| 泊位引导系统中运动物体的检测与分类 |
| |
| 引用本文： | 高庆吉,支源,胡琢,王立文.泊位引导系统中运动物体的检测与分类[J].中国民航学院学报,2007,25(1):7-11. |
| |
| 作者姓名： | 高庆吉 支源 胡琢 王立文 |
| |
| 作者单位： | 中国民航大学航空自动化学院，中国民航大学航空自动化学院，东北电力大学自动化学院，中国民航大学科技处 天津300300中国民航大学智能信号与图像处理天津市重点实验室，天津300300，天津300300，吉林吉林132012，天津300300 |
| |
| 摘 要： | 重点研究了基于视觉检测方式的机场泊位自动引导系统的图像预处理算法。提出改进的基于混合高斯模型的运动检测算法,有效抑制了分割运动缓慢且成像面积比较大的物体时易出现的“拖尾”现象。提出了基于颜色色度信息的阴影检测算法,成功地检测出运动物体的阴影。在此基础上设计了分类器,区分泊位飞机与泊位场景中其他运动物体。在实际机场泊位自动引导系统中的应用证明了所提出算法的有效性和实用性。
|
| 关 键 词： | 可视化泊位引导系统 改进混合高斯模型 阴影检测 模式识别 |
| 文章编号： | 1001-5000（2007）01-0007-05 |
| 修稿时间： | 2006-08-262006-11-06 |
Study on Detection and Classification Algorithm of Moving Object in Docking Guidance System |
| |
| GAO Qing-ji,ZHI Yuan,HU Zhuo,WANG Li-wen.Study on Detection and Classification Algorithm of Moving Object in Docking Guidance System[J].Journal of Civil Aviation University of China,2007,25(1):7-11. |
| |
| Authors: | GAO Qing-ji ZHI Yuan HU Zhuo WANG Li-wen |
| |
| Institution: | 1.College of Aerospace Automation;2.Tianjin Key Lab for Advanced Signal Processing; 3.Science and Technology Department, CAUC , Tianjin 300300, China; 5.Department of Automation Control Engineering, Northeast Dianli University, Jilin 132012, China |
| |
| Abstract: | The paper focuses on studying the image preprocessing algorithm of visual docking guidance system.The improved algorithm of motion detection based on mixture of Gaussians is proposed to eliminate the "bad-tail" appearance occurring in background subtraction of slowly moving and large objects.The new algorithm based on color hue is put forward to detect the shadow of moving object,and the shadow is detected successfully.On this basis,classifier is made in order to separate the docking aircraft from other moving objects.The experimental results in practical visual docking guidance system prove the validity and practicability of the algorithms. |
| |
| Keywords: | visual docking guidance system improved mixture of gaussians algorithm shadow detecting recognition |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
