Andi B. Kaswar, Yasser A. Djawad, Dyah D. Andayani, Jose A. Veloria, Oslan Jumadi
Rice farming land in Indonesia has been decreasing annually, affecting rice productivity. Optimal fertilizer application is crucial to maintain rice quality and yield. Previous studies focused only on nitrogen content measured by leaf color, while rice plant growth quality is determined by more than just nitrogen content. Therefore, this study proposes a classification model of rice plant health in the vegetative phase based on nitrogen content and leaf width, using artificial neural networks. The proposed model uses digital imagery and computer vision to classify rice plants into low, medium, and high health levels. The model includes image acquisition, quality improvement, segmentation, feature extraction, and classification using backpropagation neural networks. The proposed method achieved an average accuracy of 85.9% and a Misclassification Error of 14.1%. This research can assist farmers in identifying rice plant health levels for optimal fertilizer application. © 2026, Khon Kaen University,Research and Technology Transfer Affairs Division. All rights reserved.
Computer Engineering Department, Universitas Negeri Makassar, Indonesia; Electronics Engineering Department, Universitas Negeri Makassar, Indonesia; TEBRIO, N-620, Salamanca, Spain; Biology Department, Universitas Negeri Makassar, Indonesia
Research at a Glance
Register to unlock
Topics & SDG Alignment
Register to unlock
Collaboration
Register to unlock
Author Profile (Selected)
Register to unlock
References Overview
Register to unlock
Journal & Source
Register to unlock
Metadata & Integrity
Register to unlock