| 基于模糊积分的运动目标分类算法 |
| |
| 引用本文： | 李俊韬,刘丙午.基于模糊积分的运动目标分类算法[J].微计算机信息,2007,23(34):303-304. |
| |
| 作者姓名： | 李俊韬 刘丙午 |
| |
| 作者单位： | 北京物资学院,北京,100081 |
| |
| 基金项目： | 基于决策支持技术的物流管理智能化研究项目 |
| |
| 摘 要： | 交通视频监控中的运动目标分辨率较低且投影变形严重，特征提取过程存在很多的不确定性，其目标分类问题很难解决。利用模糊测度和Choquet模糊积分的方法进行目标的分类算法。由于同时利用目标的形状特征信息和运动特征信息，提高了分类器的分类精度。利用实际的交通场景的运动目标测试结果表明，分类器不仅满足一定的分类精度，而且计算简单。
|
| 关 键 词： | 视频监控 目标分类 模糊测度 Choquet模糊积分 |
| 文章编号： | 1008-0570（2007）12-1-0303-02 |
| 修稿时间： | 2007-09-03 |
Moving Objects Classification Algorithm Based on Fuzzy Integral |
| |
| LI JUNTAO,LIU BINGWU.Moving Objects Classification Algorithm Based on Fuzzy Integral[J].Control & Automation,2007,23(34):303-304. |
| |
| Authors: | LI JUNTAO LIU BINGWU |
| |
| Abstract: | Moving object of traffic video surveillance is little and distorted. Features of moving object have many uncertain factors. Object Classification is an unsolved problem. An algorithm based on fuzzy measures and fuzzy integral is presented. The Classifiers performs hybrid classification which combines motion and appearance features. Very promising experimental results indicated that the proposed algorithm is effective with strong robustness using real image sequences of traffic video. |
| |
| Keywords: | Video Surveillance Object Classification Fuzzy Measurement Choquet Fuzzy Integral |
| 本文献已被 维普 万方数据 等数据库收录！ |
