Detection Of Oryza Sativa (Rice) Leaf Diseases Using Image Processing And Deep Learning With Googlenet

Abstract

Oryza Sativa, a primary staple crop across the globe, particularly in Asia, is vulnerable to various foliar diseases that adversely impact yield and agricultural sustainability. This research introduces a deep learning and image processing-based framework to detect and classify rice leaf diseases using real-world image datasets. A custom dataset was created from field-sourced samples, comprising diverse disease categories under varying environmental conditions. The study utilizes GoogleNet, implemented via the PyTorch framework, by using Preprocessing techniques that features the extraction and classification to enhance the superiority of image and learning efficiency. The intended system achieved high accuracy in disease detection, signifying its possibility to serve as a strong resource for primary diagnosis in agricultural settings. The results emphasize the position of integrating deep learning (DL) architectures with targeted image processing to improve the consistency of automated crop disease identification.