Rapid Detection Method of Pipe Inner Diameter Based on Binocular Vision

Abstract

Abstract With the rapid development of robotics, automatic pipeline detection has become a research hotspot in recent years. The existing pipeline diameter automatic detection method needs to calculate a large number of clouds, which is inefficient and difficult to play a role in real-time pipeline detection. Therefore, this paper proposes a fast detection method based on binocular vision to achieve real-time detection of the inner diameter of the pipeline. According to the gradient diagram of the color image after the down-sampling, adaptive double-threshold edge detection is proposed to calculate the edge points. Then an improved BM stereo matching algorithm is proposed to obtain the disparity map of edge points quickly. The least square method is finally used for fitting the edge point coordinates to calculate the inner diameter of the pipeline. The experimental results show that the system has strong accuracy and high efficiency, which can meet the needs of real-time detection of the inner diameter of the pipeline.