基于改进Faster R-CNN的机场跑道道面裂缝检测方法  

A crack detection method for airport runway pavement based on improved faster R-CNN

作　　者：张璐[1] 高培伟 张芊伊 李国庆 ZHANG Lu;GAO Peiwei;ZHANG Qianyi;LI Guoqing(College of Civil Aviation,Nanjing University of Aeronautics and Astronautics,Nanjing 211106,China)

机构地区：[1]南京航空航天大学民航学院,江苏南京211106

出　　处：《粘接》2025年第5期159-162,共4页Adhesion

基　　金：江苏省自然科学基金项目(项目编号:BK20200429)。

摘　　要：民航运输在中国交通体系中占据着至关重要的地位。随着机场使用年限的延长,道面损伤问题日益严重,这对飞机滑行与起降的安全构成了重大威胁。为了降低飞机在起降过程的风险性,提出了一种基于Faster R-CNN的改进检测方法。该检测方法综合利用了GC-ASFF模块、CIoU指标、改进损失函数和迁移学习等深度学习和目标检测技术,实现了对道面裂缝的准确检测,以便利用识别到的道面裂缝特征参数来对当前道面安全状况进行评估。试验结果表明,改进后的模型识别精度较高,综合性能较优,对于飞机跑道道面损伤能够进行精准识别和检测,具有较高的可靠性。Civil aviation plays a vital role in China's transportation system.With the extension of the service life of airports,the problem of pavement damage is becoming more and more serious,which poses a major threat to the safety of aircraft taxiing,take-off and landing.In order to reduce the risk of aircraft in the process of take-off and landing,an improved detection method based on Faster R-CNN was proposed.The detection method comprehensively used deep learning and object detection technologies such as GC-ASFF module,CIoU index,improved loss function and transfer learning to achieve accurate detection of pavement cracks,so as to evaluate the current pavement safety status by using the identified characteristic parameters of pavement cracks.The experimental results showed that the improved model had high recognition accuracy and excellent comprehensive performance,and can accurately identify and detect runway pavement damage,which has high reliability.

关 键 词：裂缝检测 Faster R-CNN ASFF 交并比 损失函数 

分 类 号：TP391.92[自动化与计算机技术—计算机应用技术] U418.6[自动化与计算机技术—计算机科学与技术]