An Improved Approach to Detection of Rice Leaf Disease with GAN-Based Data Augmentation Pipeline
Source abstract
Abstract
The lack of large balanced datasets in the agricultural field is a glaring problem for researchers and developers to design and train optimal deep learning models. This paper shows that using synthetic data augmentation outperforms the standard methods on object detection models and can be crucially important when datasets are few or imbalanced. The purpose of this study was to synthesize rice leaf disease data using a Style-Generative Adversarial Network Adaptive Discriminator Augmentation (SG2-ADA) and the variance of the Laplacian filter to improve the performance of Faster-Region-Based Convolutional Neural Network (faster-RCNN) and Single Shot Detector (SSD) in detecting the major diseases affecting rice. We collected a few unbalanced raw samples of rice leaf diseases images grouped into four diseases namely; bacterial blight (BB), tungro (TG), brown-spot (BS), and rice-blast (RB) with 1584, 1308, 1440, and 1600 images, respectively. We then train StyleGAN2-ADA for 250 epochs whilst using the variance of the Laplacian filter to discard blurry and poorly generated images. The synthesized images were used for augmenting faster-RCNN and SSD models in detecting rice leaf diseases. The StyleGAN2-ADA model achieved a Fréchet Inception Distance (FID) score of 26.67, Kernel Inception Distance (KID) score of 0.08, Precision of 0.49, and Recall of 0.14. In addition, we attained a mean average precision (mAP) of 0.93 and 0.91 for faster-RCNN and SSD, respectively. The learning curves of loss over 250 epochs are 0.03 and 0.04 for Faster-RCNN and SSD, respectively. In comparison to the standard data augmentation, we achieved a t-test p-value of 9.1×10−4 and 8.3×10−5. Hence, the proposed data augmentation pipeline to improve faster-RCNN and SSD models in detecting rice leaf diseases is significant. Our data augmentation approach is helpful to researchers and developers that are faced with the problem of fewer imbalanced datasets and can also be adopted by other fields faced with the same problems.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Smart Agriculture and AI · Plant Pathogenic Bacteria Studies · Plant Pathogens and Fungal Diseases
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
School of Automation and Electrical Engineering, Beihang University, Beijing 100191, China
School of Electronics and Information Engineering, Beihang University, Beijing 100191, China
References (63)
Show source-provided references
1. Behnassi, M., Baig, M.B., Sraïri, M.T., Alsheikh, A.A., and Risheh, A.W.A.A. (2022). Food Security and Climate-Climate Smart Food Systems—An Introduction. Food Security and Climate-Smart Food Systems, Springer.10.1007/978-3-030-92738-7 ↗
2. Falsafi, P., Baig, M.B., Reed, M.R., and Behnassi, M. (2022). The Nexus of Climate Change, Food Security, and Agricultural Extension in Islamic Republic of Iran. Food Security and Climate-Smart Food Systems, Springer.10.1007/978-3-030-92738-7_12 ↗
3. Interacting effects of land-use change and natural hazards on rice agriculture in the Mekong and Red River deltas in Vietnam10.5194/nhess-21-1473-2021 ↗
4. Agronomic practices preventing local outbreaks of rice yellow mottle virus disease revealed by spatial autoregressive analysis10.1007/s13593-022-00757-9 ↗
5. A real-time ap-proach of diagnosing rice leaf disease using deep learning-based faster R-CNN framework10.7717/peerj-cs.432 ↗
6. Rice diseases detection and classification using attention based neural network and bayesian optimization10.1016/j.eswa.2021.114770 ↗
7. Automatic diagnosis of rice diseases using deep learning10.3389/fpls.2021.701038 ↗
8. Kiratiratanapruk, K., Temniranrat, P., Kitvimonrat, A., Sinthupinyo, W., and Patarapuwadol, S. (2020). Using deep learn-ing techniques to detect rice diseases from images of rice fields. International Conference on Industrial, Engi-neering and Other Applications of Applied Intelligent Systems, Springer.10.1007/978-3-030-55789-8_20 ↗
9. Identification of rice diseases using deep convolutional neural net-works10.1016/j.neucom.2017.06.023 ↗
10. Image recognition of four rice leaf diseases based on deep learning and support vector machine10.1016/j.compag.2020.105824 ↗
