With the development of information technology and intelligent technology in power system, using intelligent inspection robot to inspect the operation state of primary and secondary equipment in substation has gradually become the trend of intelligent operation inspection. In this paper, an image fusion-based method is proposed to identify the operation state of the protection platen. Firstly, the high light interference in the image is detected quickly based on OTSU threshold segmentation optimization method; then, the improved RANSAC algorithm is introduced to complete the multi-view image feature matching and find the best perspective transformation matrix, using the auxiliary image perspective transformation to the main image to repair the high-light region; finally, on the basis of image repair, the operation state is judged by the dip angle of the edge detection of the platen. The experimental results show that the proposed method can better identify the operation state of the platen in the image and enhance its anti-interference ability.