基于多尺度反馈聚合增强的YOLOv12植物病害检测框架

湖北农业科学 ›› 2026, Vol. 65 ›› Issue (7): 194-200.doi: 10.14088/j.cnki.issn0439-8114.2026.07.030

• 农业工程 • 上一篇下一篇

基于多尺度反馈聚合增强的YOLOv12植物病害检测框架

陈凌, 甘露, 陈嘉颖, 陈剑峰, 戴曲顺

福建林业职业技术学院,福建南平 353000

收稿日期:2026-04-14 出版日期:2026-07-25 发布日期:2026-07-23
通讯作者: 戴曲顺（1991-）,男,福建漳州人,讲师,硕士,主要从事植物生理与智慧农业研究工作,（电子信箱）daiqushun@fjlzy.com。
作者简介:陈凌（1990-）,女,福建南平人,助教,硕士,主要从事作物病害识别与智慧农业研究工作,（电子信箱）2023085@fjlzy.com。
基金资助:
南平市自然科学基金联合项目（N2025J014）; 福建省中青年教师教育科研项目（JZ230078）; 福建林业职业技术学院高层次人才课题（2023BK13）; 福建林业职业技术学院教研课题（YBJY202513）

YOLOv12-based framework for plant disease detection enhanced by multi-scale feedback aggregation

CHEN Ling, GAN Lu, CHEN Jia-ying, CHEN Jian-feng, DAI Qu-shun

Fujian Forestry Vocational Technical College, Nanping 353000, Fujian, China

Received:2026-04-14 Published:2026-07-25 Online:2026-07-23

摘要/Abstract

摘要： 针对植物病害检测中存在的小目标识别困难、特征融合不足及复杂背景干扰等问题,提出一种基于改进YOLOv12的多尺度反馈聚合检测模型。该模型在包含30类植物叶片病害图像的PlantDoc数据集上进行验证,旨在提升复杂环境下病斑区域的检测精度与模型的鲁棒性。通过在YOLOv12框架中引入MEMBA-F模块,利用多分支卷积与双向信息反馈机制实现跨层特征自适应融合,强化病斑边缘和细粒度特征的表达。结果表明,改进模型在PlantDoc数据集上Precision、Recall、mAP₅₀和mAP_50-95分别达到0.573、0.372、0.582和0.365,较YOLOv8分别提升7.7%、1.4%、3.7%和7.7%,同时保持27.4 ms的推理速度,具备实时检测能力。与基准YOLOv12模型相比,改进后模型的Precision、Recall、mAP₅₀和mAP_50-95等指标均有进一步提升,验证了MEMBA-F模块在多尺度特征聚合、跨层信息交互以及小目标与细粒度表征方面的有效性。改进模型平衡了检测精度与效率,可为智慧农业中植物病害识别提供参考。

关键词: 植物病害检测, YOLOv12, 多尺度特征融合, MEMBA-F模块

Abstract: To address the problems of difficult small-object recognition, insufficient feature fusion, and complex background interference in plant disease detection, a multi-scale feedback aggregation detection model based on improved YOLOv12 was proposed. The model was validated on the PlantDoc dataset containing images of 30 categories of plant leaf diseases to improve the detection accuracy and robustness of lesion regions in complex environments. By introducing the Multi-Scale Enhanced Multi-Branch Aggregation with Feedback（MEMBA-F） module into the YOLOv12 framework, the improved model employed multi-branch convolution and a bidirectional information feedback mechanism to achieve adaptive cross-layer feature fusion, thereby enhancing the representation of lesion edges and fine-grained features. The results showed that the Precision, Recall, mAP50, and mAP50-95 of the improved model on the PlantDoc dataset reached 0.573, 0.372, 0.582, and 0.365, respectively, which were 7.7%, 1.4%, 3.7%, and 7.7% higher than those of YOLOv8, while maintaining an inference speed of 27.4 ms, demonstrating real-time detection capability. Compared with the baseline YOLOv12 model, the metrics of the improved model were further enhanced, verifying the effectiveness of the MEMBA-F module in multi-scale feature aggregation, cross-layer information interaction, and the representation of small objects and fine-grained features. The improved model balanced detection accuracy and efficiency, providing a reference for plant disease recognition in smart agriculture.

Key words: plant disease detection, YOLOv12, multi-scale feature fusion, MEMBA-F module

中图分类号:

TP391
S43

陈凌, 甘露, 陈嘉颖, 陈剑峰, 戴曲顺. 基于多尺度反馈聚合增强的YOLOv12植物病害检测框架[J]. 湖北农业科学, 2026, 65(7): 194-200.

