首页|期刊导航|Computer Modeling in Engineering & Sciences|High-Performance Segmentation of Power Lines in Aerial Images Using a Wavelet-Guided Hybrid Transformer Network

High-Performance Segmentation of Power Lines in Aerial Images Using a Wavelet-Guided Hybrid Transformer NetworkOA

中文摘要

Inspections of power transmission lines(PTLs)conducted using unmanned aerial vehicles(UAVs)are complicated by the fine structure of the lines and complex backgrounds,making accurate and efficient segmentation challenging.This study presents the Wavelet-Guided Transformer U-Net(WGT-UNet)model,a new hybrid net-work that combines Convolutional Neural Networks(CNNs),Discrete Wavelet Transform(DWT),and Transformer architectures.The model’s primary contribution is based on spatial and channel attention mechanisms derived from wavelet subbands to guide the Transformer’s self-attention structure.Thus,low and high frequency components are separated at each stage using DWT,suppressing structural noise and making linear objects more prominent.The developed design is supported by multi-component hybrid cost functions that simultaneously solve class imbalance,edge sharpness,structural integrity,and spatial regularity issues.Furthermore,high segmentation success has been achieved in producing sharp boundaries and continuous line structures with the DWT-guided attention mechanism.Experiments conducted on the TTPLA dataset reveal that the version using the ConvNeXt backbone outperforms the current state-of-the-art approaches with an F1-Score of 79.33%and an Intersection over Union(IoU)value of 68.38%.The models and visual outputs of the developed method and all compared models can be accessed at https://github.com/burhanbarakli/WGT-UNET.

Burhan Baraklı;Ahmet Küçüker

Department of Electrical and Electronics Engineering,Sakarya University,Sakarya,TürkiyeDepartment of Electrical and Electronics Engineering,Sakarya University,Sakarya,Türkiye

信息技术与安全科学

Salient object detectionsuperpixel segmentationtransformersattention mechanismmulti-level fusionedge-preserving refinementmodel-driven

《Computer Modeling in Engineering & Sciences》 2026 (2)

P.772-802,31

10.32604/cmes.2026.077872

评论