华中农业大学学报2026，Vol.45Issue(3)：87-97,11.DOI:10.13300/j.cnki.hnlkxb.2026.03.008
基于改进RT-DETR的苹果病害检测算法
Apple disease detection algorithm based on improved RT-DETR
摘要
Abstract
To address the challenges of strong background interference,difficulty in recognizing multi-scale lesions,and the need for lightweight deployment in apple disease detection within complex orchard en-vironments,an improved lightweight apple disease detection model named EGA-DETR was proposed based on RT-DETR.The model was enhanced in three key aspects.First,an RGLAN lightweight feature aggregation module was designed to reduce redundant computations and enhance feature representation through feature splitting and re-parameterized convolution.Second,a BiFPN-GLSA multi-scale feature fu-sion module was developed to improve the representation of lesions at various scales via bidirectional fea-ture transmission and a global-local self-attention mechanism.Third,an Inner-Shape-IoU loss function was introduced to improve model's localization accuracy for irregularly shaped lesion targets.Experimental results showed that EGA-DETR achieved 91.8%precision,88.9%recall,and 92.1%mAP50 on the apple disease dataset,with improvements of 3.5,3.5,and 1.5 percentage points over the baseline RT-DETR-18,respectively.Meanwhile,the number of model parameters was reduced to 11.8×106,which is 40.4%fewer than that of the baseline model,and the inference speed reached 120 frames per second.In summary,the EGA-DETR model achieved a favorable balance between detection accuracy and computational efficiency,providing robust technical support for accurate and real-time apple disease detection.关键词
苹果病害检测/RT-DETR/轻量化特征聚合/双向特征金字塔网络/全局-局部自注意力/Inner-Shape-IoUKey words
apple disease detection/RT-DETR/lightweight feature aggregation/bidirectional fea-ture pyramid network/global-local self-attention/Inner-Shape-IoU分类
信息技术与安全科学引用本文复制引用
肖永帅,陈若冰,时雷,郑光,尹飞..基于改进RT-DETR的苹果病害检测算法[J].华中农业大学学报,2026,45(3):87-97,11.基金项目
河南省科技攻关项目(242102521027) （242102521027）
河南省科技研发计划联合基金项目(222301420113) （222301420113）
