基于多传感器信息融合的机场道面裂缝检测算法  

Airport pavement crack detection algorithm based on multi⁃sensor information fusion

作　　者：李海丰[1] 聂晶晶 吴治龙 彭博[2] 桂仲成 LI Haifeng;NIE Jingjing;WU Zhilong;PENG Bo;GUI Zhongcheng(College of Computer Science and Technology,Civil Aviation University of China,Tianjin 300300,China;School of Traffic and Transportation,Chongqing Jiaotong University,Chongqing 400074,China;Chengdu Guimu Robot Co.,Ltd.,Chengdu 610101,China)

机构地区：[1]中国民航大学计算机科学与技术学院,天津300300 [2]重庆交通大学交通运输学院,重庆400074 [3]成都圭目机器人有限公司,四川成都610101

出　　处：《现代电子技术》2020年第24期17-21,25,共6页Modern Electronics Technique

基　　金：国家重点研发计划课题(2019YFB1310601);国家自然科学基金青年科学基金项目(61703064);重庆市技术创新与应用示范项目(cstc2018jscx⁃msybX0295)。

摘　　要：机场道面的裂缝检测只能在夜间停航期间进行,由此带来光照条件差、图像对比度低、噪声干扰强烈等问题,致使传统基于可见光图像的裂缝检测算法难以适用。为此,提出一种融合可见光图像和红外图像的裂缝检测算法。首先通过局部差分法检测不同传感器图像中的初始裂缝;然后,通过局部区域像素的灰度和温度概率分布建立决策级信息融合模型,获得候选裂缝,并对候选裂缝进行像素级融合;最后,利用多种数学形态学约束进行筛选,获得最终的裂缝检测结果。在真实机场道面数据集上进行了测试,并与多个算法完成对比,结果显示该文算法的准确率、召回率和F值均优于对比算法,可以较精确地检测出裂缝,为机场道面裂缝的检测与管理提供了技术基础。As the crack detection of airport pavement can only be carried out during the suspension of flight at night,which brings some problems such as poor illumination conditions,low image contrast and strong noise interference,which makes the traditional crack detection algorithm based on visible images difficult to be applied,a crack detection algorithm fusing visible image and infrared image is proposed.The local difference method is used to detect the initial cracks in the images obtained by different sensors.The decision⁃level information fusion model is established according to the gray scale and temperature probability distribution of pixels within local region to obtain the candidate cracks and then perform the pixel⁃level fusion of the candidate cracks.A variety of mathematical morphology constraints are used to screen them and obtain the final results of the crack detection.The testing on a dataset of real airport pavement is conducted,and the result is compared with that of other algorithms.The results show that the precision,recall rate and F value of the proposed algorithm are better than those of the compared algorithms,and the proposed algorithm can detect cracks more accurately,which provides a technical basis for the crack detection and management of airport pavement.

关 键 词：机场道面 裂缝检测 信息融合 候选裂缝筛选 像素融合 测试分析 

分 类 号：TN912.23-34[电子电信—通信与信息系统] TP212.9[电子电信—信息与通信工程]