Face Recognition System using Histograms of Oriented Gradients and Convolutional Neural Network based on with Particle Swarm Optimization

In this paper, Histograms of Oriented Gradients dependent on the strong point of convolutional neural organization which is new methodology for evenness face data set, is introduced. A proposed face acknowledgment framework was created to be utilized for various purposes. We utilized Gabor wavelet change for include extraction of evenness face preparing information and afterward we utilized profound learning technique for acknowledgment. We executed and assessed the proposed strategy on ORL and YALE data sets with Matlab 2020b. Besides, similar trials were directed applying Particle Swarm Optimization (PSO) for include determination approach. The execution of Gabor wavelet include extraction with a high number of preparing picture tests has end up being more viable than different strategies in our examination. The acknowledgment rate while carrying out the PSO strategies on ORL data set is 86.62% while it is 92.6% with the three techniques on YALE data set. In any case, the utilization of PSO calculation has expanded the exactness rate to 95.88% for ORL information base and 95.23% on YALE data set.