Research On Space Object Detection And Tracking Algorithm

As the technology develops, the research on space object (as satellite, aircraft, fragment, planet in the out space) is the key problem in spatial attack and defense. And it includes the technology of spatial object recognition, orbit identification and estimation of position and velocity.The main purpose of the paper is the recognition and tracking of the spatial object, and has studied the mean filtering method, medium filtering method in image preprocessing algorithm; object extracting method by difference, optical flow method in object recognition algorithm; traditional matching algorithm and SSDA matching method in matching and tracking algorithm; small object detection algorithm by dynamic programming. And improved mean filtering algorithm, optical flow algorithm and SSDA matching algorithm are provided by analyzing and comparing the advantages and disadvantages of these algorithms.According to amount of emulations and experiments on platform, the image preprocessing algorithms and object recognition algorithms can achieve the propose of detection of objects in a long distance, capture moving objects and tracking, and it has made a good preparation for an intelligent vision platform system.