{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,20]],"date-time":"2026-08-20T15:22:33Z","timestamp":1787239353817,"version":"build-2736575974"},"reference-count":39,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,4,10]],"date-time":"2026-04-10T00:00:00Z","timestamp":1775779200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Plant Sci."],"abstract":"<jats:sec>\n                    <jats:title>Introduction<\/jats:title>\n                    <jats:p>Accurate plant disease identification is of great importance for ensuring agricultural productivity and food security. However, complex illumination variations, leaf occlusion, and diverse disease spot scales throughout plant growth stages significantly increase the difficulty of real-time detection, leading to limited accuracy and robustness in existing approaches.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>To address these challenges, we propose an improved RT-DETRv2-based plant disease detection model, termed AF-RT-DETR. A Bidirectional Cross Gate (BCG) module is introduced in the feature extraction stage to reduce channel redundancy and enhance discriminative feature representation through multi-level feature interactions. The original RepVGG structure is replaced with a Dynamic Channel Shift (DCS) module, effectively enlarging the receptive field and strengthening contextual feature fusion without additional computational overhead. Additionally, an improved Scale-aware Multi-level Loss (SML) emphasizes low-quality feature maps to improve detector robustness.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>The model achieves mAP50 and mAP50:95 of 93.6% and 67.2% on the Plant-Disease dataset, surpassing the baseline by 5.1% and 4.5%. Furthermore, the model was evaluated on multiple crops and growth stages under diverse field conditions, demonstrating robust performance and adaptability.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Discussion<\/jats:title>\n                    <jats:p>These results indicate that AF-RT-DETR effectively enables real-time plant disease detection in complex field environments.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.3389\/fpls.2026.1788655","type":"journal-article","created":{"date-parts":[[2026,4,10]],"date-time":"2026-04-10T06:04:11Z","timestamp":1775801051000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":3,"title":["AF-RT-DETR: Adaptive cross-scale feature interaction for real-time plant disease detection in complex field environments"],"prefix":"10.3389","volume":"17","author":[{"given":"Ming","family":"Liu","sequence":"first","affiliation":[{"name":"School of Mathematics and Computer Science, Wuhan Polytechnic University","place":["Wuhan, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jiangrong","family":"Liu","sequence":"additional","affiliation":[{"name":"School of Mathematics and Computer Science, Wuhan Polytechnic University","place":["Wuhan, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ziqi","family":"Mao","sequence":"additional","affiliation":[{"name":"School of Computer and Artificial Intelligence, Wuhan Textile University","place":["Wuhan, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Shuzhe","family":"Cheng","sequence":"additional","affiliation":[{"name":"China Railway 11th Bureau Group Xi\u2019an Construction Co., Ltd.","place":["Xi\u2019an, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xinyang","family":"Li","sequence":"additional","affiliation":[{"name":"School of Mathematics and Computer Science, Wuhan Polytechnic University","place":["Wuhan, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Shiyu","family":"Yan","sequence":"additional","affiliation":[{"name":"School of Mathematics and Computer Science, Wuhan Polytechnic University","place":["Wuhan, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2026,4,10]]},"reference":[{"key":"B1","doi-asserted-by":"crossref","first-page":"213","DOI":"10.1007\/978-3-030-58452-8_13","article-title":"End-to-end object detection with transformers","volume-title":"Computer vision \u2013 ECCV 2020","author":"Carion","year":"2020"},{"key":"B2","doi-asserted-by":"publisher","DOI":"10.3389\/fpls.2022.1034600","article-title":"Multi-peril pathogen risks to global wheat production: A probabilistic loss and investment assessment","volume":"13","author":"Chai","year":"2022","journal-title":"Front. Plant Sci."