Enhancing Potato Leaf Disease Detection: Implementation of Convolutional Vision Transformers with Synthetic Datasets from Stable Diffusion

Tri Astuti - Universitas AMIKOM Purwokerto, Purwokerto, Indonesia
Amri Umar - Universitas AMIKOM Purwokerto, Purwokerto, Indonesia
Rizki Wahyudi - Universitas AMIKOM Purwokerto, Purwokerto, Indonesia
Zanuar Rifai - Universitas AMIKOM Purwokerto, Purwokerto, Indonesia


Citation Format:



DOI: http://dx.doi.org/10.62527/joiv.8.4.2167

Abstract


Numerous studies have addressed the classification of potato plants. However, the available datasets often lack the necessary diversity to improve the accuracy of predictive classification models effectively. Our research capitalizes on synthetic datasets generated through the Stable Diffusion 1.5 image generation method to address this challenge. This study suggests a new way to solve the problem by using artificial datasets created with the Stable Diffusion 1.5 method to teach a Convolutional Vision Transformer (CvT) model how to identify diseases on potato leaves accurately. Our objective is to train the CvT model employing synthetic datasets to excel in detecting potato leaf diseases. Our methodology encompasses the model's training using synthetic datasets from Stable Diffusion 1.5. We employ a comprehensive dataset of 11,121 synthetic images to train the Convolutional Vision Transformer (CvT) model, which enables it to accurately identify various potato leaf diseases such as black leg/soft rot, mosaic, leaf roll, early blight, and late blight. We conduct evaluations at multiple training stages to gauge the model's performance and accuracy. The outcomes of our research underscore the effectiveness of employing synthetic datasets from Stable Diffusion 1.5, which significantly augments the available image data while preserving a high level of accuracy. The CvT model proficiently identifies potato leaf diseases with an evaluation accuracy of 84%. Additional testing reveals that by the fifth epoch, the CvT model attains an accuracy of 81% when assessed using 82 randomly selected images of diseased plants from Google. The implications of this research are far-reaching, particularly within the domains of image processing and agriculture. The strategy of utilizing synthetic datasets to train the CvT model presents an efficient remedy to address the limitations of original image datasets. The adept disease detection capability of the CvT model holds the potential to expedite plant condition identification, mitigate crop loss, and ultimately amplify agricultural productivity. This study effectively demonstrates that the Convolutional Vision Transformer (CvT), when leveraged with synthetic datasets from Stable Diffusion 1.5, produces a model capable of accurately identifying potato leaf diseases. These findings bear positive implications for both the agricultural and image-processing sectors. 

Keywords


Convolutional vision transformer; potato leaf disease identification; synthetic dataset; stable diffusion

Full Text:

PDF

References


U. Isbah and R. Y. Iyan, “Analisis peran sektor pertanian dalam perekonomian dan kesempatan kerja di Provinsi Riau,” Jurnal Sosial Ekonomi Pembangunan, vol. 7, no. 19, pp. 45–54, 2016.

Kementerian Pertanian, Analisis PDB Sektor Pertanian Tahun 2022 - Pusat Data dan Sistem Informasi Pertanian Kementerian Pertanian 2022, vol. 1. Jakarta Selatan: Pusat Data dan Sistem Informasi Pertanian, 2022.

Kementerian Pertanian, “Kementerian Pertanian Direktorat Jenderal Perkebunan » Kementan terus tingkatkan sumber devisa ekspor nasional dari sektor non migas.” Accessed: Jul. 16, 2023. [Online]. Available: https://ditjenbun.pertanian.go.id/kementan-terus-tingkatkan-sumber-devisa-ekspor-nasional-dari-sektor-non-migas/

Kementerian Pertanian, Pemetaan Produksi dan Konsumsi Unggulan Hortikultura Internasional, vol. 1. Jakarta Selatan: Biro Kerja Sama Luar Negeri - Kementerian Pertanian RI, 2017.

I. A. Astarini, I. G. R. M. Temaja, Kusmana, and D. Margareth, Tentang Kentang, vol. 1. Denpasar: Udayana University Press, 2019.

M. Ghosh and G. Sanyal, “Exploring the effectiveness of convolutional neural network with ensemble technique,” in Research Square, 2020. Accessed: Aug. 24, 2023. [Online]. Available: https://doi.org/10.21203/rs.2.21664/v1

C. Ren, D. K. Kim, and D. Jeong, “A Survey of Deep Learning in Agriculture: Techniques and Their Applications,” Journal of Information Processing Systems, vol. 16, no. 5, pp. 1015–1033, 2020, doi: 10.3745/JIPS.04.0187.

N. Ganatra and A. Patel, “Deep Learning Methods and Applications for Precision Agriculture,” in Lecture Notes in Networks and Systems, Springer Science and Business Media Deutschland GmbH, 2021, pp. 515–527. doi: 10.1007/978-981-15-7106-0_51.

