A Novel Deep Learning-Based Image Segmentation Approach for Accurate Plant Disease Detection

Plant diseases significantly impact agricultural productivity, leading to substantial economic losses worldwide. Early and accurate detection is essential for effective disease management; however, traditional manual inspection methods are time-consuming, subjective, and require expert knowledge. This paper presents a deep learning-based approach for automatic plant disease detection using advanced image segmentation techniques. The proposed system employs state-of-the-art models, including U-Net++, DeepLabV3, and Swin Transformer, to accurately identify and isolate diseased regions from plant leaf images. The input images are preprocessed using resizing, normalization, and data augmentation to enhance model generalization. The models are trained to generate precise binary masks representing infected regions. Performance evaluation is conducted using Dice Score and Intersection over Union (IoU). Experimental results demonstrate that U-Net++ outperforms the other models, achieving superior segmentation accuracy. Furthermore, the best-performing model is deployed using a Flask-based web application, enabling real-time user interaction. The proposed system provides an efficient, scalable, and user-friendly solution for automated plant disease detection, contributing to improved agricultural productivity and decision-making.