11. Bhattacharya, S., Mukherjee, A., and Phadikar, S. (2020). A deep learning approach for the classification of rice leaf diseas-es. Intelligence Enabled Research, Springer.10.1007/978-981-15-2021-1_8 ↗
12. Mathulaprangsan, S., Lanthong, K., Jetpipattanapong, D., Sateanpattanakul, S., and Patarapuwadol, S. (2020, January 11–14). Rice diseases recognition using effective deep learning models. Proceedings of the Joint International Conference on Digi-tal Arts, Media and Technology with ECTI Northern Section Conference on Electrical, Electronics, Computer and Telecommunications Engineering (ECTI DAMT & NCON), Pattaya, Thailand.10.1109/ECTIDAMTNCON48261.2020.9090709 ↗
13. Identification and recogni-tion of rice diseases and pests using convolutional neural networks10.1016/j.biosystemseng.2020.03.020 ↗
14. Detection of rice plant diseases based on deep transfer learning10.1002/jsfa.10365 ↗
15. Joshi, A.A., and Jadhav, B.D. (2016, January 19–21). Monitoring and controlling rice diseases using Image processing techniques. Proceedings of the International Conference on Computing, Analytics and Security Trends (CAST), Pune, India.10.1109/CAST.2016.7915015 ↗
16. A framework for predicting the production performance of unconventional resources using deep learning10.1016/j.apenergy.2021.117016 ↗
17. Richter, S.R., Vineet, V., Roth, S., and Koltun, V. (2016). Playing for data: Ground truth from computer games. European Conference on Computer Vision, Springer.10.1007/978-3-319-46475-6_7 ↗
18. X-invariant Contrastive Augmentation and Representation Learning for Semi-Supervised Skeleton-Based Action Recognition10.1109/TIP.2022.3175605 ↗
19. Kang, J., Lee, S., Kim, N., and Kwak, S. (2022, January 19–24). Style Neophile: Constantly Seeking Novel Styles for Domain Generalization. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, New Orleans, LA, USA.10.1109/CVPR52688.2022.00699 ↗
20. Generative adversarial networks: Introduction and outlook10.1109/JAS.2017.7510583 ↗
21. Training generative adversarial networks with limited data
22. Unsupervised image translation using adversarial networks for im-proved plant disease recognition10.1016/j.compag.2019.105117 ↗
23. Genera-tive adversarial networks10.1145/3422622 ↗
24. A case study of conditional deep convolutional generative adversarial networks in machine fault diagnosis10.1007/s10845-020-01579-w ↗
25. Park, S.W., Ko, J.S., Huh, J.H., and Kim, J.C. (2021). Review on generative adversarial networks: Focusing on computer vi-sion and its applications. Electronics, 10.10.3390/electronics10101216 ↗
26. A method for improving CNN-based image recognition using DCGAN
27. A simple recurrent unit model based intrusion detection system with DCGAN10.1109/ACCESS.2019.2922692 ↗
28. Arjovsky, M., Chintala, S., and Bottou, L. (2017, January 6–11). Wasserstein generative adversarial networks. Proceedings of the International Conference on Machine Learning, Sydney, NSW, Australia.
29. Guo, T., Xu, C., Huang, J., Wang, Y., Shi, B., Xu, C., and Tao, D. (2020, January 13–19). On positive-unlabeled classification in GAN. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, Seattle, WA, USA.10.1109/CVPR42600.2020.00841 ↗
30. Song, J., and Ermon, S. (2020, January 13–18). Bridging the gap between f-gans and wasserstein gans. Proceedings of the 37th International Conference on Machine Learning, Online.
31. Karras, T., Laine, S., Aittala, M., Hellsten, J., Lehtinen, J., and Aila, T. (2020, January 13–19). Analyzing and improving the image quality of stylegan. Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, Seattle, WA, USA.10.1109/CVPR42600.2020.00813 ↗
32. Differentiable augmentation for data-efficient gan training
33. Viazovetskyi, Y., Ivashkin, V., and Kashin, E. (2020). Stylegan2 distillation for feed-forward image manipulation. European Conference on Computer Vision, Springer.10.1007/978-3-030-58542-6_11 ↗
34. Kumari, N., Zhang, R., Shechtman, E., and Zhu, J.-Y. (2022, January 18–24). Ensembling off-the-shelf models for gan training. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, New Orleans, LA, USA.10.1109/CVPR52688.2022.01039 ↗
35. Thermal Face Generation Using StyleGAN10.1109/ACCESS.2021.3085423 ↗
36. Štepec, D., and Skočaj, D. (2020). Image synthesis as a pretext for unsupervised histopathological diagnosis. International Workshop on Simulation and Synthesis in Medical Imaging, Springer.10.1007/978-3-030-59520-3_18 ↗
37. GAN-based synthetic medical image augmentation for increased CNN performance in liver lesion classification10.1016/j.neucom.2018.09.013 ↗
