一种烟草病虫害混合专家检测变换模型

打开文本图片集

中图分类号：TP183；TP391.41 文献标志码：A 文章编号： 1000-5013（2026）02-0175-08

Abstract：To solve the challnge of multi-scale object detection in tobacco pest and disease identification under complexfield suroundings，a mixture-of-expert detection transformer model is proposed.The model integrates three heterogeneous expert modules according to the diffrent feature processing requirements of tobacco pest and disease images. At the shallow layer，a cross-stage detail enhancement module is used to strengthen local details and edge features;at the middle layer，an adaptive balancing module is utilized to ntelligently balanceslocal and contextual information;at the deep layer，a cross-stage eficient state-space module is designed to effectively model long-range dependencies，thereby achieving hierarchical and dynamic extraction of multi-scale pest and disease features.Experimental results show that compared with the baseline model RTDETR-R18，the proposed mixture-of-experts detection transformer model reduces parameters and floating-point operations by 27.4% and 16.3% ，respectively，while improving mAP@0. 50 from 0.770 to 0.815.

Keywords：multi-scale object detection; detection transformer; mixture-of-expert;smallobject detection;ag riculture computer vision; tobacco pest and disease

烟草是我国的重要经济作物，烟草产量约占世界总量 35% ，卷烟产量约占世界总量 32%[1] 。（剩余11654字）