该论文来源于网络,本站转载的论文均是优质论文,供学****和研究使用,文中立场与本网站无关,版权和著作权归原作者所有,如有不愿意被转载的情况,请通知我们删除已转载的信息,如果需要分享,请保留本段说明。 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。 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 中图分类号: ?34; 文献标识码: A 文章编号: 1004?373X(2020)24?0017?05 Airport pavement crack detection algorithm based on multi?sensor information fusion LI Haifeng1, NIE Jingjing1, WU Zhilong1, PENG Bo2, GUI Zhongcheng3 (1. College of Computer Science and Technology, Civil Aviation University of China, Tianjin 300300, China; 2. School of Traffic and Transportation, Chongqing Jiaotong University, Chongqing 400074, China; 3. Chengdu Guimu Robot Co., Ltd., Chengdu 610101, China) Abstract: 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