Abstract
Palm veins have become a research hotspot of biometric recognition due to their own advantages of universality, uniqueness, collectability and stability. This paper proposes a palm vein recognition algorithm based on Neighborhood Preserving Embedding (NPE) and Kernel Extreme Learning Machine (KELM). The algorithm firstly performs gray-scale normalization processing on vein images, then extracts neighborhood preserving embedding dimensionality reduction features, and finally uses extreme learning machine for classification and recognition. The method is tested on the multispectral palmprint database of Hong Kong Polytechnic University. The experimental results show that this method can effectively reduce the vein dimensions to less than 30, and achieve an ideal recognition effect, if the parameters are selected appropriately. The algorithm is also verified on the palmvein database of Tongji University and FYO palmvein database for verifying the robustness, and also get ideal experimental results.
Highlights
Biometric identification technology is a technology that uses human physiological characteristics or behavioral characteristics for identity authentication
Since the subspace dimensionality reduction algorithm can both well maintain the original structural characteristics of the sample and inherit the classification ability of the neural network, this paper proposes a palm vein recognition algorithm based on Neighborhood Preserving Embedding (NPE) and Kernel Extreme Learning Machine (KELM), and obtains relatively ideal experimental results
SELECTION OF PARAMETERS 1) THE INFLUENCE OF NPE PARAMETERS In order to verify the influence of NPE neighborhood size and dimensionality reduction on the recognition rate, this experiment randomly selects 200 samples, and each sample randomly selects 8 images as training samples, with the remaining 4 images as samples to be tested
Summary
Biometric identification technology is a technology that uses human physiological characteristics or behavioral characteristics for identity authentication. The collection process is very friendly because the use of palms is relatively natural and easy to accept by users [1]. These merits have made vein recognition technology one of the research hotspots of domestic and foreign researchers in recent years. The depth learning-based method [15]–[20] mainly uses existing deep learning networks for vein recognition. Since the subspace dimensionality reduction algorithm can both well maintain the original structural characteristics of the sample and inherit the classification ability of the neural network, this paper proposes a palm vein recognition algorithm based on NPE and KELM, and obtains relatively ideal experimental results.
Sign-in/Register to view highlights & summary 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 callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
