Study On Algorithm For Aerial Target Real-time Tracking And System Design

The target recognizing and tracking technique is widely applied to many fields such as industries, navigation, weapon controlling and guiding and so on. Its application in military field has become the research hotspot. The stability of target tracking is the most important in military applications. However, it is difficult to recognize the aerial target because of the unpredictable circumstance (light,weather,visibility) and noise interference , especially from long distance, which affects the performance of stable tracking.The article studies the problem of aerial missile tracking. The tracking algorithm in this article is based on template correlation algorithm, combining with target character matching. This tracking algorithm can meet the demands of real-time object tracking. Firstly, it utilizes the improved template correlation algorithm to achieve coarse matching, getting the most possible target points from the sequence image. Then with pattern identification methodology, it uses target character matching to achieve fine matching, thus to realize accurate tracking of target. This improved method overcomes the disadvantages of common correlation method: too complicated to be used in real-time applications. It has stronger ability of anti-disturbance. The usage of character matching guarantees the tracking accuracy. To further improve the target tracking speed and stability, the article uses the window tracking method and the target movement estimation method to forecast target location, and adopts the reliable tracking policy. These methods successfully save the tracking time and improve tracking accuracy. In this way, we can track the missile steadily, even when the target is sheltered by the barrier or disturbed by noises. The tracking algorithm does work successfully on the tracking system based on PC platform. During the tracking procedure, the system adjusts the platform's direction according to target image location, and keeps the target image around the center of the CCDs' vision fields. The article also describes the tracking system's structure and function definition. The tracking system can also measure the target distance using the double computer visions.