基于改进YOLOv5s的香菇菌棒污染识别方法

doi:10.13733/j.jcam.issn.2095‑5553.2025.02.032

摘要/Abstract

摘要： 为提高香菇菌棒污染识别的准确性和效率，提出一种改进YOLOv5s的香菇菌棒污染识别模型(YOLOv5s—CGGS)。首先，在YOLOv5s的特征提取网络引入CA(Coordinate Attention)注意力机制提高菌棒污染的可识别性和目标定位的准确性。然后，将CIoU损失函数替换为SIoU损失函数，提高模型收敛速度和推理准确性。最后，采用GSConv和GhostConv模块对特征融合网络进行改进优化，提高识别效率。该方法的精确率、召回率、mAP@0.5分别达到97.68%、97.20%、98.01%，相比于YOLOv5s分别提高2.21%、2.79%、1.31%。mAP@0.5比YOLOv4、Ghost—YOLOv4、Mobilenetv3—YOLOv4提高4.42%、4.97%、5.85%，且FPS提高2~3倍。

关键词: 香菇菌棒, 污染识别, 深度学习, 注意力机制, 损失函数

Abstract: In order to improve the accuracy and efficiency of Lentinula Edodes logs contamination identification, an improved YOLOv5s contamination identification model for Lentinula Edodes logs (YOLOv5s—CGGS) is proposed in this paper. Firstly, a CA (coordinate attention) mechanism is introduced in the feature extraction network of YOLOv5s to improve the identifiability of Lentinula Edodes logs contamination and the accuracy of target localization. Then, the CIoU (Complete—IoU) loss function is replaced by an SIoU (SCYLLA—IoU) loss function to improve the model's convergence speed and inference accuracy. Finally, the GSConv and GhostConv modules are used to improve and optimize the feature fusion network to enhance identification efficiency. The method in this paper achieves values of 97.68%, 97.20%, and 98.01% in precision, recall, and mAP@0.5, which are 2.21%, 2.79%, and 1.31% higher than that of YOLOv5s, respectively. mAP@0.5 is 4.42%, 4.97%, and 5.85% higher than YOLOv4, Ghost—YOLOv4, and Mobilenetv3—YOLOv4, and the FPS is increased by two to three times.

Key words: Lentinula Edodes logs, contamination identification, deep learning, attention mechanism, loss function

中图分类号:

S436.46+1

吴秋兰, 陈雪飞, 陈超, 张峰, 王姝妹, 赵恒. 基于改进YOLOv5s的香菇菌棒污染识别方法[J]. 中国农机化学报, 2025, 46(2): 217-223.

Wu Qiulan, Chen Xuefei, Chen Chao, Zhang Feng, Wang Shumei, Zhao Heng. Identification method of Lentinus Edodes logs contamination based on improved YOLOv5s[J]. Journal of Chinese Agricultural Mechanization, 2025, 46(2): 217-223.

