基于改进YOLOv8算法的谷子田杂草检测

doi:10.13733/j.jcam.issn.2095-5553.2025.01.028

摘要/Abstract

摘要： 针对谷子田环境复杂、杂草种类众多、杂草分布密集的特点导致识别精度低的问题，提出一种基于YOLOv8的改进模型。通过加入CloFormer结构来减少YOLOv8算法计算量并提高识别精度，使用Global和Local的注意力与c2f模块进行融合，使用AttnConv共享权重来整合局部信息，部署上下文感知权重来增强局部特征；为进一步提高识别精度，另外添加Gam注意力机制，与当前较先进的注意力机制进行对比试验，并与YOLO各系列模型进行对比试验。结果表明，YOLOv8-CG模型检测的平均精度均值为92.6%，比YOLOv5模型高4%。同时分析垄的种植密度不同对模型识别产生的影响，种植较为稀疏的10号垄比种植密集的2号垄精度高6.6%。

关键词: 杂草检测, 谷子, YOLOv8, 注意力机制, 轻量级模型

Abstract: An improved model based on YOLOv8 was proposed to solve the problem of low recognition accuracy due to the complex environment, numerous weed species and dense weed distribution in millet field. By adding CloFormer structure, the model reduced the computational load of YOLOv8 algorithm and improved the recognition accuracy. This structure mainly used Global and Local attention to integrate with c2f module, used AttnConv shared weights to integrate local information, and deployed contextaware weights to enhance local features. In order to further improve the recognition accuracy, Gam attention mechanism was added, and comparison experiments were conducted with the current more advanced attention mechanism and the YOLO series. According to the experiments, the average detection accuracy of YOLOv8-CG model was 92.6%, 4% higher than that of YOLOv5 model. At the same time, the effect of different planting density on model recognition was analyzed. The experiment showed that the precision of row 10 with sparse planting was 6.6% higher than that of row 2 with dense planting.

Key words: weed detection, millet, YOLOv8, attention mechanism, lightweight model

中图分类号:

S451

王鑫淼, 张正, 董晓威, 王林烽, 李瑞祥. 基于改进YOLOv8算法的谷子田杂草检测[J]. 中国农机化学报, 2025, 46(1): 185-189.

Wang Xinmiao, Zhang Zheng, Dong Xiaowei, Wang Linfeng, Li Ruixiang. Research on weed detection in millet field based on improved YOLOv8 algorithm[J]. Journal of Chinese Agricultural Mechanization, 2025, 46(1): 185-189.

参考文献

［1］　付豪, 赵学观, 翟长远, 等. 基于深度学习的杂草识别方法研究进展［J］. 中国农机化学报, 2023, 44（5）: 198-207.
Fu Hao, Zhao Xueguan, Zhai Changyuan, et al. Research progress on weed recognition method based on deep learning technology ［J］. Journal of Chinese Agricultural Mechanization, 2023, 44（5）: 198-207.
［2］　庄占兴, 孙文国, 范金勇, 等. 西草净对谷子田一年生杂草活性及其安全性测定［J］. 农药, 2017, 56（7）: 531-534.Zhuang Zhanxing, Sun Wenguo, Fan Jinyong, et al. Weed control effect of simetryn and its safety to millet in glasshouses ［J］. Agrochemicals, 2017, 56（7）: 531-534.
［3］　赵玉信, 杨惠敏. 作物格局、土壤耕作和水肥管理对农田杂草发生的影响及其调控机制［J］. 草业学报, 2015, 24（8）: 199-210.
Zhao Yuxin, Yang Huimin. Effects of crop pattern, tillage practice and water and fertilizer management on weeds and their control mechanisms ［J］. Acta Prataculturae Sinica, 2015, 24（8）: 199-210.
［4］　姜延军, 岳德成, 李青梅, 等. 全膜双垄沟播玉米田选用除草地膜的适宜田间杂草密度研究［J］. 植物保护, 2018, 44（1）: 110-115.Jiang Yanjun, Yue Decheng, Li Qingmei, et al. Effects of covering weeding film on the suitable weed density in doubleridge maize fields with whole plasticfilm mulching ［J］. Plant Protection, 2018, 44（1）: 110-115.
［5］　Girshick R, Donahue J, Darrell T, et al. Rich feature hierarchies for accurate object detection and semantic segmentation ［C］. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2014: 580-587.
［6］　Ren S, He K, 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, 2015, 39（6）: 1137-1149.
［7］　Liu W, Anguelov D, Erhan D, et al. SSD: Single shot multibox detector ［C］. European Conference on Computer Vision, 2016: 21-37.
［8］　Redmon J, Farhadi A. YOLOv3: An incremental improvement ［J］. arXiv:180402767, 2018.
［9］　刘莫尘, 高甜甜, 马宗旭, 等. 基于MSRCR-YOLOv4-tiny的田间玉米杂草检测模型［J］. 农业机械学报, 2022, 53（2）: 246-255, 335.
Liu Mochen, Gao Tiantian, Ma Zongxu, et al. Target detection model of corn weeds in field environment based on MSRCR algorithm and YOLOv4-tiny ［J］. Transactions of the Chinese Society for Agricultural Machinery, 2022, 53（2）: 246-255, 335.

［10］　王宇博, 马廷淮, 陈光明. 基于改进YOLOv5算法的农田杂草检测［J］. 中国农机化学报, 2023, 44（4）: 167-173.

