GAN generated picture detection method and system based on residual domain rich model

Abstract

The invention provides a GAN generated picture detection method and a system based on a residual domain rich model, and the method comprises the steps: an original image processing step: carrying outthe recognition and cutting of an original image through employing a digital image processing technology, recognizing a face, and cutting an image of a face part; a step of obtaining residual information of the original image: preprocessing the original image by using a digital image processing technology, and extracting the residual information of the original image; a convolutional neural network processing step: inputting a residual image of the original image into a set convolutional neural network, and adding a BN layer to a convolutional layer in front of each activation function; a global average pooling layer processing step: replacing a full connection layer with a global average pooling layer; and a sample training convolutional neural network processing step: training a convolutional neural network by using the sample of the data set to obtain a picture classifier, and obtaining judgment result information. According to the method, a preprocessing high-pass filter is designed, and 99% of accuracy is finally achieved through an improved convolutional neural network.

Description

technical field [0001] The present invention relates to digital image processing technology and the interdisciplinary field of artificial intelligence, in particular to a method and system for detecting pictures generated by GAN based on residual domain rich model. Background technique [0002] In 2017, NVIDIA launched ProGAN, which broke through the limit of the previous generative confrontational neural network (GAN for short), and generated high-resolution synthetic face pictures (1024×1024). Most of the pictures generated by the model are already highly deceptive, but the details are still not ideal. People can still distinguish the authenticity of the pictures through careful observation with the naked eye. On this basis, NVIDIA has launched a new generative confrontational neural network StyleGAN, whose synthetic face can be further improved to be real, and even the details have been improved. Philip Wang, a software engineer from Uber, used StyleGAN to make a website...

AI Technical Summary

Problems solved by technology

This method is only for the detection of low-resolution faces in the video, and does not recognize and detect high-resolution face images

Images