Face deep counterfeiting identification method based on deep learning and related equipment

The invention discloses a deep learning-based face deep counterfeiting identification method and related equipment. The scheme comprises the following steps of: 1, training, learning and reconstructing distribution characteristics of a face image and a real face image through an Encoder-Decoder deep network; 2, intercepting a real face image and a forged face image of a target in the image; 3, aligning the face of the real face image and the face of the forged face image; step 4, performing reconstruction processing on the real face image and the forged face image by using an Encoder-Decoder deep network, and extracting distribution features of the reconstructed real face image and the reconstructed forged face image; and 5, calculating the cosine similarity of the distribution characteristics of the real face reconstruction image and the forged face reconstruction image, and classifying the real face image and the forged face image according to the cosine similarity to obtain an identification result. Accord