Face recognition method for unlocking smartphones based on fusion of global and local features

In order to solve the problems of low accuracy and poor response speed in smartphone face recognition unlocking, a smartphone unlocking face recognition method based on the fusion of global and local features is proposed. Firstly, perform bilinear interpolation and greyscale processing on the facial image. Secondly, the LBP algorithm and ASM model are used to extract local and global features of facial images. Again, integrate local and global features. Finally, based on SVM, the optimal hyperplane for classifying facial image sample data is constructed, and a kernel function is introduced to unlock facial recognition for smartphones. Through experiments, it has been shown that the unlocking recognition error of the proposed method is always below 0.2%, the equilibrium point in the P-R curve is at 0.83, and the average response time of unlocking recognition remains below 0.5s. The experimental data is superior to the comparative method, and the accuracy of face recognition is high, the response speed is fast, and it has good application effects.