Integrated Learning-Based Pest and Disease Detection Method for Tea Leaves
<p>Tea pest and disease category (<b>a</b>) indicates <span class="html-italic">Leaf blight</span> (<b>b</b>) indicates <span class="html-italic">Apolygus lucorμm</span> (<b>c</b>) indicates both <span class="html-italic">Leaf blight</span> and <span class="html-italic">Apolygus lucorμm</span>.</p> "> Figure 2
<p>Representative images in the tea dataset, including (<b>a</b>,<b>b</b>) individual tea tree photos (<b>c</b>,<b>d</b>) group tea tree photos.</p> "> Figure 3
<p>YOLOv5 structure picture.</p> "> Figure 4
<p>The overview of GAM.</p> "> Figure 5
<p>Channel attention submodule.</p> "> Figure 6
<p>Spatial attention submodule.</p> "> Figure 7
<p>The overview of CBAM.</p> "> Figure 8
<p>Channel attention submodule.</p> "> Figure 9
<p>Spatial attention submodule.</p> "> Figure 10
<p>Schematic representation of the WBF and NMS processing multiple predictions with the red box being the true labeled box and the blue box being the predictions made by multiple models.</p> "> Figure 11
<p>The process of merging two prediction boxes into one box through the fusion box formula.</p> "> Figure 12
<p>Integration Model Architecture Diagram.</p> "> Figure 13
<p>(<b>a</b>–<b>c</b>) is the effect of YOLOv5 + GAM detection, which can be found to be sensitive to <span class="html-italic">Apolygus lucorμm</span> over a large area but not to identify all <span class="html-italic">Leaf blight</span>. (<b>d</b>–<b>f</b>) represent YOLOv5 + CBAM detection, which can be found to detect most of the <span class="html-italic">Leaf blight</span> but not sensitive to <span class="html-italic">Apolygus lucorμm</span>, and there is leakage. (<b>g</b>–<b>i</b>) show the integrated fused model, which can be seen to be able to combine two models to detect both <span class="html-italic">Apolygus lucorμm</span> and <span class="html-italic">Leaf blight</span>.</p> "> Figure 13 Cont.
<p>(<b>a</b>–<b>c</b>) is the effect of YOLOv5 + GAM detection, which can be found to be sensitive to <span class="html-italic">Apolygus lucorμm</span> over a large area but not to identify all <span class="html-italic">Leaf blight</span>. (<b>d</b>–<b>f</b>) represent YOLOv5 + CBAM detection, which can be found to detect most of the <span class="html-italic">Leaf blight</span> but not sensitive to <span class="html-italic">Apolygus lucorμm</span>, and there is leakage. (<b>g</b>–<b>i</b>) show the integrated fused model, which can be seen to be able to combine two models to detect both <span class="html-italic">Apolygus lucorμm</span> and <span class="html-italic">Leaf blight</span>.</p> ">
Abstract
:1. Introduction
2. Materials and Methods
2.1. Datasets
2.2. YOLOv5
2.3. GAM Attention Mechanism
2.4. CBAM Attention Mechanism
2.5. Integrated Learning
2.6. Fusion Model CBAM_Fusion_GAM
2.7. Model Evaluation
2.8. Training
3. Results
3.1. Experimental Result
3.2. Comparison
4. Discussion
5. Conclusions
Author Contributions
Funding
Data Availability Statement
Conflicts of Interest
References
- Chen, S.L. Tea pest and disease control technology. Agrotech. Serv. 2009, 26, 52–54. [Google Scholar]
- Bian, K.O.; Yang, N.; Lu, Y.H. A review of deep learning applications in agricultural pest detection and identification. Softw. Guide 2021, 20, 26–33. [Google Scholar]
- Girshick, R.; Donahue, J.; Darrell, T.; Malik, J. Rich feature hierarchies for accurate object detection and semantic segmentation. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Columbus, OH, USA, 23–28 June 2014; pp. 580–587. [Google Scholar]
