The processing of palmprint image based on CUDA

This paper processed the palmprint feature image with neural network by utilizing the parallel computing technology based on CUDA, at the same time, this paper also processed the same neural network based on Matlab as contrast. By comparing the results of processing speed, we came to the conclusion that parallel computing technology based on CUDA has a huge advantage on processing speed. The algorithm was deduced and discussed, and it was also verified by the experiment. The result showed that parallel computing technology can improve the speed of processing greatly and the performance of palmprint recognition processing.