Agricultural Advancement: Leveraging EfficientNetB4 for Enhanced Paddy Disease Classification

In agricultural settings, the threat of paddy diseases looms large, precipitating considerable yield losses and economic hardships for farmers. Timely and precise disease detection is crucial for implementing effective control measures. This study seeks to tackle this challenge by leveraging convolutional neural networks (CNNs) to devise an automated framework for paddy disease classification. With a focus on constructing a resilient model capable of accurately discerning various paddy diseases from images and harnessing transfer learning to optimize model performance, the project aims to develop a robust and efficient classification system. The methodology entails a concise overview of the methods employed, encompassing dataset preparation, model implementation, and evaluation metrics. Through rigorous evaluation, the study achieved an impressive accuracy of 98.2 % in paddy disease classification, signifying a significant advancement in automated disease detection in agriculture and holding promise for enhancing crop yield protection and ensuring food security.