| 基于Hough变换的车窗提取算法 |
| |
| 引用本文： | 李冬梅,郝晓莉. 基于Hough变换的车窗提取算法[J]. 中国测试技术, 2007, 33(3): 50-52,83 |
| |
| 作者姓名： | 李冬梅 郝晓莉 |
| |
| 作者单位： | 北京交通大学电子信息工程学院,北京,100044 |
| |
| 基金项目： | 北京交通大学校科研和教改项目 |
| |
| 摘 要： | 在HOV乘客数检测系统中，对车窗的定位与提取，可以极大地减少计算量，提高系统的运算速度以及检测的准确度。针对图像中车窗边缘的图像特点，提出了一种基于相位编组法进行图像分块，在图像块内进行决速Hough变换的直线检测，并结合积分投影方法对车窗进行定位与提取。实验结果表明该算法具有较快的运算速度和较高的检测准确率。
|
| 关 键 词： | 车窗定位 Hough变换 积分投影 灰度变换 模板滤波 |
| 文章编号： | 1672-4984（2007）03-0050-03 |
| 修稿时间： | 2006-10-202006-12-27 |
An window location algorithm based on Hough transform |
| |
| LI Dong-mei,HAO Xiao-li. An window location algorithm based on Hough transform[J]. China Measurement Technology, 2007, 33(3): 50-52,83 |
| |
| Authors: | LI Dong-mei HAO Xiao-li |
| |
| Affiliation: | School of Electronics and Information Engineering, Beijing Jiaotong University,Beijing 100044, China |
| |
| Abstract: | It is important to count vehicle occupants in H0V implementation. Locating and detecting vehicles windows can greatly reduce the calculation work. According to the characteristics of vehicle windows, we employed Hough transform and Integration projection to cut the window regions. Experiments show that the algorithm has merits of high computation speed and has value in practice. |
| |
| Keywords: | Vehicle window localization Hough transform Integration projection Gradient image Template filtering |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
