适用于机场跑道异物检测的区域生长改进算法  

Improved region growth algorithm applicable to detection of foreign object debris on airport runway

作　　者：郑浪 罗天洪 王成琳 何泽银[1] 李忠涛 ZHENG Lang;LUO Tianhong;WANG Chenglin;HE Zeyin;LI Zhongtao(School of Mechanotronics&Vehicle Engineering,Chongqing Jiaotong University,Chongqing 400074,China;School of Intelligent Manufacturing Engineering,Chongqing University of Arts and Sciences,Chongqing 402160,China)

机构地区：[1]重庆交通大学机电与车辆工程学院,重庆400074 [2]重庆文理学院智能制造工程学院,重庆402160

出　　处：《现代电子技术》2020年第9期51-54,共4页Modern Electronics Technique

基　　金：重庆市研究生教育创新基金项目资助(2018S0140,CYS18223)。

摘　　要：检测机场跑道异物对飞机安全起飞着陆具有重大的意义。为此,文中提出一种适用于机场跑道异物检测的改进区域生长算法。提出的算法在对机场跑道图像进行区域种子点选择的基础上,以其为中心,进行8连通区域生长;采用Sobel算法对区域生长后的图像进行边缘增强;以颜色相似、区域相邻及吞噬小面积区域三要素为准则,融合特征相似区域,从而得到目标分割图像。利用所提算法对所测试异物的识别率分别达到91.7%和93.8%,表明提出的算法能准确地分割机场跑道异物区域,并能有效地抑制背景干扰,成功地解决了传统区域生长算法产生的过分割问题。The detection of foreign object debris(FOD)on airport runway is of great significance for aircrafts to take off and land safely. For this reason,an improved region growth algorithm applicable to FOD detection on airport runway is proposed. In the algorithm,the regional seed point of the airport runway image is selected and taken as the center to perform 8-connected region growth. The Sobel algorithm is adopted to enhance the image edge after the image is subjected to region growth.Regions with similar features are fused on the basis of the criteria of similar colors,adjacent regions and swallowing small regions to obtain the target segmentation image. The recognition rates of FOD obtained with the proposed algorithm are 91.7%and 93.8% respectively,which shows that the algorithm can accurately segment the FOD region of airport runway and effectively suppress background interference. Therefore,it successfully deals with over-segmentation occurred in the application of the traditional region growth algorithm.

关 键 词：改进区域生长算法 异物检测 机场跑道 区域合并 SOBEL算法 图像分割 

分 类 号：TN911.1-34[电子电信—通信与信息系统] TP242.62[电子电信—信息与通信工程]