CHEN Ling, GAN Lu, CHEN Jia-ying, CHEN Jian-feng, DAI Qu-shun. YOLOv12-based framework for plant disease detection enhanced by multi-scale feedback aggregation[J]. HUBEI AGRICULTURAL SCIENCES, 2026, 65(7): 194-200.

参考文献

[1] 付艳君,郭敏.基于YOLOv8n改进的植物病害检测算法[J].计算机时代,2025(9):1-6.
[2] 张和群,蔺素银,初禹彤,等.基于改进YOLOv8n的苹果叶片病害检测[J].中国农机化学报,2026,47(4):147-154,170.
[3] 薛霞,刘鹏,周文.基于改进YOLOv8算法的实时细粒度植物病害检测[J].中国农机化学报,2024,45(5):188-194.
[4] 江侯涛,马善农.基于半监督的植物病害智能检测研究[J].机电工程技术,2024,53(2):221-224,290.
[5] JAFAR A, BIBI N, ALI NAQVI R, et al.Revolutionizing agriculture with artificial intelligence: Plant disease detection methods, applications, and their limitations[J]. Frontiers in plant science, 2024, 15: 1356260.
[6] UPADHYAY A, CHANDEL N S, SINGH K P, et al.Deep learning and computer vision in plant disease detection: A comprehensive review of techniques, models, and trends in precision agriculture[J]. Artificial intelligence review, 2025, 58(3): 92.
[7] DEMILIE W B.Plant disease detection and classification techniques: A comparative study of the performances[J]. Journal of big data, 2024, 11(1): 5.
[8] DEY P, MAHMUD T, NAHAR S R, et al.Plant disease detection in precision agriculture: Deep learning approaches[A].2024 2nd International Conference on Intelligent Data Communication Technologies and Internet of Things (IDCIoT)[C].Bengaluru, India: IEEE, 2024. 661-667.
[9] HEMALATHA S, JAYACHANDRAN J J B. A multitask learning-based vision transformer for plant disease localization and classification[J]. International journal of computational intelligence systems, 2024, 17(1): 188.
[10] WANG H X, NGUYEN T H, NGUYEN T N, et al.PD-TR: End-to-end plant diseases detection using a transformer[J]. Computers and electronics in agriculture, 2024, 224: 109123.
[11] RAHMAN K N, BANIK S C, ISLAM R, et al.A real time monitoring system for accurate plant leaves disease detection using deep learning[J]. Crop design, 2025, 4(1): 100092.
[12] KHAN Z, SHEN Y, LIU H.ObjectDetection in agriculture: A comprehensive review of methods, applications, challenges, and future directions[J]. Agriculture, 2025, 15(13): 1-36.
[13] DALAL M, MITTAL P.A systematic review of deep learning-based object detection in agriculture: Methods, challenges, and future directions[J]. Computers, materials & continua, 2025, 84(1): 57-91.
[14] SINGH D, JAIN N, JAIN P, et al.PlantDoc: A dataset for visual plant disease detection[A].Proceedings of the 7th ACM IKDD CoDS and 25th COMAD[C]. New York: ACM, 2020: 249-253.
[15] REN S Q,HE K M,GIRSHICK R,et al.Faster R-CNN:towards real-time object detection with region proposal networks[J].IEEE transactions on pattern analysis and machine intelligence,2017, 39(6):1137-1149.
[16] KHANAM R, HUSSAIN M. What is YOLOv5: A deep look into the internal features of the popular object detector[EB/OL].2024: arXiv: 2407.20892. https://arxiv.org/abs/2407.20892.
[17] SOHAN M, SAI RAM T, RAMI REDDY C V. A review on YOLOv8 and its advancements[A].Data Intelligence and Cognitive Informatics[C]. Singapore: Springer, 2024.529-545.
[18] ZHU X Z, SU W J, LU L W, et al. Deformable DETR: Deformable transformers for end-to-end object detection[EB/OL].2020:arXiv: 2010.04159. https://arxiv.org/abs/2010.04159.
[19] LV W Y, ZHAO Y A, CHANG Q Y, et al.RT-DETRv2: Improved baseline with bag-of-freebies for real-time detection transformer[EB/OL].2024:arXiv: 2407.17140. https://arxiv.org/abs/2407.17140.
[20] LIU S, LI F, ZHANG H, et al.DAB-DETR: Dynamic anchor boxes are better queries for DETR[A]International Conference on Learning Representations(ICLR)[C].Online:OpenReview.net, 2022.1-20.

相关文章 1

编辑推荐

Metrics

本文评价