The Application of Computer Deep Learning in Relics Landscape Restoration

Abstract

With the rapid disappearance and deterioration of monument landscapes, it is particularly important to carry out systematic and scientific research on conservation and restoration. In this study, we propose an artificial intelligence approach based on the remote sensing images of the relic landscape satellite as the basic research data, and use an adversarial generative neural network (GAN) to repair the damaged and missing sections of the relic landscape by feature learning and feature restoration, so as to provide new orientation interpretation and theoretical support for the study of the spatial layout of the relic landscape and subsequent conservation work. The results show that the spatial feature data and logical relationships can be effectively learned and restored in the monument landscape by using the deep learning specific data learning and parsing ability of computer.