Computer Vision-based Monitoring Method for Differential Settlement of Shield Tunnels

Abstract

Shield tunnels in soft soils are inevitably subject to large differential settlement during construction and long-term operation, which can lead to various diseases and then impair structural safety and durability. Due to the inefficiency of traditional inspection methods, timely and accurate monitoring methods are necessary to ensure the safe operation and long-term maintenance of tunnels. In this study, a computer vision-based monitoring method for differential settlement of shield tunnels was developed and implemented. The monitoring area is located in an interval tunnel of a subway line in Shanghai during the freezing construction of its side channel. The influence of camera position’s micro-change was considered. The field equipment mainly consists of high-resolution industrial CMOS (complementary metal oxide semiconductor) digital camera sets, LED (light emitting diode) lights and power modules. The differential settlement relative to datum points are obtained and vertical displacement of one ring acquired by total station is compared with the proposed method to verify the feasibility of the monitoring system. The results indicate that the monitoring system is reliable and can be further applied.