【摘要】 提出一种自行设计的视频序列目标跟踪系统,采用C/S模式将视频数据传送给Linux终端,通过颜色直方图反投影,将每帧RGB输入图像转换为二维的颜色概率分布图像,基于CAMShift算法计算目标跟踪窗口的位置和大小。Linux终端的图形用户界面采用GTK+图形库设计。为了提高数据的传输效率,在客户端引入定时器,对解压后的视频数据定时读取。实验结果表明,在目标有遮挡、旋转等情况下,系统仍能有效跟踪,能够满足实际应用的需求。更多还原

【Abstract】 Proposes an object tracking system in video sequences, the compressed video data is transmitted to the Linux terminal based on C/S mode. For each video frame, the raw image is converted to a 2-D skin color probability distribution image via the color histogram back-projection. CAMShift algorithm is then applied to find the mode (peak) of t he probability distribution, which estimates the center and size of tracking window. The GUI of Linux terminal was designed by GTK+. In order to improve data transmission efficiency, timer was adopted to read the decompressed image data periodically on the client side. The experimental results show that the program tracks reliably in shading and rolling environment, and can meet the practical application’s demands.更多还原