基于改進YOL0v8n的輕量化工地堆放木材異常檢測算法

打開文本圖片集

中圖分類號：TP183：TP391 文獻標識碼：A 文章編號：2096-4706（2025）07-0058-07

Abstract：When the timber materialsare stackedontheconstruction site，theoutdoorenvironment is prone to abnormal problemssuchas moisture deformationand drycrackingonthe surfaceofthe timber.Aimingatthe problems ofpooraccuracy andhighcomputationalcomplexityoftheexisting detectionalgorithmsonthesurfaceof timber materials，alightweightsmall target detection algorithm（YOLO-ESN）basedonYOLOv8nis proposed.Thealgorithm introduces the SpatialandChannel ReconstructionConvolution （SCConv） module and the Normalized WassrsteinDistance （NWD）lossfunction forsmal target detection.Atthesametime，itembeds theEffcientMulti-ScaleAtention（EMA）modulebasedonCross-SpatialLeaminginto thebackbone network toreduce the impact ofoccusionandbackgroundinterference.Theimprovedalgorithmis experimentally verified on the timber defect dataset. Compared with the original algorithm，its is increased by 3 . 6 % ，and the parameter quantity is reduced by 23 . 3 % ，which realizes the real-time and accurate detection of the abnormal situation of stacked timbermaterials.

Keywords： improvedYOLOv8nalgorithm; constructionsite timber anomalydetection; lightweight; smalltargetdetection

0 引言

木材作為生物材料常年暴露在室外，易受人為操作不當、蟲蛀等因素影響而受損。（剩余8317字）