基于改進YOLOv5s的SAR影像鐵道檢測技術研究

打開文本圖片集

中圖分類號：TP391 文獻標志碼：A 文章編號：2095-2945（2025）13-0050-05

Abstract：Gaofen-SARsatelitesforrailroadinspectionhavetheadvantagesofwidecoverage，all-weatherandmetal sensitivity，butneedtosolvetheproblemofinstantdetectionofrlwaytargets.Forthisreasonanimprovedmodelcalled Lightweight-YOLOv5sisproposedbasedonYOLOv5s，whichisespeciallysuitablefordetectionofrailroadtargetsinSARimage. ByreducingtheMobile-Darknetbackbonefeatureextractionnetwork layers，weoptimizedthenetworkstructure；byaddingHDC andCBMmechanisms，weadjustedthesmalltargetsensoryfieldweightsandstrengthenthesmalltargetlinefacilityfeature extraction;byusing FPGMpruning，weeliminatedtheredundantfeaturemodulesandachievealightweightmodel;byusing VarifocalLossasthelossfunction，weequalizedthepositiveandnegativecategoriesandhighlightthecontributionofpositive examples.The results show that，the accuracy of Lightweight-YOLOv5s model achieves 9 7 . 6 % ，and the inference time reduces to 6.87ms.Comparedwiththeclasicalalgorithmsfordetectinglineartargetsinremotesensingimages，theperformanceisgreatly improved for instant detection of railroad targets.

Keywords： SAR; railroad inspection; rail target detection; YOLOv5s; lightweight network

鐵道線路覆蓋范圍廣、距離長，所處地區(qū)地形、地貌條件復雜多樣，尤其是氣候及地質(zhì)條件復雜的山區(qū)，地震、洪水、泥石流等自然災害易發(fā)，會嚴重威脅鐵路系統(tǒng)安全穩(wěn)定運行，需要即時準確地監(jiān)測鐵路所處環(huán)境情況，才能有效保障鐵路安全。（剩余5705字）