电子学报2025，Vol.53Issue(6)：1755-1770,16.DOI:10.12263/DZXB.20250186
DFRNet:融合扩散-聚焦物理机制的语义分割模型研究
DFRNet:A Semantic Segmentation Method Inspired with Physical Mechanism of Diffusion-Focus
摘要
Abstract
To address the information loss induced by downsampling in image semantic segmentation tasks,as well as the widespread limitations of existing upsampling methods:such as inadequate global perception,blurred fine-grained re-construction,unstable generation processes,and redundant information handling in various scenarios,this paper proposes a lightweight semantic segmentation model,DFRNet,which incorporates a physics-inspired diffusion-focusing mechanism.Specifically,inspired by the surface tension of liquids,the model introduces a diffusion-focusing mechanism and designs a dynamic context window selection(DWS)module to optimize information flow,thereby implementing the physics-inspired energy propagation upsampling(PIEPU)framework.PIEPU comprises three core modules:diffusion,focusing,and regula-tion.These modules collaboratively enhance global contextual propagation,critical region feature reinforcement,and opti-mized information flow,thereby significantly improving fine-grained perception and semantic consistency across complex scenarios.Extensive experiments conducted on 14 datasets covering 7 semantic categories demonstrate that DFRNet consis-tently achieves superior performance over state-of-the-art methods in terms of mean intersection over union(mIoU),F1 score,and Accuracy.Specifically,mIoU improvements range from 0.165％to 4.259％,F1 score gains span 0.140％to 2.888％,and Accuracy enhancements vary from 0.035％to 1.386％across diverse datasets.These results validate the robust-ness and generalization capability of the proposed approach.Notably,DFRNet has a model size of only 3.34 MB,making it suitable for lightweight real-time applications.关键词
语义分割/上采样/扩散-聚焦/全局上下文Key words
semantic segmentation/upsampling/diffusion focus/global context分类
信息技术与安全科学引用本文复制引用
黄依莎,姜林,管亚菲,张亚莎,梁欣,曾伟豪,方晓萍..DFRNet:融合扩散-聚焦物理机制的语义分割模型研究[J].电子学报,2025,53(6):1755-1770,16.基金项目
湘江实验室重大项目(No.23XJ01003,No.23XJ01009) （No.23XJ01003,No.23XJ01009）
湖南省教育厅科学研究重点项目(No.22A0441) Major Project of Xiangjiang Laboratory(No.23XJ01003,No.23XJ01009) （No.22A0441）
Key Scientific Re-search Project of Hunan Provincial Department of Education(No.22A0441) （No.22A0441）
