2026-10-12T14:37:22Z https://api.figshare.com/v2/oai

oai:figshare.com:article/33816294 2026-09-15T18:15:43Z category_21 category_39 category_734 category_931 portal_5 item_type_3 month_year_09_2026

<p>Comparison with representative modules.</p> Shenming Qu (11442020) Zhiheng Liu (341545) Huazhen Zhao (24951048) Yichao Wu (259129) Yang Yang (45629) Biotechnology Ecology Biological Sciences not elsewhere classified Information Systems not elsewhere classified proposed method achieves minimal computational overhead major staple crop enhanced feature representation enabling effective suppression diverse morphological characteristics complex field backgrounds pest infestations throughout model &# 8217 directional pest textures module adaptively fuses lightweight convolutional design ipmaize dataset demonstrate introduces global self level feature representation thereby improving low improved yolov11n framework attention weighted aggregation tomato pest based model module enhances global significance dataset evaluations level transformer deeper level xlink "> scale features reduce yield random seeds often leading missed detections ip102 datasets highly vulnerable growth cycle experimental results background noise averaged map 63 %. <div><p>As a major staple crop with global significance, maize is highly vulnerable to pest infestations throughout its growth cycle, which can substantially limit plant development and reduce yield. In practice, accurate detection remains challenging due to pronounced morphological variations across pest life stages, as well as interference from complex field backgrounds, often leading to missed detections. To address these challenges, we propose WSD-YOLO (YOLOv11n with WindmillConv, Single-Scale High-Level Transformer, and Dual-Attention Weighted Aggregation), a high-accuracy maize pest detection model built upon an improved YOLOv11n framework. The WindmillConv (WMConv) module enhances the model’s sensitivity to multi-directional pest textures, thereby improving low-level feature representation while maintaining a lightweight convolutional design. At a deeper level, the Single-Scale High-Level Transformer (SHLT) introduces global self-attention, enabling effective suppression of background noise with minimal computational overhead. In addition, the Dual-Attention Weighted Aggregation (DAWA) module adaptively fuses same-scale features, improving the detection of pests with diverse morphological characteristics. Experimental results from three repeated training runs (random seeds: 42, 2026, 3407) on the IPMaize dataset demonstrate that the proposed method achieves an averaged mAP@0.5 of 78.77% (79.7% with seed = 42) and averaged mAP@0.5:0.95 of 52.63%. Cross-dataset evaluations on the Tomato Pest&Diseases and IP102 datasets further confirm the model’s strong generalization capability.</p></div> 2026-09-15T18:15:12Z Dataset Dataset 10.1371/journal.pone.0357034.t011 https://figshare.com/articles/dataset/_p_Comparison_with_representative_modules_p_/33816294 CC BY 4.0