Research On Detecting And Tracking Small Target In Infrared Image Sequences And DSP Implementation

The technology of small target detecting and tracking is a meaningful subject in computer vision, It has a lot of applications in many fields such as missile controlling-and-guiding, video surveillance, medical treatment, traffic control, etc. Since IR imaging possesses many advantages, it has become the trend and direction for current development and study of tracking technologies worldwide.This thesis introduced basic theories and characteristics of infrared imaging and made particular studies on small target recognition and target tracking.This paper introduced a label algorithm based on downwards region growing combined with image processing algorithm. Compared with the traditional two-value target recognition algorithm, the new algorithm is easy to understand and use and with higher computation speed. This paper also introduced three tracking algorithms of small target as polynomial prediction, motion estimation and pixel labeling combined with window tracking algorithm. Compared with the traditional window tracking algorithm , the new algorithms have strong ability of resistance to the noise and better tracking performance. At last, this paper described the system operational principle, introduction of hardware, and the realization of algorithm in great detail.The emulation results on the hardware platform showed that the hardware run stably and was with strong anti-interference ability. The software programmed according to the algorithm could recognize the target accurately and quickly in real-time well, which satisfied the design requirements.