- Girshick, R. Fast R-CNN. In Proceedings of the 2015 IEEE International Conference on Computer Vision, Santiago, Chile, 7–13 December 2015; IEEE: New York, NY, USA, 2015; pp. 1440–1448. [Google Scholar]
- Ren, S.; He, K.; Girshick, R.; Sun, J. Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks. IEEE Trans. Pattern Anal. Mach. Intell. 2017, 39, 1137–1149. [Google Scholar] [CrossRef] [PubMed]
- Liu, W.; Anguelov, D.; Erhan, D.; Szegedy, C.; Reed, S.; Fu, C.Y.; Berg, A.C. Ssd: Single shot multibox detector. In Proceedings of the European Conference on Computer Vision, Amsterdam, The Netherlands, 11–14 October 2016; Springer: Cham, Switzerland, 2016; pp. 21–37. [Google Scholar]
- Jeong, J.; Park, H.; Kwak, N. Enhancement of SSD by concatenating feature maps for object detection. In Proceedings of the British Machine Vision Conference 2017, London, UK, 4–7 September 2017. [Google Scholar]
- Zhou, X.; Wang, D.; Krhenbuhl, P. Objects as points. arXiv 2019, arXiv:1904.07850. [Google Scholar]
- Redmon, J.; Divvala, S.; Girshick, R.; Farhadi, A. You only look once: Unified, real-time object detection. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Las Vegas, NV, USA, 27–30 June 2016; pp. 779–788. [Google Scholar]
- Zhang, J.; Nong, C.-R.; Yang, Z.-Y. A review of target detection algorithms based on convolutional neural networks. J. Arms Equip. Eng. 2022, 1–12. [Google Scholar]
- Wang, Y. Research on UAV-Based Tea Pest Control System. Master’s Thesis, China University of Mining and Technology, Beijing, China, 2019. [Google Scholar] [CrossRef]
- Xue, Z.; Xu, R.; Bai, D.; Lin, H. YOLO-Tea: A Tea Disease Detection Model Improved by YOLOv5. Forests 2023, 14, 415. [Google Scholar] [CrossRef]
- Bao, W.; Fan, T.; Hu, G.; Liang, D.; Li, H. Detection and identification of tea leaf diseases based on AX-RetinaNet. Sci. Rep. 2022, 12, 2183. [Google Scholar] [CrossRef] [PubMed]
- Yang, N.; Yuan, M.; Wang, P.; Zhang, R.; Sun, J.; Mao, H. Tea Diseases Detection Based on Fast Infrared Thermal Image Processing Technology. J. Sci. Food Agric. 2019, 99, 3459–3466. [Google Scholar] [CrossRef] [PubMed]
- Lee, S.; Lin, S.; Chen, S. Identification of tea foliar diseases and pest damage under practical field conditions using a convolutional neural network. Plant Pathol. 2020, 69, 1731–1739. [Google Scholar] [CrossRef]
- Li, H.; Shi, H.; Du, A.; Mao, Y.; Fan, K.; Wang, Y.; Shen, Y.; Wang, S.; Xu, X.; Tian, L.; et al. Symptom recognition of disease and insect damage based on Mask R-CNN, wavelet transform, and F-RNet. Front. Plant Sci. 2022, 13, 922797. [Google Scholar] [CrossRef] [PubMed]
- Srivastava, A.R.; Venkatesan, M. Tea leaf disease prediction using texture-based image processing. In Emerging Research in Data Engineering Systems and Computer Communications, Proceedings of CCODE 2019; Springer: Singapore, 2020; pp. 17–25. [Google Scholar]
- Liu, Y.; Shao, Z.; Hoffmann, N. Global Attention Mechanism: Retain Information to Enhance Channel-Spatial Interactions. arXiv 2021, arXiv:2112.05561. [Google Scholar]
- Woo, S.; Park, J.; Lee, J.Y.; Kweon, I.S. CBAM: Convolutional block attention module. In Computer Vision-ECCV 2018; Springer International Publishing: Cham, Switzerland, 2018; pp. 3–19. [Google Scholar] [CrossRef]
- Solovyev, R.; Wang, W.; Gabruseva, T. Weighted boxes fusion: Ensembling boxes from different object detection models. Image Vis. Comput. 2021, 107, 104117. [Google Scholar] [CrossRef]