参考文献

[ 1 ] 曹子健, 王守现, 郑素月, 等. 香菇菌棒污染菌宛氏拟青霉的鉴定及其与香菇菌丝的相互作用[J]. 北方园艺, 2022(14): 116-125
Cao Zijian, Wang Shouxian, Zheng Suyue, et al. Identification of Paecilomyces variotii on contaminated artificial Bed‑logs of Lentinula Edodes and their interactions [J]. North Horticulture, 2022(14): 116-125.
[ 2 ] Gong W, Zhang W, Bilal M, et al. Efficient web APIs recommendation with privacy‑preservation for mobile App development in industry 4.0 [J]. IEEE Transactions on Industrial Informatics, 2021, 18(9): 6379-6387.
[ 3 ] Wang D, Wang J, Ren Z, et al. DHBP: A dual‑stream hierarchical bilinear pooling model for plant disease multi‑task classification [J]. Computers and Electronics in Agriculture, 2022, 195: 106788.
[ 4 ] Wang G, Wang J, Yu H, et al. Research on identification of corn disease occurrence degree based on improved ResNeXt network [J]. International Journal of Pattern Recognition and Artificial Intelligence, 2022, 36(2): 2250005.
[ 5 ] Zu Dawei, Zhang Feng, Wu Qiulan, et al. Disease identification of Lentinus Edodes logs based on deep learning model [J]. Complexity, 2022: 1-9.
[ 6 ] Zu Dawei, Zhang Feng, Wu Qiulan, et al. Sundry bacteria contamination identification of Lentinula edodes logs based on deep learning model [J]. Agronomy, 2022, 12(9): 2121.
（下转第244页）
（上接第223页）
[ 7 ] Redmon J. YOLOv3: An incremental improvement [J]. arxiv Preprint arxiv: 1804.02767, 2018.
[ 8 ] Bochkovskiy A, Wang C Y, Liao H Y M. YOLOv4: Optimal speed and accuracy of object detection [J]. arXiv Preprint arXiv: 2004.10934, 2020.
[ 9 ] Zheng Z, Wang P, Ren D, et al. Enhancing geoindicator factors in model learning and inference for object detection and instance segmentation [J]. IEEE Transactions on Cybernetics, 2021, 52(8): 8574-8586.
[10] Gevorgyan Z. SIoU loss: More powerful learning for bounding box regression [J]. arXiv Preprint arXiv: 2205.12740, 2022.
[11] Tan X, Peng H. Improved YOLOv5 SAR image ship target detection [J]. Computer Engineering and Application, 2022, 58(4): 247-254.
[12] Hou Q, Zhou D, Feng J. Coordinate attention for efficient mobile network design [C]. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2021: 13713-13722.
[13] Li Hulin, Li Jun, Wei Hanbing, et al. Slim‑neck by GSConv: A better design paradigm of detector architectures for autonomous vehicles [J]. arXiv preprint arXiv: 2206.02424, 2022.
[14] Han K, Wang Y, Tian Q, et al. GhostNet: More features from cheap operations [C]. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2020: 1580-1589.
[15] Wong L J, Michaels A J. Transfer learning for radio frequency machine learning: A taxonomy and survey [J]. Sensors, 2022, 22(4): 1416.
[16] Selvaraju R R, Cogswell M, Das A, et al. Grad—CAM: Visual explanations from deep networks via gradient‑based localization [C]. Proceedings of the IEEE International Conference on Computer Vision, 2017: 618-626.

[1]	戴敏, 孙文靖, 缪宏. 基于轻量化CBAM—GoogLeNet的辣椒病虫害识别[J]. 中国农机化学报, 2025, 46(2): 224-229.
[2]	彭雨侬, 柳平增, 张艳, . 基于深度学习的玉米生产过程知识图谱构建[J]. 中国农机化学报, 2025, 46(2): 245-252.
[3]	李涛, 买买提明⋅艾尼, 古丽巴哈尔⋅托乎提, 杨佳雨. 基于改进YOLOv8的小棚架下无核白葡萄果梗识别[J]. 中国农机化学报, 2025, 46(2): 259-263.
[4]	段小勇, 何超, 刘学渊. 基于改进DeepLabV3+的非结构化道路可行驶区域检测[J]. 中国农机化学报, 2025, 46(2): 271-278.
[5]	李先旺, 刘赛虎, 黄忠祥, 章霞东. 基于XLNet—BiLSTM—AFF—CRF的谷物收割机械维修知识命名实体识别[J]. 中国农机化学报, 2025, 46(2): 319-325.
[6]	夏子林, 张新洲, 王文波, 夏先飞, 陈兰, 顾寄南. 基于机器视觉的蚕豆荚高精度检测方法研究[J]. 中国农机化学报, 2025, 46(1): 157-163.
[7]	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.
[8]	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.
[9]	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.
[10]	李春雨, 郭肖琴, 杨晶晶, 李忠华. 基于深度学习的水果图像识别[J]. 中国农机化学报, 2025, 46(1): 198-203.
[11]	牛潘婷, 张宝林, 潘丽杰, 郭建鹏(. 基于改进GoogLeNet的玉米叶片病害识别及其可解释性研究[J]. 中国农机化学报, 2025, 46(1): 204-212.
[12]	彭勇, 乔印虎, 张春燕, 姚杰, 包殿令. 改进YOLOv5模型的草莓检测方法研究[J]. 中国农机化学报, 2025, 46(1): 213-219.
[13]	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.
[14]	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.
[15]	李志臣, 凌秀军, 李鸿秋. 基于声音深度学习的发动机失火故障诊断[J]. 中国农机化学报, 2024, 45(9): 134-140.