Face recognition algorithm based on particle swarm optimization and image feature compensation

In order to improve face recognition accuracy, an IFC face recognition algorithm based on image feature compensation strategy for face recognition is proposed in this paper. The strategy adopts the method of extracting the image feature description by using the computational factors to compensate the features of the original image. Since the size of the feature compensation coefficients directly affects the recognition rate of the face recognition system, a modified PSO algorithm is proposed for solving the optimal combination of feature compensation coefficients for multiple computation factors, so that the recognition rate of the system can be improved. Firstly, this paper introduces the relative concepts of feature compensation strategy and calculation factor. Secondly, the process and implementation of face recognition algorithm using feature compensation strategy are described. Thirdly, the improved PSO algorithm for solving the optimal solution of the feature compensation coefficient is designed, during which the evaluation method is constructed to evaluate the merits of the feature compensation coefficient, and the implementation process of the improved PSO algorithm is conceived. Finally, the algorithm proposed in this paper is verified by simulation experiment platform. It is demonstrated that the modified PSO algorithm can significantly enhance the recognition rate of the system following its application to the feature compensation strategy face recognition system. On the ORL dataset, the recognition rate can be enhanced from 93.05% to 99.00% when the training image reaches 5, meanwhile, it still exhibits good performance on other datasets.