Research on autonomous landing of UAV based on vision sequence image

Through the analysis of the various landing modes of the current UAVs, based on the full use of the passive characteristics of visual information, a full visual information guided landing method based on visual sequence images is proposed, and a new color guiding function is designed. Multi-graphic cooperation goal, using the method of air-ground cooperative visual matching, through the selection of cooperative target and the construction process of coordinate system, the specific calculation process of flight attitude information of UAV is described, and the real-time performance of visual information processing during UAV landing process is guaranteed. Under the premise of accuracy and accuracy, the recognition and processing of cooperation targets are completed by means of grayscale transformation, HSV color transformation and HU constant distance matching. The landing mode study provides a valuable reference for the UAV's full visual autonomous landing.