A Modified Deep Learning And SVM-Based Model for Accurate Pomegranate Disease Diagnosis

Abstract

Pomegranate is an important fruit crop in regions like Nashik, Maharashtra, but it is often affected by diseases such as bacterial blight and anthracnose, which reduce both yield and fruit quality. This research proposes a modified model that combines Deep Learning (CNN) and Support Vector Machine (SVM) to accurately detect and classify pomegranate fruit diseases at an early stage. The model uses Convolutional Neural Networks to extract important features like color, texture, and shape from fruit images taken with mobile phones or simple cameras. These features are then classified using SVM which gives accuracy result 85% and speed of detection is increased.The main aim of this system is to help local farmers who may not have access to advanced tools or expert knowledge. Once a disease is detected, the system will provide useful suggestions such as pesticide recommendations and preventive actions in a user-friendly format. This proposed model is designed to be low-cost, easy to use, and suitable for real farming conditions in Maharashtra. It supports early detection, reduces chemical usage, and helps in better crop management, ultimately promoting sustainable and smart farming for small and medium-scale farmers.