中国民航大学学报

• 工程技术 • 上一篇    下一篇

基于区域生长和最外边缘追踪的路面损坏检测

王维1,孟丽娜1,闫兆进2   

  1. (1.中国民航大学机场学院,天津300300;2.山东科技大学测绘科学与工程学院,山东青岛266590)
  • 收稿日期:2016-03-22 修回日期:2016-05-10 出版日期:2016-12-26 发布日期:2016-12-28
  • 作者简介:王维(1960—),男,河北丰南人,教授,硕士,研究方向为机场工程.
  • 基金资助:

    中央高校基本科研业务费专项(Y16-18)

Asphalt pavement damage detection method based on region growing and outermost border following

WANG Wei1, MENG Lina1, YAN Zhaojin2   

  1. (1. College of Airport Engineering, CAUC, Tianjin 300300, China; 2. College of Geomatics,Shandong University of Science and Technology, Qingdao 266590, China)
  • Received:2016-03-22 Revised:2016-05-10 Online:2016-12-26 Published:2016-12-28

摘要:

为提高沥青路面损坏检测的效率和准确度,提出一种基于区域生长和最外边缘追踪的沥青路面损坏检测方法。该方法先对路面图像进行区域生长分割,得到病害的连续区域,然后利用数学形态学中的闭运算进行处理,连接邻近物体和平滑边界,接着通过边界标记和获取最外边界唯一值得到最外边界的边界点坐标,最后通过坐标转换,生成相应的GIS 图层。通过路面数据验证,证明了该方法的有效性。

关键词: 道路工程, 路面破损, 图像识别, 区域生长, 边缘追踪

Abstract:

In order to improve the efficiency and accuracy of asphalt pavement damage detection, a damage detection method is proposed to the asphalt pavement based on region growing and border following of outermost borders. Firstly,road image is divided into different regions according to region growing to get the continuous area of diseases.Secondly, adjacent objects and smooth border are connected by closing operation of mathematical morphology, and then border point coordinates of outermost border are obtained by marking the border and getting unique value of outermost border. Finally, the appropriate GIS layers are generated through coordinate conversion, proving the effectiveness of this method through road data testing.

Key words: road engineering, pavement damage, image recognition, regional growth, border following

中图分类号: