ABSTRACT
The rapid development of urban areas necessitates strict control over the impacts of underground construction. Ground loss during tunnel excavation is a major contributor to ground settlement, but accurate methods for estimating ground loss on-site are limited. In Earth Pressure Balance shield tunneling, the excavated muck volume is closely related to tunneling-induced ground loss. This study introduces an innovative method using a depth camera to continuously capture depth images of muck for estimating excavated soil volume. An image processing workflow specifically designed for muck depth images captured in shield tunneling environments and a detailed calculation process for estimating the volume of muck flowing on the conveyor belt are presented in the research. This method is applied to Metro Line 18 in Shanghai and it achieves non-contact, high-precision, automated real-time estimation of muck volume for each construction ring. This is of great significance for predicting and controlling the excavation-induced ground settlement.
