DETECTION ET RECONNAISSANCE DE VISAGE DANS UNE IMAGE PAR LE DEEP LEARNING.

Abstract

Deep learning methods, in particular convolutional neural networks(CNN) , have achieved great success in the field of computer vision, and this success shows great superiority over face detection and recognition algorithms and gives it detection speed and accuracy, especially in real time. The aim of this work is to study and apply this type of networks for the detection and recognition of faces subjected to in-depth learning. we used this type of networks for the classification and recognition of faces in images, so we proposed two models with different architectures (the number of convolution layers, pooling layers, fully connected layers and the number of era). The results obtained showed that the choice of the number of epochs and the size of the image base as well as the depth of the grating have a great influence to have the best results.