IDENTIFICATION OF BACTERIAL SPOT DISEASES ON PAPRIKA LEAVES USING CNN AND TRANSFER LEARNING

M. Ilhamsyah; Ultach Enri

doi:10.33480/pilar.v18i1.2755

M. Ilhamsyah ^(1*) University of Singaperbangsa
Ultach Enri ⁽²⁾ University of Singaperbangsa

(*) Corresponding Author

DOI: https://doi.org/10.33480/pilar.v18i1.2755

Keywords: Deep Learning, Transfer learning, Classification, Paprika, Bell Peppers

Abstract

Paprika, often called bell peppers, is a plant with the Latin name Capsicum annuum var. gross. Paprika in Indonesia has a high selling value, so the opportunity for cultivating the paprika plant itself is enormous. However, the cultivation of this plant cannot be separated from the threat of disease that can affect the yield of paprika. Bacterial spot is one of them, and it is a disease that is very dangerous for paprika plants because the disease infects all parts of the plant. In this case, early detection is needed to carry out appropriate treatment to minimize the effects caused by bacterial spots. Detection of bacterial spots on paprika can be done by direct observation or conducting laboratory tests, but this requires people who have the appropriate knowledge and experience. Based on the above problems, the identification system can be an option in identifying bacterial spot disease in paprika. This research chose the Convolutional Neural Network (CNN) algorithm in the identification system. Because CNN is one of the algorithms that can receive output in the form of an image which is very suitable for the case of bacterial spots on peppers, this research dataset is divided into healthy leaves and leaves infected with bacterial spots. In this study, the implementation of CNN with transfer learning obtained results from a test accuracy of 90%, training accuracy 97% with a loss of 8.5%, validation accuracy of 97.5% with a loss of 6.9%.

Downloads

Download data is not yet available.

References

Andika, L. A., Pratiwi, H., & Handajani, S. S. (2019). Klasifikasi penyakit pneumonia menggunakan metode convolutional neural network dengan optimasi adaptive momentum. Indonesian Journal of Statistics and Its Applications, 3(3), 331–340.

adan Pusat Statistik Indonesia. (2020). Produksi Tanaman Sayuran 2020. https://www.bps.go.id/indicator/55/61/1/produksi-tanaman-sayuran.html

Biswas, T., Guan, Z., & Wu, F. (2018). An overview of the US bell pepper industry. EDIS, 2018(2).

Bogatzevska, N., Vancheva-Ebben, T., Vasileva, K., Kizheva, Y., & Moncheva, P. (2021). An overview of the diversity of pathogens causing bacterial spot on tomato and pepper in Bulgaria. Bulgarian Journal of Agricultural Science, 27(1), 137–146.

Food and Agriculture Organization of the United Nations Statistics. (2020). Countries by commodity. https://www.fao.org/faostat

Gowdra, N., Sinha, R., MacDonell, S., & Yan, W. (2020). Maximum Categorical Cross Entropy (MCCE): A noise-robust alternative loss function to mitigate racial bias in Convolutional Neural Networks (CNNs) by reducing overfitting.

Hakim, D. M., & Rainarli, E. (2019). Convolutional Neural Network untuk Pengenalan Citra Notasi Musik. Techno. Com, 18(3), 214–226.

Ilahiyah, S., & Nilogiri, A. (2018). Implementasi Deep Learning Pada Identifikasi Jenis Tumbuhan Berdasarkan Citra Daun Menggunakan Convolutional Neural Network. JUSTINDO (Jurnal Sistem Dan Teknologi Informasi Indonesia), 3(2), 49–56.

Rahman, M., & Miah, N. A. (2020). Bacterial Leaf Spot of Pepper. WVU Extension. https://extension.wvu.edu/

Reza, P. M. A., Syuhriatin, S., & Rahayu, S. M. (2021). Analisis Pertumbuhan Tanaman Paprika (Capsicum annuum var. grossum) Berdasarkan Pola Tanam. LOMBOK JOURNAL OF SCIENCE, 3(1), 23–32.

Rezende, E., Ruppert, G., Carvalho, T., Theophilo, A., Ramos, F., & de Geus, P. (2018). Malicious software classification using VGG16 deep neural network’s bottleneck features. In Information Technology-New Generations (pp. 51–59). Springer.

Rozaqi, A. J., Sunyoto, A., & rudyanto Arief, M. (2021). Deteksi Penyakit Pada Daun Kentang Menggunakan Pengolahan Citra dengan Metode Convolutional Neural Network. Creative Information Technology Journal, 8(1), 22–31.

Rusiecki, A. (2019). Trimmed categorical cross‐entropy for deep learning with label noise. Electronics Letters, 55(6), 319–320.

Simonyan, K., & Zisserman, A. (2018). Very deep convolutional networks for large-scale image recognition. arXiv. ArXiv Preprint ArXiv:1409.1556.

Suminar, J. R., Karolina, C. M., & Ratnasari, E. (2019). Lumbung Paprika Indonesia: Desa Pasirlangu Studi Kasus Komunikasi Pertanian di Desa Pasirlangu Kabupaten Bandung Barat sebagai Lumbung Pertanian Paprika di Indonesia. Studia Komunika: Jurnal Ilmu Komunikasi, 2(2), 33–42.

Umri, B. K., & Delica, V. (2021). Penerapan transfer learning pada convolutional neural networks dalam deteksi covid-19. JNANALOKA, 53–61.