Wang Yubo, Ma Tinghuai, Chen Guangming. Weeds detection in farmland based on a modified YOLOv5 algorithm ［J］. Journal of Chinese Agricultural Mechanization, 2023, 44（4）: 167-173.
［11］　张伟康, 孙浩, 陈鑫凯, 等. 基于改进YOLOv5的智能除草机器人蔬菜苗田杂草检测研究［J］. 图学学报, 2023, 44（2）: 346-356.〖JP2〗Zhang Weikang, Sun Hao, Chen Xinkai, et al. Research on weed detection in vegetable seedling fields based on the improved YOLOv5 intelligent weeding robot ［J］. Journal of Graphics, 2023, 44（2）: 346-356.
［12］　Chen Jiqing, Wang Huabin, Zhang Hongdu, et al. Weed detection in sesame fields using a YOLO model with an enhanced attention mechanism and feature fusion ［J］. Computers and Electronics in Agriculture, 2022, 202.
［13］　Fan Qihang, Huang Huaibo, Guan Jiyang, et al. Rethinking local perception in lightweight vision transformer ［J］. arXiv:230317803, 2023.
［14］　Liu Yichao, Shao Zongru, Nico Hoffmann. Global attention mechanism: Retain information to enhance channelspatial interactions ［J］. arXiv:211205561, 2021.
［15］　Park J, Woo S, Lee J Y, et al. BAM: Bottleneck attention module ［J］. arXiv:180706514, 2018.
［16］　Woo S, Park J, Lee J Y, et al. CBAM: Convolutional block attention module ［C］. Proceedings of the European Conference on Computer Vision, 2018: 3-19.
［17］　Jie Hu, Li Shen, Samuel Albanie, et al. Squeeze and excitation networks ［J］. arXiv:170901507,2019.
［18］　Ouyang Daliang, He Su, Zhang Guozhong, et al. Efficient multiscale attention module with crossspatial learning ［J］. arXiv: 230513563, 2023.
［19］　Li Yehao, Yao Ting, Pan Yingwei, et al. Contextual transformer networks for visual recognition ［J］. arXiv: 210712292, 2021.
［20］　杨断利, 王永胜, 陈辉, 等. 基于改进YOLO v6-tiny的蛋鸡啄羽行为识别与个体分类［J］. 农业机械学报, 2023, 54（5）: 268-277.
Yang Duanli, Wang Yongsheng, Chen Hui, et al. Feather pecking abnormal behavior identification and individual classification method of laying hens based on improved YOLO v6-tiny ［J］. Transactions of the Chinese Society for Agricultural Machinery, 2023, 54（5）: 268-277.

[1]	高芳征, 温鑫, 黄家才, 陈光明, 金少宇, 赵雪迪. 基于AD-YOLOX-Nano的茶叶嫩芽识别算法[J]. 中国农机化学报, 2025, 46(1): 178-184.
[2]	高天赐, 王克俭, 陈晨, 韩宪忠, 王超, 李会平, . 基于DCP-ShuffleNetV2的轻量级森林害虫识别方法[J]. 中国农机化学报, 2025, 46(1): 190-197.
[3]	卢明, 余心杰, 郭俊先. 基于改进YOLOv8算法的水稻田间人工施肥行为识别[J]. 中国农机化学报, 2025, 46(1): 220-226.
[4]	韩鹏飞, 宋其江, 贾梦实. 基于改进轻量化 EfficientNet-V2模型的小麦种子分类[J]. 中国农机化学报, 2024, 45(9): 111-117.
[5]	张传栋, 亓璐, 丁华立. 基于改进 YOLOv8n模型的多品种葡萄簇检测方法[J]. 中国农机化学报, 2024, 45(9): 220-226.
[6]	李洁, 高尚兵, 余骥远, 陈新, 李士丛, 袁星星. 基于PMMS-Net和叶绿素荧光成像的绿豆叶斑病抗性鉴定方法[J]. 中国农机化学报, 2024, 45(8): 210-216.
[7]	刘星, 顾寄南, 黄则栋, 张文浩, 张伟. 基于深度学习的苹果点云语义分割方法研究[J]. 中国农机化学报, 2024, 45(8): 223-227.
[8]	苏斐, 王光辉, 史艳霞, 贾然, 闫银发, 祖林禄, . 融合通道注意力机制的ResUnet作物病斑分割模型[J]. 中国农机化学报, 2024, 45(8): 228-233.
[9]	冯晓, 张辉, 刘运超, 张微, 李小红, 马中杰, . 基于改进YOLOv5s的硬核期葡萄簇检测[J]. 中国农机化学报, 2024, 45(8): 240-245.
[10]	周思捷, 刘天奇, 陈天华. 基于改进YOLOv5算法的水稻病害识别研究[J]. 中国农机化学报, 2024, 45(8): 246-253.
[11]	王克涛, 陈世锋, 陈贵, 韦锦, 蒙丽雯, 陈泉成. 基于YOLOv5s的园林球形绿篱检测方法[J]. 中国农机化学报, 2024, 45(8): 262-269.
[12]	岳耀华, 张伟, 亓立强, . 基于改进YOLOv5的田间大豆花朵生长状态检测方法研究[J]. 中国农机化学报, 2024, 45(7): 188-193.
[13]	吴利刚, 吕媛媛, 周倩, 陈乐, 张梁, 史建华. 基于改进YOLOv5的轻量级黄花成熟检测方法[J]. 中国农机化学报, 2024, 45(7): 235-242.
[14]	张紫恒, 陶桂香, 衣淑娟, 王帏, 岳晴, 李睿. 谷子风筛式清选装置设计与试验[J]. 中国农机化学报, 2024, 45(6): 119-127.
[15]	雷浩, 苑迎春, 何振学. 基于多尺度卷积和注意力机制的枣品种识别[J]. 中国农机化学报, 2024, 45(6): 135-141.