Application of BP Neural Network in the Fingerprint Identification
Honglong Chang, Xiaoli Zhao, Bin Li, Xu Zhou · 2012
Fingerprint identification is an individual recognition technology, which is based on the uniqueness and invariance of persons' fingerprint characteristics. In recent years, this means of identification has made substantial progress and development, which is one of the most mature technology in the development of biometric. Fingerprint identification has been widely used in social life and production fields. However, due to a variety of noise, there is a great deal of difficulty in the process of fingerprint identification and confirmation. This paper uses BP neural network for fingerprint feature extraction and recognition, discusses the network to set the initial weights of the principle, and puts forward the selection of principle of hidden layer number of neurons, reduces training time and calculated amount to improve the identification of accuracy. Keyword: Pattern recognition; Neural network; Fingerprint identification; Back Propagation