| 基于粒子视频的高密度人群主流运动检测 |
| |
| 引用本文： | 章东平,童超,芦亚飞.基于粒子视频的高密度人群主流运动检测[J].电子技术应用,2012,38(4):123-125. |
| |
| 作者姓名： | 章东平 童超 芦亚飞 |
| |
| 作者单位： | 中国计量学院信息工程学院,浙江杭州,310018 |
| |
| 基金项目： | 浙江省自然科学基金资助项目，浙江省科技计划资助项目 |
| |
| 摘 要： | 采用粒子视频流获得视频序列中的特征点运动轨迹,并对获得的运动轨迹进行提取,然后利用最长共同子序列LCS(Longest Common Subsequence)聚类轨迹,得到运动的主流方向。该算法可以有效检测实际场景中的主流运动方向。
|
| 关 键 词： | 人群分析 轨迹聚类 最长公共子序列 粒子轨迹 |
Dominant motions detection in dense crowds based on particle video |
| |
| Zhang Dongping
,
Tong Chao
,
Lu Yafei.Dominant motions detection in dense crowds based on particle video[J].Application of Electronic Technique,2012,38(4):123-125. |
| |
| Authors: | Zhang Dongping Tong Chao Lu Yafei |
| |
| Affiliation: | (College of Information Engineering China Jiliang University,Hangzhou 310018,China) |
| |
| Abstract: | In this paper,we present a novel method to automatically identify dominant motion pattern in crowd scenes.The algorithm begins by using long range motion estimation-particle video to obtain the trajectories of feature points.Compared to many approaches using optical flow obtain feature point trajectory,particle video can obtain more reliable particle trajectory.Then those particle trajectories can be clustered into dominant motions based on longest common subsequences.Results on real video se-quences demonstrate that the approach can successfully identify both dominant in crowded scenes. |
| |
| Keywords: | crowd analysis trajectory clustering longest common subsequences particle trajectory |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
|
