Research on intelligent detection method of ancient building damage characteristics in deep learning environment

Abstract

In view of the problem that the strength value of damage features obtained by existing intelligent feature detection methods is too small, which leads to poor coupling between building damage features, this paper designs an intelligent detection method of ancient building damage features in deep learning environment. After the long-term load numerical model of ancient buildings is constructed by using deep learning technology, the damage characteristics of bearing columns of ancient buildings are calibrated and processed into a distance term and an angle term, and the intelligent detection numerical relationship of damage characteristics is constructed. According to the ancient building damage feature extraction results, the GNSS coordinates of damage points are collected, and the detection methods based on low-level topology, damage mechanism and the designed detection methods are used for experiments. The experimental results show that the designed intelligent detection method has the largest value of damage feature strength, the best coupling between damage feature indexes, and the practical application effect is good.