Image-based nutrient deficiency detection in banana (Musa acuminata) leaves through support vector machine and neural network stacking
Paper Details
Image-based nutrient deficiency detection in banana (Musa acuminata) leaves through support vector machine and neural network stacking
Abstract
Banana production plays a vital role in the agricultural economy of tropical countries such as the Philippines, where maintaining crop health is essential for sustaining yield and ensuring farmer livelihoods. Nutrient deficiencies in banana leaves, often manifested through discoloration, leaf deformation, or necrosis, pose challenges to timely diagnosis, especially for smallholder farmers with limited access to laboratory analysis or expert assessment. This study aims to develop an image-based nutrient deficiency detection system using a stacking ensemble model that integrates Convolutional Neural Networks (CNN) and Support Vector Machines (SVM). A dataset consisting of approximately 5,000 banana leaf images exhibiting deficiencies in potassium, boron, magnesium, sulphur, iron, zinc, manganese, and calcium was preprocessed using the Inception v3 embedding technique in Orange Data Mining software. The stacking ensemble was constructed by combining CNN and SVM as base learners. Model evaluation followed the Knowledge Discovery in Databases (KDD) process. Results demonstrate strong predictive capability, with the ensemble achieving an accuracy of 0.874 and a precision of 0.874. Confusion matrix findings show high classification accuracy for healthy and manganese leaves, although zinc and magnesium classes exhibited higher misclassification rates due to overlapping visual symptoms. The study highlights the potential for integrating the model into mobile applications to support real-time, accessible nutrient deficiency detection for farmers and agricultural practitioners, contributing to precision agriculture and sustainable banana production.
Bharadwaj, Prakash KB, Kanagachidambaresan GR. 2021. Pattern recognition and machine learning. Programming with TensorFlow: solution for edge computing applications, 105–144.
Buet A, Galatro A, Ramos-Artuso F, Simontacchi M. 2019. Nitric oxide and plant mineral nutrition: current knowledge. Journal of Experimental Botany 70(17), 4461–4476.
FAOStat. 2025. Food and agriculture organization statistics. https://www.fao.org/faostat/en/#data/QCL/visualize
Fayyad U, Piatetsky-Shapiro G, Smyth P. 1996. The KDD process for extracting useful knowledge from volumes of data. Communications of the ACM 39(11), 27–34.
Gallagher J. 2023. What is an image embedding? https://blog.roboflow.com/what-is-an-image-embedding
Iparraguirre-Villanueva O, Guevara-Ponce V, Roque Paredes O, Sierra-Liñán F, Zapata-Paulini J, Cabanillas-Carbonell M. 2022. Convolutional neural networks with transfer learning for pneumonia detection. International Journal of Advanced Computer Science and Applications 13(9), 544–551.
Kumari P, Gaur SS, Tiwari RK. 2023. Banana and its by-products: A comprehensive review on its nutritional composition and pharmacological benefits. EFood 4(5), e110.
Lahav E, Israeli Y. 2019. Mineral deficiencies of banana. In: Handbook of diseases of banana, abacá and enset, 475–486.
Olufemi OS. 2024. Exploring banana production in Africa for food security and economic growth: a short review. Food Nutrition Chemistry 2(1). https://doi.org/10.18686/fnc.v2i1.125
PCAARRD. 2024. Banana: industry strategic science and technology program.
Pishchenko O. 2023. Ecological and economic security of the agricultural sector. Herald of Khmelnytskyi National University: Economic Sciences.
Rana GK, Singh Y, Mishra SP, Rahangdale HK. 2018. Potential use of banana and its by-products: a review. International Journal of Current Microbiology and Applied Sciences 7(6), 1827–1832. https://doi.org/10.20546/ijcmas.2018.706.218
Scott GJ. 2020. A review of root, tuber and banana crops in developing countries: past, present and future. International Journal of Food Science and Technology 56(3), 1093–1114. https://doi.org/10.1111/ijfs.14778
Sowmiya M, Krishnaveni S. 2022. Convolutional neural network based plant nutrient deficiency detection. ECS Transactions 107(1), 5625.
Sudhakar M, Priya RS. 2024. IncepV3Dense: deep ensemble based average learning strategy for identification of micro-nutrient deficiency in banana crop. IEEE Access 12, 73779–73790. https://doi.org/10.1109/ACCESS.2024.3405027
Sunitha P, Uma B, Channakeshava S, Babu S. 2023. A fully labelled image dataset of banana leaves deficient in nutrients. Data in Brief 48, 109155.
Talukder MSH, Sarkar AK. 2023. Nutrients deficiency diagnosis of rice crop by weighted average ensemble learning. Smart Agricultural Technology 4, 100155.
University of Regina. n.d. Overview of the KDD process. https://www2.cs.uregina.ca/~dbd/cs831/notes/kdd/1_kdd.html
Von Ryan PM, Lagarteja JG. 2020. Corzea: portable maize (Zea mays L.) nutrient deficiency identifier. International Journal of Scientific and Technology Research 9(2), 4210–4214.
Xie K, Cakmak I, Wang S, Zhang F, Guo S. 2021. Synergistic and antagonistic interactions between potassium and magnesium in higher plants. The Crop Journal 9(2), 249–256.
Niel Francis B. Casillano*, 2026. Image-based nutrient deficiency detection in banana (Musa acuminata) leaves through support vector machine and neural network stacking. Int. J. Biosci., 28(4), 51-58.
Copyright © 2026 by the Authors. This article is an open access article and distributed under the terms and conditions of the Creative Commons Attribution 4.0 (CC BY 4.0) license.