| 视频运动目标检测方法的对比分析 |
| |
| 引用本文： | 赵文哲,秦世引. 视频运动目标检测方法的对比分析[J]. 科技导报(北京), 2009, 27(10) |
| |
| 作者姓名： | 赵文哲 秦世引 |
| |
| 作者单位： | 北京航空航天大学自动化科学与电气工程学院,北京,100191;北京航空航天大学自动化科学与电气工程学院,北京,100191 |
| |
| 基金项目： | 国家高技术研究发展计划(863计划),国家自然科学基金,中国科技部国际科技合作项目,中国教育部高等学校博士学科点专项科研基金 |
| |
| 摘 要： | 针对视频运动目标检测领域的研究现状,对帧间差分法、背景建模法、图像分割法、聚类分析法、运动矢量场法等视频运动目标检测方法的算法适应性、复杂度等性能进行了对比分析.辅以典型的视频运动目标检测实例,比较了各种方法的适用范围与优劣特点,可为不同应用场合下视频运动目标检测方法的选择提供参考依据.指出了现阶段视频运动目标检测技术研发的瓶颈所在,以及在新的应用需求背景下所面临的挑战,并对进一步的发展趋势作了预测和评估.
|
| 关 键 词： | 目标检测 帧间差分法 背景建模法 AdaBooat分类 运动矢量场 |
Comparative Study on Detection Methods for Video Motion Targets |
| |
| ZHAO Wenzhe,QIN Shiyin School of Automation Science , Electrical Engineering,Beihang University,Beijing ,China. Comparative Study on Detection Methods for Video Motion Targets[J]. Science & Technology Review, 2009, 27(10) |
| |
| Authors: | ZHAO Wenzhe QIN Shiyin School of Automation Science Electrical Engineering Beihang University Beijing China |
| |
| Affiliation: | ZHAO Wenzhe,QIN Shiyin School of Automation Science , Electrical Engineering,Beihang University,Beijing 100191,China |
| |
| Abstract: | This paper reviews the current development of detection methods for video motion targets, with a comparison of a series of related algorithms, including frame difference method, background subtract method, background modeling, method based on image segmentation, motion vector field analysis method and others, together with their application ranges and complexity. Performances of several target detection methods are tested with video samples, based on which the accuracy, the level of realtime and independenc... |
| |
| Keywords: | target detection frame difference background modeling AdaBoost classification motion vector |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
