基于改进的DSSIM跟踪算法的行人异常行为检测-《集美大学学报（自然科学版）》

文章信息/Info

Title:: Abnormal Behavior Detection of the Pedestrian Based on Modified Differential Structural Similarity

Author(s):: LI Teng-fang¹; CHEN Shui-li²; (1.College of Mathematics and Computer ScienceFuzhou UniversityFuzhou 350002China;2.School of ScienceJimei University Xiamen 361021China)

Keywords:: DSSIM; pedestrian tracking; affine transformation; gray prediction; abnormal behavior

摘要:: 提出了改进型微分结构相似度 (Differential Structural Similarity ，DSSIM）算法，用核均值亮度计算SSIM中的亮度均值，并在跟踪过程可能发生遮挡、跟错的情况下，利用GM（1,1）灰色预测模型预测行人可能出现的位置，提高算法的跟踪效果.在稳定跟踪的基础上，对跟踪窗口进行仿射变换，使跟踪窗口可以进行旋转，从而能够检测探身、摔倒.实验结果表明，算法能够在简单场景的跟踪过程中快速有效地检测出游荡、探身、摔倒的人体异常行为

Abstract:: The modified Differential Structural Similarity (DSSIM) is presented in this paper.The kernel mean luminance is adopted to calculate the luminance in SSIM and the GM(1,1) gray prediction model is employed to predict the approximate pedestrian location when pedestrians blocking each other or the tracking may be wrong.The method can improve the result of tracking.By making affine transformation of the tracking window ，the tracking window can be rotated to detect bending and tripping. The experiment results show the abnormal behavior of pedestrian such as loitering，bending and tripping in the simple scenario can be effectively detected