The Face recognition based on the sparse autoencoder

The sparse autoencoder was used for the construction of the face recognition. Firstly, the face image database we used is the MIT face image database, In order to achieve the better results, image preprocessing is essential. It included the shear of the effective face area and Local Binary Pattern of the effective face area. Secondly, through training the hidden layer, initializing the network and training the forward network, a machine for Face recognition can be built. The experimental results showed that the sparse autoencoder improved the accuracy of face recognition and achieved good results in practical operation.