Deep Learning-Based Diagnosis of Diseases in Durian Plants

This project focuses on developing a deep learning-based system for detecting diseases in durian plants, targeting both leaves and fruits. For leaf disease detection, models such as DenseNet and ResNet are used to classify various conditions, including 'algal_leaf_spot', 'allocaridara_attack', 'healthy_leaf', 'leaf_blight', And 'phomopsis_leaf_spot'.. For fruit disease detection, YOLOv11 and YOLOv12 models are employed to identify issues like 'damage', 'fungus', and 'worm' infestations on durian fruits. A user-friendly web application, built with Flask, HTML, CSS, and JavaScript, enables users to register, log in, and upload images of durian leaves and fruits. The system processes these images and provides disease predictions. By utilizing deep learning and computer vision, this solution helps farmers and agricultural experts in early detection of plant diseases, enabling proactive measures to minimize crop damage and improve overall yield. The models are trained on annotated image datasets and optimized for accuracy and efficiency, offering a practical tool for enhancing durian farming practices and ensuring healthy crops.

Keywords:

Durian, Disease Detection, Deep Learning, DenseNet, ResNet, YOLOv11, YOLOv12, Leaf Disease, Fruit Disease, Flask, Web Application, Machine Learning, Image Classification, Agriculture, Disease Prevention, Crop Yield