一种用于运动目标跟踪与测量的目标边缘差分算法研究

Research on a Target Edge Difference Algorithm for Moving Target Tracking and Measurement

为提高对运动目标的有效追踪,同时有效避免传统差分算法对目标有效边缘造成的“空洞”破坏,提出了一种改进型目标边缘差分算法。一方面,通过引入一种基于INMF的目标跟踪算法以完成对目标区域的动态捕捉和分割;另一方面,对目标图像的边缘处理进行Canny灰度值特征提取,通过一种改进型边缘帧差分算法完成对分割区域内目标对象的边缘运动识别。最后,将该算法同传统几种典型的跟踪识别算法进行实验验证。结果表明:在目标部分遮挡、光线变化和运动模糊环境中,该算法在检测精度和稳定性方面表现较为优异。

In order to improve the effective tracking of moving targets,while effectively avoiding the"hole"damage caused by the traditional difference algorithm to the effective edge of the target,an improved target edge difference algorithm is proposed.On the one hand,by introducing an INMF-based target tracking algorithm to complete the dynamic capture and segmentation of the target area;on the other hand,the edge processing of the target image is performed by Canny gray value feature extraction,through an improved edge frame difference The algorithm completes the edge motion recognition of the target object in the segmented area.Finally,the algorithm is verified by experiments with several typical traditional tracking and recognition algorithms.The results show that the algorithm performs better in detection accuracy and stability in the environment of partial occlusion,light changes and motion blur.