| 基于动态粒子流场的视频异常行为自动识别 |
| |
| 引用本文： | 仉长崎,管业鹏. 基于动态粒子流场的视频异常行为自动识别[J]. 光电子．激光, 2015, 26(12): 2375-2380 |
| |
| 作者姓名： | 仉长崎 管业鹏 |
| |
| 作者单位： | 上海大学 通信与信息工程学院,上海 200444;上海大学 通信与信息工程学院,上海 200444 ;新型显示技术及应用集成教育部重 点实验室,上海 200072 |
| |
| 基金项目： | 国家自然科学基金(11176016,60872117)和高等学校博士学科点专项科研基金(20123108110014)资助项目 (1.上海大学 通信与信息工程学院,上海 200444； 2.新型显示技术及应用集成教育部重点实验室,上海 200072) |
| |
| 摘 要： | 为了有效实现视频异常 行为的自动识别,基于动态粒子流场,将视频运动对象的运 动行为,映射为有效 反映其运动变化状态的动态粒子流,同时提取度量不同场景内容下的运动方式各异的异常行 为的显著性运动特 征,进行异常行为的分类与识别。对来自不同场景并具有不同运动行为方式的公开视频 测试序列的实验表明,本文方法无需跟踪运动对象,也无需预先采集 异常行为样本进行学习与训 练,可在多种条件下实现视频运动对象异常行为的有效自动识别。
|
| 关 键 词： | 动态粒子流场 异常行为识别 显著性特征提取 智能视频监控 |
| 收稿时间： | 2015-08-18 |
Dynamic particle flow field based automatic recognition of video abnormal behavior |
| |
| ZHANG Chang-qi and GUAN Ye-peng. Dynamic particle flow field based automatic recognition of video abnormal behavior[J]. Journal of Optoelectronics·laser, 2015, 26(12): 2375-2380 |
| |
| Authors: | ZHANG Chang-qi and GUAN Ye-peng |
| |
| Affiliation: | School of Communication & Information Engineering,Shanghai University,Shang hai 200444,China;School of Communication & Information Engineering,Shanghai University,Shang hai 200444,China ;Key Laboratory of Advanced Display and System Applications,M inistry of Education of China,Shanghai 200072,China |
| |
| Abstract: | How to efficiently realize automatic r ecognition of abnormal behavior for intelligent video surveillance is a key problem.A method has been developed that the dynamic particle flow field from video is got based on the Lagrange dynamic system equation and self-adaptive determination of time interval in the equation.Some motion behaviors for motion objects in video are mapped to the dynamic particle flows w hich can be used to describe their motion variation states.Some significant motion features for abnormal behavior with different motion styles from different scenes have been extracted to classify and recognize the abnormal beha viors.Some open video test sequences from different scenarios with different behavioral patterns are selected to perf orm experimental verifications and comparisons.Experimental results show that abnormal behavior can be automaticall y recognized efficiently in various conditions where it is not necessary to track motion object or collect abnorm al behavior sample in advance for learning and training. |
| |
| Keywords: | dynamic particle flow field abnormal behavior recognition significant feature extraction intelligent video surveillance |
|
| 点击此处可从《光电子．激光》浏览原始摘要信息 |
|
点击此处可从《光电子．激光》下载全文 |
|
