光学精密工程2025，Vol.33Issue(22)：3564-3576,13.DOI:10.37188/OPE.20253322.3564
基于Transformer的边缘检测网络
Edge detection network based on transformer
摘要
Abstract
The current mainstream edge detection method based on convolutional neural network has limi-tations in receptive field range and fine-grained edge perception.With the development of Vision Trans-former,its global modeling ability and flexible information interaction mechanism bring new possibilities for edge detection tasks.To solve this issue,this paper proposed an encoder-decoder model named TFEdge,which combined Transformer,Multi-Level Aggregation Feature Pyramid(MLAFP),and Multi-Scale Attention Aggregation(MSAA)modules for high-precision edge detection.The model intro-duced the Dilated Neighborhood Attention Transformer as the backbone network and extracted the global context information and local edge clues of the image through a multi-stage cascade design.Simultaneous-ly,the Multi-Level Aggregation Feature Pyramid was designed to aggregate the deep and shallow features of each stage,endowing the shallow features with more abundant semantic features to suppress image noise and improve the detection ability of indistinct boundaries.Finally,the Multi-Scale Attention Aggre-gation module,based on an attention mechanism,was proposed to further enhance feature representation by aggregating the cross-scale spatial and channel attention information of feature maps.The experiment is evaluated on the BSDS500 and NYUDv2 datasets.The ODS and OIS F-scores of TFEdge on the BS-DS500 are 0.857 and 0.874,respectively,while on the NYUDv2 they are 0.788 and 0.801,respective-ly.Compared with many existing methods,TFEdge shows superior edge detection performance in both quantitative and qualitative results.关键词
边缘检测/Transformer/注意力机制/多级聚合特征金字塔/多尺度注意力增强Key words
edge detection/transformer/attention mechanism/multi-level aggregation feature pyramid/multi-scale attention aggregation分类
信息技术与安全科学引用本文复制引用
LIN Jianpu,LI Xianguang,LIN Shanling,LÜ Shanhong,LIN Zhixian..基于Transformer的边缘检测网络[J].光学精密工程,2025,33(22):3564-3576,13.基金项目
国家重点研发计划资助(No.2023YFB3609400) （No.2023YFB3609400）
