Novel System for Rapid Investigation and Damage Detection in Cultural Heritage Conservation Based on Deep Learning

Abstract

Rapid investigation and damage assessment are crucial for cultural heritage conservation. At present, mobile crowd sensing (MCS) techniques are very effective for cultural heritage investigation and data collection. Unfortunately, data collected based on MCS techniques cannot be fully utilized and analyzed. To overcome this limitation, this study combines MCS techniques and a state-of-the-art deep learning algorithm to realize rapid investigation and damage detection of the Great Wall in China. The GreatWatcher system, based on MCS techniques and a deep learning algorithm, was developed in this study, focusing on big data collection and damage detection for the Great Wall. The system highlights the significance and emerging revolution of the combination MCS techniques with deep learning methods in the cultural heritage field. System components include a mobile client (data collection), web platform (data storage database), and computing terminal (data analysis and automatic damage detection). Two field investigations and data collection for the Great Wall were performed to verify the feasibility and effectiveness of the system. Based on the collected data, a deep learning method was used to automatically analyze damage to the Great Wall at the computing terminal. Moreover, various validation experiments of different conditions were performed to verify the good performance of the deep learning method.