Lestari, Jelita (2026) Klasifikasi Penyakit Pada Batang Tanaman Jagung Menggunakan Algoritma Convolutional Neural Network (CNN). Other thesis, Universitas Islam Riau.

Text skripsi_213510559_watermark.pdf - Published Version Restricted to Registered users only Download (6MB) | Request a copy

Abstract

This research aims to develop a maize stalk disease classification system using a Convolutional Neural Network (CNN) based on the EfficientNet architecture as an effort to assist farmers in performing early disease detection quickly and accurately. The research workflow begins with collecting the “Maize Stalk Diseases Dataset” from Kaggle, consisting of 300 original images, which were expanded to 1,000 images through data augmentation to enrich data variation. The subsequent stages include image preprocessing, normalization, resizing, and splitting the dataset into training, validation, and testing sets. The method involves training the CNN model using three different learning rate scenarios—0.01, 0.001, and 0.0001—to determine the most optimal performance in classifying three categories: Gibberella stalk rot, Anthracnose stalk rot, and healthy maize stalk. Model performance evaluation was carried out using a confusion matrix, classification report, and accuracy-loss curves. The results show that the learning rate scenario of 0.0001 produced the most optimal performance with an accuracy of 96% and no signs of overfitting, making it the selected final model. The trained model was then integrated into a web-based application using Flask, enabling users to upload maize stalk images and obtain automatic classification results. In conclusion, the developed system has proven to be effective and can serve as a practical tool for detecting maize stalk diseases, thereby improving field identification efficiency and minimizing potential yield losses.

Actions (login required)

View Item