified into three categories according the types of palmprint images [17]: highresolution palmprint [18], low-resolution palmprint [7] and three-dimensional (3D) palmprint recognition [19] methods
In this paper, we propose a triple-type feature descriptor (TFD) for palmprint representation and recognition
Summary
As one of the most important solutions for performing personal authentication our modern society, biometric recognition can effectively and efficiently identify an individual based on one’s physiological or behavioral traits [1,2,3]. Genovese et al [36] proposed a PalmNet method by applying the Gabor responses and PCA into the convolutional networks These learning-based methods usually require many labeled samples to learn and extract the discriminative features. Unlike single-type feature descriptor, our proposed method can completely represent the multiple important and inherent characteristics of palmprint images. Unlike the recently learning-based methods which require many training samples, our proposed method can effectively extract the discriminative feature manually without requiring any training samples, such that our proposed method is suitable for the few-shot and even zero-shot biometric recognition tasks We conduct both palmprint verification and palmprint identification experiments on three widely used challenging databases and the experimental results demonstrate that our proposed method consistently outperforms previous state-of-the-art methods.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
