The Ship Automatic Tracking Technique And Implementation Based On VTS Images

With the continuous development of science and technology now, vessel traffic analysis has a very important role in the field of marine traffic engineering, whether the formation of marine traffic engineering theory, or sea transport planning and management can not do without doing a lot of vessel traffic analysis. The ship tracking is a prerequisite for vessel traffic analysis. The traditional ship tracking technique is mainly to use the vessel traffic videos or lots of vessel traffic images from the VTS center, and then hand-draw the ship tracks map in artificial eye tracking methods to analyze the ship traffic in the waters. But there are many disadvantages with this method. First, when the target vessels are similar in size and their distance is close, using this method is easy to track the wrong target; second getting a lot of video images from the VTS center takes a large storage media, so it's time-consuming and costly, and the ship parameters are not precise enough through it. Therefore we want to use computer digital image processing method, develop a ship automatic tracking algorithm based on VTS images, and it will be verified and implemented through the MATLAB simulation.This paper introduces the study significance, the theoretical basis and the research status of ship tracking. A survey in ship tracking research is applied in this paper at first. Then, on the basis of collecting and analyzing large data in ship tracking field, this paper concludes a model that is suitable for the engineering application. To resolve the question of ship tracking under the fixed VTS image background, use Kalman prediction algorithm and a cost function algorithm of multi-target tracking have been analyzed in depth. In this thesis, with the help of MATLAB simulation tool, it accomplished the tracking of two ship targets and completed the algorithm emulation of every phase in ship tracking. From the satisfied result of simulation, the validity and the accuracy of the ship tracking algorithm has been proved.