The novel coronavirus infection (COVID-19), which appeared at the end of 2019 has developed into a global pandemic with numerous deaths, and has also become a serious social concern. The most important and basic measure for preventing infection is hand hygiene. In this study, by photographing palm images of nursing students after hand-washing, using fluorescent lotion to conduct hand-washing training and as a black light, we developed a hand hygiene evaluation system using pix2pix, which is a type of the generative adversarial network (GAN). In pix2pix, the input image adopted was a black light image obtained after hand-washing, and the ground truth image was a binarized image obtained by extracting the residue left on the input image by a trained staff member. We adopted 443 paired-images after hand-washing as training models, and employed 20 images as verification images, which included 10 input images with 65% or more of the residue left, and 10 input images with 35% or less of the residue left in the ground truth images. To evaluate the training models, we calculated the percentage of residue left in the estimated images generated from the verification images, and conducted two-class discriminant analysis using the Mahalanobis distance. Consequently, misjudgment only occurred in one image for each image group, and the proposed system with pix2pix exhibited high discrimination accuracy.