Research Of Air-Ground Visible Image-Guidance-Tracking

To solve the problem of air-ground visible Image Guidance Tracking in a complex background, this paper presents an air-ground visible Image Guidance Tracking algorithm based on image feature matching. Characteristic information of the target are obtained from target area of current frame image through feature points sampling by combining uniformly-spaced sampling and improved Kirsch edge detection method. Then analyzes the obtained samples set of feature point by optical flow to get the change of feature points in adjacent frame of image target area. The template matching can realize feature recognition and filtrate from samples of feature points by compare the adjacent two frame images, and then the target area is restored. According to the change of target area in two adjacent frames, the guidance tracing of ground target can achieve. The result show that in the process of tracking, the average time of the algorithm is less than7.75ms, the maximum frame rate is129Hz. Speed, accuracy and stability fully meet the goals of modern spatial maneuvering target intelligent tracking and guidance system.