Real-time Image Processing Technology On Location Solver Of Motion Target

Aerial refueling technology, which is as known as the air force multiplier, has been widely usedin modern war. It is a hotspot of research. At present, aerial refueling technology in our country is stillrelatively backward. The process of the refueling can be succeeded after many docking incoordination of guide member and pilot, so it is very complicated and risky. Thus, it is necessary todevelop a set of assistant aerial refueling guidance system. This thesis will research on the detection,tracking and positioning of the target taper sleeve with image technology.Firstly, the thesis will research on the technology of image-preprocessing. Thresholdsegmentation, Morphological Filter and Gaussian filter will be applied to V channel which can extractcolor channel image from RGB model of the target image. Through combination of differentimage-preprocessing methods, piecewise multiple image-preprocessing will be achieved which canextract image feature information needed in different distances of the taper sleeve and camera indifferent stages of image-processing.Secondly, the accuracy of the detection algorithm for the target will be studied on. The secondarydetection algorithm of the target based on Least Squares Fit will be applied when the taper sleevedistance is far. While in the case of close range, detection algorithm based on the contour extractionwill be applied. According to the difference of distance between taper sleeve and camera, differentdetection algorithms will be applied, so as to realize accurate detection of target taper sleeve indifferent distance.Furthermore, the stable tracking algorithm will be researched on. According to the features ofvideo image sequences, detection and tracking methods in different distance modes will be designed,and the handoff scheme between modes will also be put forward. The efficiency of tracking algorithmwill be improved through the template matching method which ensures the stability and real-time ofthe tracking algorithm. Based on the pinhole camera model, the taper sleeve monocular visionlocalization algorithm will be studied.Finally, the system of real-time collecting and processing of images are designed and built bythis paper to test the detection algorithm、tracking algorithm、localization algorithm of target tapersleeve under different outdoor backgrounds、illumination and motion states. The experiment resultshows that the methods which has good real-time processing capacity put forward in this thesis canrealize the effective detection, stable tracking and precise positioning of the target taper sleeve.