A pointer meter reading recognition method based on improved ORB algorithm for substation inspection robot

Abstract Currently, pointer meters in substations are read manually, which causes a waste of human resources. To realize automatic reading such meters, an accurate method for recognizing and was proposed the inspection robot substation. The new consists three steps. Step 1: image is preprocessed with grayscale, noise reduction piecewise linear grayscale transformation. 2: Gaussian scale space used to enhance invariace ORB algorithm. Meanwhile, RANSAC algorithm screen matching point pairs, improves accuracy feature matching. 3: Hough transform utilized fit pointer. Finally, result obtained according angle prior information meter. approach can recognize meter accurately. It has high reliability engineering application value.