Author(s): WU Chunsheng, FENG Caigang, CHI Xuebin, Beijing Criminal Science Institution, Computer Network Information Center, Chinese Academy of Sciences, University of Chinese Academy of Sciences
Pages: 571-576
Year: 2015 Issue:  4
Journal: Journal of the Graduate School of the Chinese Academy of Sciences
Keyword:  palmprint recognition; minutia feature point; GPU;
Abstract: The problems of palmprint application in public security department,such as large database and poor quality of scene palmprint,are described in this paper. Firstly,the computational complexity of palmprint feature point matching algorithm is analyzed. The three-level feature point matching algorithm is presented based on the results of the analysis. Secondly,the working process of three-level feature point matching algorithm is described in detail. Finally the algorithm is transformed to the GPU parallel and the test results are given. The GPU technology is first applied to palmprint recognition in our study. The results show that the speed of palmprint feature matching increases about three times compared with others,and the speed may increase more than 15 times after using GPU technology.