【摘要】 ASM是一种基于统计模型的目标定位方法。在多种目标跟踪研究的基础上,提出了一种基于ASM的彩色视频中运动物体的定位方法,该方法充分利用了彩色视频中各种不同的颜色信息,同时为提高ASM方法的通用性,将原有的一维搜索空间扩展到二维邻域搜索空间,提高了定位过程的抗干扰性。实验结果表明,该方法在定位的准确性和鲁棒性上有较大提高。更多还原

【Abstract】 Active Shape Model(ASM) is a method for object localization based on statistic model.On the basis of the existing research of object tracking,this paper presented a method for localizing the moving object in color video based on ASM.This method made the best use of different color information in color video.To improve the generalization,it extended one-dimension search space to two-dimension neighborhood search space so that it performs very well when being applied to object localization.The experimental results show that the proposed method acquires much better precision and robustness than the conventional one.更多还原