Multiscale Feature Extraction of Finger-Vein Patterns Based on Wavelet and Local Interconnection Structure Neural Network

Abstract

We propose a multiscale feature extraction method of finger-vein patterns based on wavelet and local interconnection structure neural networks. The finger-vein image is performed the multiscale self-adaptive enhancement transform. A neural network with local interconnection structure is designed to extract the features of the finger-vein pattern. This method has three features: Firstly, by applying the multiscale self-adaptive enhancement transform to the finger-vein image, the finger-vein pattern is emphasized and noises are refrained. Secondly, we use different receptive fields to deal with different size finger-rein patterns. This and the multiscale property of the wavelet analysis lead to accurate extraction of different size finger-rein modes. Thirdly, our method is very fast by using the integral image method. The experimental results show the proposed method is superior to other methods and solve the problem of extracting features from the unclear images efficiently. The EER of the proposed method is 0.130% in personal identification