Archive | 2019
The Implementation of Automatic Inspection Algorithm for Underwater Vehicles Based on Hough Transform
Abstract
An image detection method based on Hough transform was proposed to realize the target detection of autonomous underwater vehicles based on the detection and tracking of underwater pipelines by using single vision system. The image pixels were converted from 2D space coordinates to polar coordinate space. The strength of each point was normalized in polar coordinate space to make it be between 0 and 255. The pixel points of 2D space were found out according to that R value of polar coordinates was equal to the input parameter (the radius of the circle). Result color was given to the spatial pixels. Then, the results were returned to the 2D spatial pixel set and the detection of the oil spill was completed. At the same time, based on the characteristics of underwater pipeline image, a judgment method based on slope was proposed, which improved the extraction effect of pipeline edge and corner. Such algorithm was compared with the traditional algorithm, and it was proved that the proposed algorithm could effectively enhance the detection rate of the target and had good real-time performance. In addition, the feasibility and effectiveness of the system were verified through the pool test.