CEA-RT-DETR:一种军事迷彩伪装目标检测算法OA

Military camouflage target detection is a challenging task due to the high similarity between targets and complex backgrounds.To address the issues of low detection precision and poor real-time performance,this paper proposes an improved algorithm named CEA-RT-DETR.First,a cross-level edge feature integration backbone network is designed to focus on edge regions by incorporating a gradient edge information extractor and a progressive feature aggregation module.Second,an enhanced spatial feature pyramid network is adopted in the neck layer,which utilizes enhanced linear attention and semantic feature fusion to preserve multi-scale features.Finally,an additive-weighted feature mixer attention mechanism is implemented in the intra-scale feature interaction layer to capture texture and structural details.Experimental results demonstrate that the improved model significantly enhances detection performance.