基于改进FasterR-CNN的机场跑道道面裂缝检测方法

打开文本图片集

中图分类号：TP391.92;U418.6 文献标志码：A文章编号：1001-5922（2025）05-0159-04

Abstract：Civil aviation plays a vital rolein China's transportationsystem.With theextension of theservice life of airports，the problem of pavement damage is becoming more and more serious，which poses a major threat to the safetyof aircraft taxing，take-offand landing.Inorder to reduce theriskofaircraft inthe processof take-off and landing，an improved detection method based on Faster R-CNN was proposed.The detection method comprehensivelyused deep learningand objectdetection technologiessuch as GC-ASFF module，CIoU index，improved loss functionand transfer learning to achieve accurate detectionof pavement cracks，so as to evaluate thecurent pavement safety status by using the identified characteristic parameters of pavement cracks.The experimental results showedthat the improved model had high recognition accuracyand excelent comprehensive performance，and can accurately identify and detect runway pavement damage，which has high reliability.

Key Words：crack detection；faster R-CNN；ASFF ；merging ratio；loss function

机场跑道是飞机起降的主要区域，在荷载和环境因素的相互作用下，机场道面很容易出现各种病害，裂缝病害则是大多数问题的早期表现之一，若不及时进行处理，则会造成巨大的安全隐患[1]。（剩余7028字）