| 基于高斯背景模型的车辆检测改进算法 |
| |
| 引用本文： | 胡学刚,刘忠振.基于高斯背景模型的车辆检测改进算法[J].计算机工程与设计,2011,32(12):4111-4114. |
| |
| 作者姓名： | 胡学刚 刘忠振 |
| |
| 作者单位： | 1. 重庆邮电大学数理学院,重庆,400065
2. 重庆邮电大学计算机学院,重庆,400065 |
| |
| 基金项目： | 国家自然科学基金项目，重庆市教委科研基金项目，重庆邮电大学自然科学基金项目 |
| |
| 摘 要： | 针对车辆检测中实时性、准确性以及自适应性很难兼顾的问题,提出了一种快速有效的车辆检测算法。该算法采用阈值判定背景区域和更新区域,根据背景变化的程度,利用动态权值更新学习率,使用基于单高斯背景模型方法,结合逻辑"或"运算检测出车辆。实验结果表明,该算法能够快速准确地检测出车辆,并且在背景更新过程中实现了自适应,并具有一定的鲁棒性。
|
| 关 键 词： | 车辆检测 单高斯背景模型 背景更新 背景差分 二次帧差 |
Improved vehicle detection algorithm based on Gaussian background model |
| |
| HU Xue-gang,LIU Zhong-zhen.Improved vehicle detection algorithm based on Gaussian background model[J].Computer Engineering and Design,2011,32(12):4111-4114. |
| |
| Authors: | HU Xue-gang LIU Zhong-zhen |
| |
| Affiliation: | HU Xue-gang1,LIU Zhong-zhen2(1.College of Mathematics and Physics,Chongqing University Posts and Telecommunications,Chongqing 400065,China,2.College of Computer,China) |
| |
| Abstract: | For the problem that it is difficult to take into account real-time,accuracy and adaptability together in vehicle detection,a fast and effective vehicle detection algorithm is proposed.By using threshold values,the background area and the update region are determined,and by using dynamic weights,learning rates are updated in accordance with the degree of changes in background.With logical or operator,vehicles are detected based on a single Gaussian background model.The experimental results show that this ... |
| |
| Keywords: | vehicles detection single-Gaussian background model background updating background subtraction dual subtraction |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
|