- Dietterich, T.G. Ensemble learning. Handb. Brain Theory Neural Netw. 2002, 2, 110–125. [Google Scholar]
- Sagi, O.; Rokach, L. Ensemble learning: A survey. Wiley Interdiscip. Rev. Data Min. Knowl. Discov. 2018, 8, e1249. [Google Scholar] [CrossRef]
- Zhou, D.; Fang, J.; Song, X.; Guan, C.; Yin, J.; Dai, Y.; Yang, R. IoU loss for 2D/3D object detection. In Proceedings of the 2019 International Conference on 3D Vision (3DV), Québec City, QC, Canada, 16–19 September 2019; IEEE: New York, NY, USA, 2019; pp. 85–94. [Google Scholar] [CrossRef]
- Lin, H.; Tang, C. Intelligent Bus Operation Optimization by ntegrating Cases and Data Driven Based on Business Chain and Enhanced Ouantum Genetic Algorithm. IEEE Trans. Intell. Transp. Syst. 2021, 23, 9869–9882. [Google Scholar] [CrossRef]
- Lin, H.; Tang, C. Analysis and Optimization of Urban Public Transport Lines Based on Multiobjective Adaptive Particle Swarm Optimization. IEEE Trans. Intell. Transp. Syst. 2021, 23, 16786–16798. [Google Scholar] [CrossRef]
- Xue, X.; Jin, S.; An, F.; Zhang, H.; Fan, J.; Eichhorn, M.P.; Jin, C.; Chen, B.; Jiang, L.; Yun, T. Shortwave radiation calculation for forest plots using airborne LiDAR data and computer graphics. Plant Phenom. 2022, 2022, 9856739. [Google Scholar] [CrossRef] [PubMed]
Pest and Disease Categories | Apolygus lucorμm | Leaf blight |
---|---|---|
Label name | D00 | D10 |
Number | 289 (112 pictures of pest and disease mix) | 273 (112 pictures of pest and disease mix) |
Model | Test Case 1 | Test Case 2 | Test Case 3 |
---|---|---|---|
✓ | ✓ | ✗ | |
✓ | ✗ | ✓ | |
✗ | ✓ | ✓ | |
Integration | ✓ | ✓ | ✓ |
Model | Test Case 1 | Test Case 2 | Test Case 3 |
---|---|---|---|
✓ | ✓ | ✗ | |
✓ | ✓ | ✗ | |
✓ | ✓ | ✗ | |
Integration | ✓ | ✓ | ✗ |
Model | Test Case 1 | Test Case 2 | Test Case 3 |
---|---|---|---|
✓ | ✗ | ✗ | |
✗ | ✓ | ✗ | |
✗ | ✗ | ✓ | |
Integration | ✗ | ✗ | ✗ |
Test Environment | Details |
---|---|
Programming language | Python 3.9 |
Operating system | Windows 11 |
Deep learning framework | Pytorch 1.12.1 |
GPU | NVIDIA GeForce RTX3060 |
Training Parameters | Details |
---|---|
Epochs | 250 |
Batch-size | 8 |
img-size (pixels) | 640 ∗ 640 |
Optimization algorithm | SGD |
Initial learning rate | 0.01 |
Dataset | Train | Val | Test |
---|---|---|---|
Number | 360 | 45 | 45 |
Model | P (%) | mAP (%) | AVG P (%) | ||
---|---|---|---|---|---|
D10 | D00 | D10 | D00 | ||
YOLOv5 | 68.6 | 68.3 | 74.1 | 69.3 | 68.4 |
YOLOv5 + CBAM | 73.3 | 67.9 | 74.4 | 63.3 | 70.6 |
YOLOv5 + GAM | 66.2 | 73.1 | 72.2 | 66.3 | 69.7 |
YOLOv4 | 66.3 | 64.6 | 69.3 | 65.3 | 65.4 |
YOLO v5 + transformer_layer | 70.3 | 67.8 | 74.9 | 66.6 | 69 |
YOLOv3 | 67.5 | 62.6 | 68.4 | 64.2 | 65 |
CBAM_fusion_GAM | 80.0 | 78.6 | 74.8 | 68.5 | 79.3 |
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. |
© 2023 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
Share and Cite
Wang, Y.; Xu, R.; Bai, D.; Lin, H. Integrated Learning-Based Pest and Disease Detection Method for Tea Leaves. Forests 2023, 14, 1012. https://doi.org/10.3390/f14051012
Wang Y, Xu R, Bai D, Lin H. Integrated Learning-Based Pest and Disease Detection Method for Tea Leaves. Forests. 2023; 14(5):1012. https://doi.org/10.3390/f14051012
Chicago/Turabian StyleWang, Yinkai, Renjie Xu, Di Bai, and Haifeng Lin. 2023. "Integrated Learning-Based Pest and Disease Detection Method for Tea Leaves" Forests 14, no. 5: 1012. https://doi.org/10.3390/f14051012