Research of Obstacle Recognition Based on Vision for High Voltage Transmission Line Inspection Robot

Abstract

Obstacle recognition and orientation is one of the key techniques in high voltage transmission line autonomous inspection robot.In light of the structure of 220 kV single split transmission line,a method of obstacle recognition based upon vision sensor is put forward.First of all,the shot image is processed by expansion,erosion,Gauss smoothing treatment to relieve the noise.And then,Otsu is used to improve Canny algorithm so as to detect the edge of the image,which can reduce the influence of the light changing.At last,some graph cells are detected and given structure constraint in the edge of image to achieve obstacles recognition.A lot of tests proved that structures such as wire,counterweight,strain clamp and suspension on high voltage transmission line can be effectively recognized using the method.