Article Details

Title · Authors · Abstract · Keywords · DOI · References

Research on Tomato Leaf Disease Recognition Based on Contrastive Learning

HUANG Zhongping

(School of Computer Science and Engineering, Anhui University of Science and Technology, Huainan 232001, China)

Section:Artificial Intelligence · 2025, Vol. 9, No. 11 · Pages:43-48 · 2096-4706(2025)11-0043-06 · CLC number:TP391.4 · Document code:A · DOI:10.19850/j.cnki.2096-4706.2025.11.009

Abstract

In view of the current situation that traditional Deep Learning models exhibit low recognition accuracy and limited generalization ability in the process of tomato leaf disease recognition, this paper proposes a recognition model based on the Supervised Contrastive Learning method. The model proposes the AMECA module based on Channel Attention, which effectively captures the dependencies among channels, enhances the model's channel fusion ability, and improves the recognition performance. The AMECA module is integrated into ResNet18 model as an image feature extractor, and a high-precision tomato leaf disease recognition model is trained through the Supervised Contrastive Learning method. The experimental results on the tomato leaf disease dataset show that the accuracy of the model reaches 99.198%, which is 3.208% higher than that of the original ResNet18 model. Compared with some other traditional Convolutional Neural Networks, it has higher recognition accuracy and can better recognize tomato leaf diseases. It is applicable to tomato leaf images obtained in natural scenes, and demonstrates strong practicability.

Keywords

Attention Mechanism; tomato leaf disease recognition; image recognition; Contrastive Learning

References

  1. [1] 张志刚,董涛,李瑞云.基于专利分析的中国番茄产业发展研究 [J].农业图书情报学刊,2018,30(11):72-76.
  2. [2] 王振奎.日光温室番茄高效栽培技术 [J].农业工程技术,2024,44(15):50-51.
  3. [3] 刘笠溶.基于深度学习的西红柿叶片病害识别研究及应用 [D].昆明:云南农业大学,2023.
  4. [4] HE K,ZHANG X,REN S,et al. Deep Residual Learning for Image Recognition [C]//Proceedings of the IEEE Conference on Computer Vision and Pattern recognition.Las Vegas:IEEE,2016:770-778.
  5. [5] VASWANI A,SHAZEER N,PARMAR N,et al. Attention is All You Need [C]//NIPS'17:Proceedings of the 31st International Conference on Neural Information Processing Systems.
  6. [6] 王汉生,姚建斌.基于ResNet和ViT双流网络的小麦病虫害识别 [J].农业技术与装备,2024(2):18-21.
  7. [7] WANG Q,WU B,ZHU P,et al. ECA-Net: Efficient Channel Attention for Deep Convolutional Neural Networks [C]//Proceedings of the IEEE/CVF conference on computer vision and pattern recognition.Seattle:IEEE,2020:11534-11542.
  8. [8] 叶荣,马自飞,高泉,等.基于改进YOLOv5s-ECA-ASFF算法的茶叶病害目标检测 [J].中国农机化学报,2024,45(1):244-251.
  9. [9] 曹义亲,曹鑫晨.融合膨胀卷积与ECA的钢材缺陷检测算法 [J].计算机工程与设计,2024,45(11):3312-3319.
  10. [10] WOO S,PARK J,LEE J Y,et al. CBAM: Convolutional Block Attention Module [C]//Proceedings of the European conference on computer vision (ECCV).Munich:Springer,2018:3-19.
  11. [11] 张重生,陈杰,李岐龙,等.深度对比学习综述 [J].自动化学报,2023,49(1):15-39.
  12. [12] CHEN T,KORNBLITH S,NOROUZI M,et al. A Simple Framework for Contrastive Learning of Visual Representations [C]//International Conference on Machine Learning. PMLR,2020:1597-1607.
  13. [13] 李希,刘喜平,李旺才,等.对比学习研究综述 [J].小型微型计算机系统,2023,44(4):787-797.
  14. [14] HE K,FAN H,WU Y,et al. Momentum Contrast for Unsupervised Visual Representation Learning [C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition.Seattle:IEEE,2020:9729-9738.
  15. [15] KHOSLA P,TETERWAK P,WANG C,et al. Supervised Contrastive Learning [J].Advances in Neural Information Processing Systems,2020,33:18661-18673.
  16. [16] 陈承功.基于混合图像特征与区域特征的监督对比学习算法研究 [D].开封:河南大学,2024.
  17. [17] 张艳,车迅,汪芃,等.基于双节点-双边图神经网络的茶叶病害分类方法 [J].农业机械学报,2024,55(3):252-262.

About the authors

Huang Zhongping (1984-), male, Han ethnicity, from Anqing, Anhui Province, is currently pursuing a master's degree in computer vision.