Rice Leaf Disease Detection Using CNN On Raspberry PI
Journal on Electronic and Automation Engineering · 01.05.2026 · 5 (2) · 10
DOI: 10.46632/jeae/5/2/2 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal on Electronic and Automation Engineering, online erschienen am 01.05.2026. Verfasst von Deepika P. Die Arbeit nennt 44 Literaturangaben.
Abstract (Originalsprache)
Rice (Oryza sativa) is the primary caloric staple for over 3.5 billion people worldwide, yet annual yields suffer losses of 10–30% due to devastating leaf diseases. This paper presents a complete end-to-end automated rice leaf disease detection and classification system utilizing a custom Convolutional Neural Network (CNN) deployed on a Raspberry Pi 4 Model B edge computing platform. The system captures rice leaf images in real time via a USB camera, preprocesses them using OpenCV, and classifies them into four categories: Healthy, Leaf Blast (Magnaporthe oryzae), Brown Spot (Cochliobolus miyabeanus), and Bacterial Leaf Blight (Xanthomonas oryzae pv. oryzae). The CNN model was trained on a balanced dataset of 4,500 annotated images over 50 epochs, achieving a training accuracy of 88.5% and a validation accuracy of 85.2% with an F1-score of 0.84. The trained Keras model is converted to TensorFlow Lite format with INT8 post-training quantization, reducing model size by 4× and enabling average inference at 120 ms per frame ( 8–10 FPS) on the Raspberry Pi. Classification results are communicated via a 16×2 I2C LCD display with audible alerts through a Piezo buzzer for disease-positive detections. The proposed system offers a portable, low-cost, and connectivity-independent diagnostic solution for smallholder farmers in resource-constrained agricultural environments.
Zitieren
APA
P, D. (2026). Rice Leaf Disease Detection Using CNN On Raspberry PI. Journal on Electronic and Automation Engineering, 5(2), 10. https://doi.org/10.46632/jeae/5/2/2
Vancouver
P D. Rice Leaf Disease Detection Using CNN On Raspberry PI. Journal on Electronic and Automation Engineering. 2026;5(2):10. doi:10.46632/jeae/5/2/2
BibTeX
@article{P2026,
author = {P, Deepika},
title = {Rice Leaf Disease Detection Using CNN On Raspberry PI},
journal = {Journal on Electronic and Automation Engineering},
year = {2026},
volume = {5},
number = {2},
pages = {10},
doi = {10.46632/jeae/5/2/2}
}
EUVIMED