| 交通视频背景估计动态高斯混合模型 |
| |
| 引用本文： | 李红宝,徐建闽,林培群,黄玲. 交通视频背景估计动态高斯混合模型[J]. 交通与计算机, 2009, 27(4) |
| |
| 作者姓名： | 李红宝 徐建闽 林培群 黄玲 |
| |
| 作者单位： | 1. 华南理工大学土木与交通学院,广州,510640;广州市中心区交通项目领导小组办公室,广州,510030
2. 华南理工大学土木与交通学院,广州,510640 |
| |
| 基金项目： | 国家自然科学基金资助项目，广州市科技攻关重点项目 |
| |
| 摘 要： | 为适应交通环境多变情况下实时交通信息动态采集的要求,提出了1种基于单高斯分布假设、图像分块处理、利用帧差法进行样本选取的动态混合模型对交通信息的图像背景进行快速估计,并在实际道路环境下对多种方法进行实验比较,证明该模型有效地削弱了运动物体和噪声对于背景估计的影响,具有较好的正确性和有效性.
|
| 关 键 词： | 智能交通系统 动态交通信息采集 图像处理 背景估计 高斯分布 |
Dynamic Gaussian Mixed Model for Traffic Image Background Estimation |
| |
| LI Hongbao,XU Jianmin,LIN Peiqun,HUANG Ling. Dynamic Gaussian Mixed Model for Traffic Image Background Estimation[J]. Computer and Communications, 2009, 27(4) |
| |
| Authors: | LI Hongbao XU Jianmin LIN Peiqun HUANG Ling |
| |
| Affiliation: | School of Civil Engineering and Transportation;South China University of Technology;Guangzhou 510640;China;Guangzhou City Center Transport Project Office;Guangzhou 510030;China |
| |
| Abstract: | For real-time traffic information collection in changing environments,a hybrid model was proposed based on single Gaussian estimation hypothesis,image separation and frame-difference method for fast background estimation.Moreover,some comparative experiments were conducted on different background estimation methods under real traffic situations.The results prove that the proposed model can effectively reduce the impacts of moving objects and noise on background estimation with quite high correctness and eff... |
| |
| Keywords: | intelligent transportation systems dynamic traffic information collection image processing background estimation Gaussian distribution |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
|
