Image Data Enhancement Method Based on Generative Adversarial Network for Millimeter-wave Body Scanner
Yijie Niu, Pengzhe Li, Haitao Qin, Qi Ye, Yuejun Han · 2021
A large number of serious artifact image data sets are necessary for the training of image recognition algorithm of millimeter wave body scanner. It is very difficult to obtain serious artifact images in practical work. The generative adversarial network is introduced for data enhancement by building an appropriate loss function. This method can create conditions for training multi-class classification networks and improve the effectiveness of millimeter wave body image evaluation.