Abstract: Traditional methods for measuring segment dislocation in shield tunnels suffer from disadvantages such as high equipment costs and low computation efficiency. Therefore, an intelligent inspector based on the C/S architecture was designed and developed. In terms of hardware, the Raspberry Pi 4B was employed as the main control module, integrated with a binocular camera equipped with a CMOS sensor and an LED display, to enable real-time image acquisition and display. On the software side, a decoupling of the front-end and back-end functions was implemented based on the TCP/IP protocol. The client (front-end), deployed on the intelligent inspector, handled camera invocation, image acquisition, data transmission and result visualization. The server (back-end) deployed on a consumer-grade computer, was responsible for the data listening, invoking the core binocular vision ranging algorithm, and returning the results. The proposed inspector was applied to a section of Fuzhou Binhai Express Line shield tunnel. The experimental results showed that the calculated results agreed well with the measurements from a weld gauge, achieving a sub-millimeter accuracy in segment dislocation measurement. A single detection took only 5.40 seconds, effectively improving inspection efficiency while ensuring measurement accuracy. This research provides a new approach for the rapid and accurate inspection of segment dislocation in shield tunnels.
