Search

Three-dimensional (3D) reconstruction of substation fittings is of great significance for live working robots. However, the key problem is that active 3D cameras cannot work in outdoor strong light environment, and the passive 3D cameras cannot extract features of weak texture targets. This paper proposes an outdoor 3D reconstruction method based on multi-line laser and binocular vision. To solve the problem that weak texture target has few features, we use multi-line laser to create artificial features. To reduce the interference of natural light on the laser in the images, the frame-difference method is proposed for natural light filtering. Then we use the gray-centroid method to position the multi-line laser accurately. Finally, the binocular vision model is used to complete 3D reconstruction of the target. The experiments show that, compared with traditional 3D reconstruction methods, our 3D reconstruction method can realize 3D reconstruction of outdoor weak texture targets effectively.

The identification and positioning of insulator shed feature is of vital significance for insulator cleaning. However, there exist few texture features the insulator and its surface is reflective, which makes it difficult to identify and locate the convex points on the surface of the insulator shed. In order to identify and locate the insulator shed, we propose a method to extract the convex points of the insulator skirt based on the binary image of the insulator and the position of the central axis. Binocular vision technology is applied to obtain the depth change curve of the insulator on its axis, with the help of the depth threshold method and principal component analysis(PCA) method. At the same time, the pixel coordinates on the central axis are used to traverse the depth on the central axis to extract the shed bumps, thereby achieving the spatial positioning of the insulator shed bumps. The experiments show that the bumps on the insulator’s central axis can be well marked and used in the insulator cleaning robot system.

Current methods for insulator extraction are mostly used for aerial imagery. There are few insulator extraction methods suitable for live working robots on mobile platforms. This paper proposes an insulator segmentation method based on depth image and color image (ISBDC). Firstly, ISBDC divides the image according to the depth information acquired by the depth camera, and removes the background image with a long distance to obtain the approximate position of the insulator in the image. Then, using OSTU to extract contours of the insulator. Finally, according to the distribution of the pixels, the background noise is further removed and the small connected region is deleted to get the final result. Compared with other algorithms, ISBDC can suppress more background noise (especially when there are multiple interference insulators in the image) and obtain enough insulator details while ensuring simple structure and not requiring much data to train.