优化CamShift算法的适应度函数的船舶跟踪改进算法

Improved Algorithm of Ship Tracking by Optimizing the Fitness Function of the CamShift Algorithm

针对内河船舶对象在动态背景下的目标跟踪展开研究,考虑船舶在内河运行环境的特点,以Cam Shift跟踪算法为基础,对其适应度函数进行优化设计,将期望值最优作为适应度函数的性能参量。设计中以目标区域颜色直方图概率密度和预测区域颜色直方图概率的期望值作为适应度函数,以此适应度作为视频帧中选择目标区域的参量。内河运行船舶实际摄制视频的跟踪分析结果表明:相对于传统的Cam Shift算法,改进后的算法提高了跟踪精度和系统稳定性。

The target tracking of ships in inland river is studied. Considering the characteristics of the ship in the inland river, the fitness function is optimized based on the CamShift tracking algorithm, and the best performance parameters of the fitness function are optimized by the expected value. In the design the expected value of the probability density of the region color histogram is used as the fitness function, which becomes a parameter of selecting the target area in the video frame. Through the tracking analysis of the actual video of the ship in inland river, compared with the traditional CamShift algorithm, the improved algorithm designed in this paper improved the tracking accuracy and stability of the system.