Leveraging MobileNet, InceptionV3, and CropNet to Classify Cassava Plant Disease

GRADY MATTHIAS OKTAVIAN; HANDRI SANTOSO

doi:10.26760/mindjournal.v6i2.183-193

Leveraging MobileNet, InceptionV3, and CropNet to Classify Cassava Plant Disease

GRADY MATTHIAS OKTAVIAN, HANDRI SANTOSO

Sari

Abstrak

Singkong adalah tanaman yang tumbuh di sub-saharan Africa dan sering dijadikan sumber karbohidrat bagi manusia. Namun, tanaman singkong tersebut memiliki banyak penyakit yang dapat mengancam ketersediaan bahan makanan bagi jutaan orang. Terdapat banyak upaya dan penlitian yang menggunakan kecerdasan buatan dalam bentuk computer vision agar dapat membantu petani mendiagnosa apakah tanaman singkong mereka sehat atau tidak hanya dengan mengambil gambar dari daun tanaman mereka. Pada publikasi ini, penulis melatih tiga jaringan saraf artifisial yang bernama CropNet, MobileNet, dan InceptionV3 untuk dapat mengklasifikasikan gambar-gambar berupa penyakit tanaman singkong. Pembaruan yang dibawa penulis adalah dengan membuat sebuah algoritma gabungan yang mengkombinasikan hasil prediksi dari ketiga jaringan saraf artifisial yang telah dilatih guna mendapatkan hasil prediksi yang lebih akurat. Ternyata, metode penggabungan algoritma ini mampu memberikan nilai akurasi lebih tinggi 6.8% ketimbang nilai rata-rata akurasi dari masing-masing model.

Kata kunci: pembelajaran mesin, visi komputer, klasifikasi gambar, jaringan saraf artifisial, kecerdasan buatan, penyakit tanaman

Abstract

In sub-Saharan Africa, cassava is widely grown and considered to be a large source of carbohydrates for human food. However, the plant is plagued with diseases which can threaten food supply for millions of people. By using computer vision, researchers attempted to create an image classification model that can tell farmers whether the plant is sick or not by taking pictures of their leaves. In this short paper, the author attempts to train three Convolutional Neural Network: CropNet, MobileNet, and InceptionV3 that can classify cassava plant diseases based on visual data. As a novelty, the author creates an ensemble voting classifier that combines the prediction of CropNet, MobileNet, and InceptionV3 to create a better prediction. Turns out, creating an ensemble voting classifier enables us to achieve an accuracy score which is 6.8% higher than the average individual scores of each model.

Keywords: machine learning, computer vision, image classification, convolutional neural network, artificial intelligence, plant diseases

Teks Lengkap:

PDF (English)

Referensi

Amanda Ramcharan, K. B. (2017). Deep Learning for Image Based Cassava Disease Detection. Frontiers in Plant Science, 1852.

Andrew Howard, M. S.-C. (2019). Searching for MobileNetV3. Retrieved from https://arxiv.org/pdf/1905.02244.pdf

Begomoviruses: Occurrence and Management in Asia and Africa. (2017). Singapore: Springer Singapore.

C. Szegedy, V. V. (2016). Rethinking the Inception Architecture for Computer Vision. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, (pp. 2818-2826).

Cock, J. H. (2019). Cassava: New Potential For A Neglected Crop. United States: CRC Press.

Gutowski, A., Wohlmuth, K., Hassan, N. M., Alabi, R. A., Nour, S. S., & Knedlik, T. (2018). Science, Technology and Innovation Policies for Inclusive Growth in Africa. Austria: Lit Verlag.

H. Pan, Z. P. (2020). A New Image Recognition and Classification Method Combining Transfer Learning Algorithm and MobileNet Model for Welding Defects. IEEE Access, 119951-119960.

I. H. Witten, M. A. (2016). Data Mining: Practical Machine Learning Tools and Techniques. Netherlands: Elsevier Science.

Iffat Zafar, G. T. (2018). Hands-On Convolutional Neural Networks with TensorFlow: Solve Computer Vision Problems with Modeling in TensorFlow and Python. United Kingdom: Packt Publishing.

Krizhevsky, A., Sutskever, I., & Hinton, G. E. (2012). ImageNet Classification with Deep Convolutional Neural Networks. Advances in Neural Informatics Processing Systems 25 (pp. 1097-1105). Massachusetts: NIPS'12.

Kumar, R. V. (2019). Geminiviruses: Impact, Challenges and Approaches. Germany: Springer International Publishing.

Lebot, V. (2019). Tropical Roots and Tuber Crops: Casava, Sweet Potato, Yams and Aroids. United Kingdom: CABI.

Mahabub, A. (2020). A Robust Technique of Fake News Detection using Ensemble Voting Classifier and Comparison with Other Classifiers. Springer Applied Science, 525.

Makerere University AI Lab. (2021, 11 11). Cassava Leaf Disease Classification. Retrieved from Kaggle: https://www.kaggle.com/c/cassava-leaf-disease-classification

Millstein, F. (2020). Convolutional Neural Networks In Python: Beginner's Guide To Convolutional Neural Networks In Python. CreateSpace Independent Publishing Platform.

Roossinck, M. (2020). Virus: 101 Incredible Microbes from Coronavirus to Zika. United Kingdom: Ivy Press.

Thind, B. S. (2019). Phytopathogenic Bacteria and Plant Diseases. United Kingdom: CRC Press.

Vasilev, I. (2019). Advanced Deep Learning with Python: Design and Implement Advanced Next-generation AI Solutions Using TensorFlow and PyTorch. United Kingdom: Packt Publishing.

Wang, W., Li, Y., Zou, T., Wang, X., You, J., & Luo, Y. (2020). A Novel Image Classification Approach via Dense-MobileNet Models. Mobile Information Systems.

Zhu, F. (2015). Composition, Structure, Physicochemical Properties, and Modifications of Cassava Starch. Carbohydr Polym. doi:10.1016/j.carbpol.2014.10.063

DOI: https://doi.org/10.26760/mindjournal.v6i2.183-193