},{"key":"B3","doi-asserted-by":"publisher","first-page":"17","DOI":"10.1007\/s44297-024-00037-w","article-title":"Enhancing Agrobacterium-mediated soybean transformation efficiency with an auxiliary solution","volume":"2","author":"Chen","year":"2024","journal-title":"Crop Health"},{"key":"B4","doi-asserted-by":"publisher","first-page":"326","DOI":"10.1111\/ppa.13794","article-title":"Development of a molecular diagnostic to discriminate between Fomitiporia species and advancements in detection of the main grapevine-decline-related pathogens","volume":"73","author":"Christopoulou","year":"2024","journal-title":"Plant Pathol."},{"key":"B5","first-page":"11963","article-title":"Scaling up your kernels to 31\u00d731","author":"Ding","year":"2022"},{"key":"B6","first-page":"15162","article-title":"DEIM: DETR with improved matching for fast convergence","author":"Huang","year":"2025"},{"key":"B7","doi-asserted-by":"publisher","DOI":"10.1016\/j.jii.2025.100843","article-title":"Toward agri-food supply chain viability under pest spread risk","volume":"45","author":"Kalantari Khalil Abad","year":"2025","journal-title":"J. Ind. Inf. Integr."},{"key":"B8","doi-asserted-by":"publisher","DOI":"10.3390\/rs15235615","article-title":"Weed\u2013crop segmentation in drone images with a novel encoder\u2013decoder framework enhanced via attention modules","volume":"15","author":"Khan","year":"2023","journal-title":"Remote Sens."},{"key":"B9","doi-asserted-by":"publisher","first-page":"20607","DOI":"10.1007\/s11042-024-19860-y","article-title":"Classification of plant diseases in images using dense-inception architecture with attention modules","volume":"84","author":"Khan","year":"2025","journal-title":"Multimed. Tools Appl."},{"key":"B10","article-title":"YOLOv11: An overview of the key architectural enhancements","volume-title":"arXiv","author":"Khanam","year":"2024"},{"key":"B11","first-page":"15189","article-title":"Lite DETR: An interleaved multi-scale encoder for efficient DETR","author":"Li","year":"2023"},{"key":"B12","first-page":"25281","article-title":"Shift-ConvNets: Small convolutional kernel with large kernel effects","author":"Li","year":"2025"},{"key":"B13","first-page":"13609","article-title":"DN-DETR: Accelerate DETR training by introducing query denoising","author":"Li","year":"2022"},{"key":"B14","doi-asserted-by":"publisher","DOI":"10.1016\/j.compag.2024.108788","article-title":"A DETR-like detector-based semi-supervised object detection method for Brassica chinensis growth monitoring","volume":"219","author":"Li","year":"2024","journal-title":"Comput. Electron. Agric."},{"key":"B15","doi-asserted-by":"publisher","DOI":"10.1016\/j.compag.2022.107345","article-title":"An improved YOLOv5-based vegetable disease detection method","volume":"202","author":"Li","year":"2022","journal-title":"Comput. Electron. Agric."},{"key":"B16","first-page":"2999","article-title":"Focal loss for dense object detection","author":"Lin","year":"2017"},{"key":"B17","article-title":"More convNets in the 2020s: scaling up kernels beyond 51\u00d751 using sparsity","author":"Liu","year":"2023"},{"key":"B18","first-page":"25","article-title":"DAB-DETR: Dynamic anchor boxes are better queries for DETR","author":"Liu","year":"2022"},{"key":"B19","first-page":"191","article-title":"Bitter gourd leaf disease detection based on Faster R-CNN","volume-title":"Advances in intelligent systems and computing","author":"Liu","year":"2021"},{"key":"B20","first-page":"3651","article-title":"Conditional DETR for fast training convergence","author":"Meng","year":"2021"},{"key":"B21","doi-asserted-by":"publisher","DOI":"10.3390\/agriengineering7110375","article-title":"Benchmarking YOLO models for crop growth and weed detection in cotton fields","volume":"7","author":"Raza","year":"2025","journal-title":"AgriEngineering"},{"key":"B22","doi-asserted-by":"publisher","DOI":"10.1016\/j.mib.2022.102233","article-title":"Climate change and plant pathogens","volume":"70","author":"Raza","year":"2022","journal-title":"Curr. Opin. Microbiol."