38. Saiz, F.A., Alfaro, G., Barandiaran, I., and Graña, M. (2021). Generative Adversarial Networks to Improve the Robustness of Visual Defect Segmentation by Semantic Networks in Manufacturing Components. Appl. Sci., 11.10.3390/app11146368 ↗
39. Detecting money laundering transactions with machine learning10.1108/JMLC-07-2019-0055 ↗
40. Sheema, D., Ramesh, K., Renjith, P.N., and Lakshna, A. (2021, January 25–27). Comparative Study of Major Algorithms for Pest Detection in Maize Crop. Proceedings of the International Conference on Intelligent Technologies, Hubbali, India.10.1109/CONIT51480.2021.9498280 ↗
41. Recent advances in image processing techniques for automated leaf pest and disease recognition–A review
42. Multi-Granularity Anchor-Contrastive Representation Learning for Semi-Supervised Skeleton-Based Action Recognition
43. Resampling and data augmentation for short-term PV output predition based on an imbalanced sky images dataset using convolutional neural networks10.1016/j.solener.2021.05.095 ↗
44. Hu, C., Sapkota, B.B., Thomasson, J.A., and Bagavathiannan, M.V. (2021). Influence of image quality and light consistency on the performance of convolutional neural networks for weed mapping. Remote Sens., 13.10.3390/rs13112140 ↗
45. Learning plannable representations with causal in-fogan
46. Huang, X., Li, Y., Poursaeed, O., Hopcroft, J., and Belongie, S. (2017, January 21–26). Stacked generative adversarial networks. Proceedings of the IEEE conference on computer vision and pattern recognition, Honolulu, HI, USA.10.1109/CVPR.2017.202 ↗
47. Zhong, Z., Zheng, L., Kang, G., Li, S., and Yang, Y. (2020, January 7–12). Random erasing data augmentation. Proceedings of the AAAI Conference on Artificial Intelligence, New York, NY, USA.
48. Zhang, H., Cisse, M., Dauphin, Y.N., and Lopez-Paz, D. (2017). Mixup: Beyond empirical risk minimization. arXiv.10.1007/978-1-4899-7687-1_79 ↗
49. Pandian, J.A., Geetharamani, G., and Annette, B. (2019, January 13–14). Data augmentation on plant leaf disease image dataset using im-age manipulation and deep learning techniques. Proceedings of the IEEE 9th International Conference on Advanced Computing (IACC), Tiruchirapalli, India.
50. Intrusion detection system for healthcare systems using medical and network data: A comparison study10.1109/ACCESS.2020.3000421 ↗
51. Mao, J., Xiao, T., Jiang, Y., and Cao, Z. (2017, January 21–26). What can help pedestrian detection?. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Honolulu, HI, USA.10.1109/CVPR.2017.639 ↗
52. Energy theft detection using gradient boosting theft detector with feature engineer-ing-based preprocessing10.1109/TSG.2019.2892595 ↗
53. Tests for the as-sessment of sport-specific performance in Olympic combat sports: A systematic review with practical recom-mendations10.3389/fphys.2018.00386 ↗
54. Gikunda, P.K., and Jouandeau, N. (2019). State-of-the-art convolutional neural networks for smart farms: A review. Intelligent Computing, Proceedings of 2019 Computing Conference, London, UK, 16–17 July 2019, Springer.10.1007/978-3-030-22871-2_53 ↗
55. Deep learning models for plant disease detection and diagnosis10.1016/j.compag.2018.01.009 ↗
56. Liu, W., Anguelov, D., Erhan, D., Szegedy, C., Reed, S., Fu, C.Y., and Berg, A.C. (2016). SSD: Single shot multibox detector. European Conference on Computer Vision, Springer.10.1007/978-3-319-46448-0_2 ↗
57. Autonomous structural visual inspection us-ing region-based deep learning for detecting multiple damage types10.1111/mice.12334 ↗
58. Zeiler, M.D., and Fergus, R. (2014). Visualizing and understanding convolutional networks. European Conference on Computer Vision, Springer.10.1007/978-3-319-10590-1_53 ↗
59. Binkowski, M., Simonyan, K., Donahue, J., Clark, A., Dieleman, S.E.L., Elsen, E.K., and Casagrande, N. (2021). High Fidelity Speech Synthesis With Adversarial Networks. (U.S. Patent 17/032,578).
60. Assessing generative models via precision and recall
61. Deep feature based rice leaf disease identification using support vector machine10.1016/j.compag.2020.105527 ↗
62. (2022, March 22). Tzutalin. LabelImg. Git Code. Available online: https://github.com/tzutalin/labelImg.
63. Clark, A. (2022, March 25). Pillow (PIL Fork) Documentation. Available online: https://buildmedia.readthedocs.org/media/pdf/pillow/latest/pillow.pdf.