| 基于多特征融合的运动物体识别方法研究 |
| |
| 引用本文： | 纪筱鹏,魏志强,冯业伟,王鹏.基于多特征融合的运动物体识别方法研究[J].哈尔滨工业大学学报,2009,41(3):169-171,235. |
| |
| 作者姓名： | 纪筱鹏 魏志强 冯业伟 王鹏 |
| |
| 作者单位： | 纪筱鹏,魏志强,冯业伟,JI Xiao-peng,WEI Zhi-qiang,FENG Ye-wei(中国海洋大学,计算机科学系,山东,青岛,266100);王鹏,WANG Peng(中国电信集团公司,青岛电信分公司,山东,青岛,266000)
|
| |
| 基金项目： | 国家自然科学基金，国际科技合作计划项目，山东省重点自然科学基金，国家高技术研究发展计划(863计划) |
| |
| 摘 要： | 为了实现运动目标的实时检测,本文基于多特征融合,提出了一种运动物体分级识别方法.首先通过对运动目标的分割,分析得到各个物体的形状矩特征、面积大小及运动速度.对上述特征,采用支持向量机方法对其矩特征进行快速的一级分类,然后对一级分类结果及各物体的面积、速度等信源根据其对分类的重要程度,采用模糊积分技术进行多特征融合分类.实验表明,本方法具有较强的学习能力及较高的识别精度,可有效地应用于道路监控等系统.
|
| 关 键 词： | 多特征融合 支持向量机 模糊积分 矩特征 目标分类与识别 |
Moving object recognition based on multi-feature fusion |
| |
| JI Xiao-peng,WEI Zhi-qiang,FENG Ye-wei,WANG Peng.Moving object recognition based on multi-feature fusion[J].Journal of Harbin Institute of Technology,2009,41(3):169-171,235. |
| |
| Authors: | JI Xiao-peng WEI Zhi-qiang FENG Ye-wei WANG Peng |
| |
| Affiliation: | 1.Computer Science Department,Ocean University of China,Qingdao 266100,China;2.Qingdao Municipal Branch Shandong Provincial Branch Corporation CHINA TELECOM,Qingdao 266000,China) |
| |
| Abstract: | To realize the real-time detection of moving targets,a hierarchy of image recognition based on multi-feature for moving objects is proposed in this paper.Firstly,the shape moment features,size and velocity of each object are obtained through the segmentation of moving objects.Secondly,support vector machines are used to apply the first classification to above features.Then fuzzy integral is used to apply the multi-feature fusion classification to the first classification result,the size and the velocity based on the importance degree.The experiments indicate that the learning capability and accuracy of this approach are good,and it can be applied in road monitoring system efficiently. |
| |
| Keywords: | multi-feature fusion support vector machine fuzzy integral moment feature target classification and recognition |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
|
