Abstract: In order to improve the autonomous inspection ability of power poles and towers, through the combination of TOF technology and multi-visual navigation camera, the UAV can accurately identify the key components of power poles and towers, and realize the all-round inspection of poles and towers. In the inspection process, the advanced image processing algorithm is used to quickly analyze the collected images, so as to improve the positioning accuracy, optimize the inspection path planning, enhance the stability and accuracy of the UAV visual navigation, and effectively improve the efficiency and quality of the power pole and tower inspection. The visual navigation camera is used for visual navigation, and the position positioning during the UAV inspection process is realized. The test showed that the visual navigation system in this study has high positioning accuracy, error within 1 cm, and fast autonomous inspection speed.