|
|||||
|
|
| 一种基于监控视频的出租车识别方法 | |
| 引用本文： | 朱伟,叶飞,马超,张重阳.一种基于监控视频的出租车识别方法[J].电视技术,2013,37(7). |
| 作者姓名： | 朱伟 叶飞 马超 张重阳 |
| 作者单位： | 1. 上海市公安局青浦分局科技科,上海201700;上海交通大学图像通信与网络工程研究所,上海200240 2. 上海交通大学图像通信与网络工程研究所,上海,200240 3. 上海交通大学图像通信与网络工程研究所,上海200240;上海市数字媒体处理与传输重点实验室,上海200240 |
| 摘 要： | 研究了一种基于视频监控的出租车识别算法.对已经完成跟踪的车辆,通过提取车辆的方向梯度直方图(HOG)特征,作为支持向量机(SVM)分类检测的输入,进行车辆是否为出租车的分类识别.通过多窗口投票机制,增强了分类识别算法的准确性与鲁棒性.实验证明,该方法能准确进行出租车的分类识别,基于实际的标清监控视频,出租车的分类准确率达到90％左右. |
| 关 键 词： | 出租车识别 方向梯度 直方图 支持向量机 多窗口投票机制 |
| 收稿时间： | 8/7/2012 12:00:00 AM |
| 修稿时间： | 8/7/2012 12:00:00 AM |
Surveillance Video based Taxi-Car Identifying |
|
| ZHU Wei,YE Fei,MA Chao and ZHANG Chong Yang.Surveillance Video based Taxi-Car Identifying[J].Tv Engineering,2013,37(7). | |
| Authors: | ZHU Wei YE Fei MA Chao and ZHANG Chong Yang |
| Affiliation: | Qingpu Branch of Shanghai Public Security Bureau, Shanghai 201700,Shanghai Jiao Tong University,Shanghai Jiao Tong University,Shanghai Jiao Tong University |
| Abstract: | Identifying the type of vehicles using video or image is one of the hot spots in computer vision. This paper presents a taxi-car identifying system based on surveillance video: For one vehicle that have been tracked in a surveillance video, its HOG (Histogram of Direction Gradient) features are extracted firstly, which are then taken as the input of SVM (Support Vector Machine) based classifier to indentify its catalog: Taxi-Car or Not. One multi-window voting mechanism is developed in this system to improve the accuracy and robustness of the classifying. The experiments results show that this method can identify the taxi in the surveillance effectively. |
| Keywords: | Taxi-Car Identifying System HOG SVM Multi-Windows Voting Mechanism |
| 本文献已被 万方数据 等数据库收录！ | |
| 点击此处可从《电视技术》浏览原始摘要信息 | |
| 点击此处可从《电视技术》下载全文 | |