L. Picek, M. Šulc, J. Matas, J. Heilmann-Clausen, T. S. Jeppesen, and E. Lind, “Automatic fungi recognition: Deep learning meets mycology,” Sensors, vol. 22, no. 633, pp. 1–22, Jan. 2022, doi: 10.3390/s22020633.

H. Wu et al., “CvT: Introducing convolutions to Vision Transformers,” in ArXiv, 2021, pp. 1–10. [Online]. Available: http://arxiv.org/abs/2103.15808

K. W. Seebold, “Blackleg & bacterial soft rot of potato,” 2014.

E. V. Rogozina, N. V. Mironenko, N. A. Chalaya, Y. Matsushita, and H. Yanagisawa, “Potato mosaic viruses which infect plants of tuber-bearing Solanum spp. growing in the VIR field gene bank,” Vavilovskii Zhurnal Genet Selektsii, vol. 23, no. 3, pp. 304–311, 2019, doi: 10.18699/VJ19.495.

S. Syifa Fitriyati, K. Hamzah Mutaqin, and T. Asmira Damayanti, “Taksasi kehilangan hasil oleh penyakit kerdil pada kentang di Jawa Tengah,” Jurnal Ilmu Pertanian Indonesia (JIPI), vol. 25, no. 2, pp. 205–212, 2020, doi: 10.18343/jipi.25.2.205.

U. Jayasinghe, “Potato leafroll virus PLRV,” 1988.

L. Soesanto, E. Mugiastuti, and D. R. F. Rahayuniati, “Inventarisasi dan identifikasi patogen tular-tanah pada pertanaman kentang di Kabupaten Purbalingga,” J. Hort, vol. 21, no. 3, pp. 254–264, 2011.

J. E. Van Der Waals, L. Korsten, and T. A. S. Aveling, “A review of early blight of potato,” Afr Plant Prot, vol. 7, no. 2, pp. 91–102, 2001.

J. W. Henfling, “Late Blight of Potato Phytophthora infestans,” 1987.

A. Ghosh, A. Sufian, F. Sultana, A. Chakrabarti, and D. De, “Fundamental Concepts of Convolutional Neural Network,” Intelligent Systems Reference Library, vol. 172, pp. 519–567, 2019, doi: 10.1007/978-3-030-32644-9_36.

D. Wodajo and S. Atnafu, “Deepfake video detection using convolutional vision transformer,” in arXiv, 2021, pp. 1–9. [Online]. Available: http://arxiv.org/abs/2102.11126

A. Vaswani et al., “Attention is all you need,” in ArXiv, California, 2017, pp. 1–15. [Online]. Available: http://arxiv.org/abs/1706.03762

A. Dosovitskiy et al., “An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale,” Oct. 2020, [Online]. Available: http://arxiv.org/abs/2010.11929

T. V. T Turay, “Toward Performing Image Classification and Object Detection With Convolutional Neural Networks in Autonomous Driving Systems: A Survey,” IEEE Access, 2022.

H. Wu et al., “CvT: Introducing Convolutions to Vision Transformers,” Proceedings of the IEEE/CVF international conference on computer vision, 2021.

R. Rombach, A. Blattmann, D. Lorenz, P. Esser, and B. Ommer, “High-resolution image synthesis with latent diffusion models,” in ArXiv, 2021, pp. 1–45. [Online]. Available: http://arxiv.org/abs/2112.10752

A. Watson, “Deep Learning Techniques for Super-Resolution in Video Games,” 2020, doi: https://doi.org/10.48550/arXiv.2012.09810.

P. Gholami and R. Xiao, “Diffusion Brush: A Latent Diffusion Model-based Editing Tool for AI-generated Images,” May 2023, [Online]. Available: http://arxiv.org/abs/2306.00219

X. Ying, “An Overview of Overfitting and its Solutions,” in Journal of Physics: Conference Series, Institute of Physics Publishing, Mar. 2019. doi: 10.1088/1742-6596/1168/2/022022.

A. Stöckl, “Evaluating a synthetic image dataset generated with Stable Diffusion,” in ArXiv, 2022. [Online]. Available: http://arxiv.org/abs/2211.01777

A. M. Lesmana, R. P. Fadhillah, and C. Rozikin, “Identifikasi Penyakit pada Citra Daun Kentang Menggunakan Convolutional Neural Network (CNN),” Jurnal Sains dan Informatika, vol. 8, no. 1, pp. 21–30, Jun. 2022, doi: 10.34128/jsi.v8i1.377.

M. H. Al-Adhaileh, A. Verma, T. H. H. Aldhyani, and D. Koundal, “Potato Blight Detection Using Fine-Tuned CNN Architecture,” Mathematics, vol. 11, no. 6, Mar. 2023, doi: 10.3390/math11061516.