},{"key":"B23","doi-asserted-by":"publisher","first-page":"1137","DOI":"10.1109\/TPAMI.2016.2577031","article-title":"Faster R-CNN: Towards real-time object detection with region proposal networks","volume":"39","author":"Ren","year":"2017","journal-title":"IEEE Trans. Pattern Anal. Mach. Intell."},{"key":"B24","unstructured":"Detecting rice crop diseases object detection dataset"},{"key":"B25","unstructured":"Disease detection object detection dataset"},{"key":"B26","unstructured":"Plant Disease v2 512\u00d7512"},{"key":"B27","doi-asserted-by":"publisher","first-page":"229","DOI":"10.1016\/j.aiia.2020.10.002","article-title":"A review of imaging techniques for plant disease detection","volume":"4","author":"Singh","year":"2020","journal-title":"Artif. Intell. Agric."},{"key":"B28","doi-asserted-by":"publisher","DOI":"10.1007\/s44297-023-00001-0","article-title":"MoScd2 is involved in appressorium formation and pathogenicity via the Pmk1 MAPK pathway in Magnaporthe oryzae","volume":"1","author":"Sun","year":"2023","journal-title":"Crop Health"},{"key":"B29","article-title":"YOLOv12: Attention-centric real-time object detectors","volume-title":"arXiv","author":"Tian","year":"2025"},{"key":"B30","unstructured":"Ultralytics YOLOv8"},{"key":"B31","article-title":"Attention is all you need","volume-title":"Advances in neural information processing systems","author":"Vaswani","year":"2017"},{"key":"B32","doi-asserted-by":"publisher","first-page":"20160451","DOI":"10.1098\/rsif.2016.0451","article-title":"Trade-off between disease resistance and crop yield: A landscape-scale mathematical modelling perspective","volume":"13","author":"Vyska","year":"2016","journal-title":"J. R. Soc Interface"},{"key":"B33","first-page":"107984","article-title":"YOLOv10: real-time end-to-end object detection","volume-title":"Advances in neural information processing systems 37","author":"Wang","year":"2024"},{"key":"B34","first-page":"1","article-title":"YOLOv9: Learning what you want to learn using programmable gradient information","volume-title":"Computer vision \u2013 ECCV 2024","author":"Wang","year":"2024"},{"key":"B35","first-page":"2423","article-title":"When shift operation meets vision transformer: An Extremely Simple Alternative to Attention Mechanism","author":"Wang","year":"2022"},{"key":"B36","doi-asserted-by":"publisher","DOI":"10.1016\/j.compag.2022.106921","article-title":"Early diagnosis and pathogenesis monitoring of wheat powdery mildew using hyperspectral imaging","volume":"197","author":"Xuan","year":"2022","journal-title":"Comput. Electron. Agric."},{"key":"B37","first-page":"1","article-title":"DINO: DETR with improved denoising anchor boxes","author":"Zhang","year":"2023"},{"key":"B38","first-page":"8514","article-title":"VarifocalNet: An IoU-aware dense object detector","author":"Zhang","year":"2021"},{"key":"B39","first-page":"3","article-title":"Deformable DETR: Deformable transformers for end-to-end object detection","author":"Zhu","year":"2021"}],"container-title":["Frontiers in Plant Science"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fpls.2026.1788655\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,4,10]],"date-time":"2026-04-10T06:04:18Z","timestamp":1775801058000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fpls.2026.1788655\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,4,10]]},"references-count":39,"alternative-id":["10.3389\/fpls.2026.1788655"],"URL":"https:\/\/doi.org\/10.3389\/fpls.2026.1788655","relation":{},"ISSN":["1664-462X"],"issn-type":[{"value":"1664-462X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,4,10]]},"article-number":"1788655"}}