deep neural network multiple patch combination METHOD FOR RECOGNIZING FACE USING MULTIPLE PATCH COMBINATION BASED ON DEEP NEURAL NETWORK WITH FAULT TOLERANCE AND FLUCTUATION ROBUSTNESS IN EXTREME SITUATION

摘要

In the present invention, in a face recognition method using a multiple patch combination based on a deep neural network, (a) when a face image having a first size is obtained, the face recognition apparatus A feature extraction network using an image-The feature extraction network is characterized in that at least one feature is extracted using a learning face image having a second size, and the second size is smaller than the first size. In this way, the feature extraction network generates a feature map by applying at least one convolution operation to the face image having the first size, and applying a sliding pooling operation to the feature map to generate a plurality of features. Step to do; And (b) the face recognition apparatus inputs the plurality of features to a learned neural aggregation network, and causes the neural aggregation network to aggregate the plurality of features to determine at least one optimal feature for face recognition. It relates to a method comprising a; to output.