A Study on Vision-based Rust Boundary Recognition for Derusting Robot

Abstract

In this paper, a new type of surface rust detection for wall-climbing robot is proposed in order to surmount the shortcomings of current wall-climbing-cleaning robots. It takes advantage of OpenCV to establish a detection system, and uses QT to do the experimental analysis. The experimental results show that the detection system can complete the rust spot recognition of the image of the wall-climbing-cleaning robot, which makes the foundation for the intelligent trajectory planning of the wall-climbing-cleaning robot.