Bangladeshi rice & tea leaf disease detection using machine learning

Abstract

Bangladesh's economy is based primarily on agriculture, with the production of tea and rice being essential to maintaining livelihoods and food security. The persistent risk of illnesses in rice and tea plants, however, makes agricultural productivity and financial stability extremely difficult. The goal of this research project is to create reliable and effective disease detection models for Bangladeshi rice and tea leaves by utilizing machine learning, more especially deep learning and image analysis approaches as these two items are consumed in a huge amount on a daily basis from poor to rich. Given Bangladesh's strong agricultural economy, early disease identification in these crops is essential to maintaining both food security and economic stability. The research seeks to offer user-friendly software tools for rapid and precise disease diagnosis, encouraging precision agriculture, lowering pesticide usage, and improving food security. This will be accomplished by training machine learning models on a big datasets of photos of healthy and diseased plants. The project is projected to play a key role in the modernization and resilience of Bangladesh's agriculture by empowering smallholder farmers through technology transfer, promoting sustainable agricultural practices, and supporting government activities. Alongside it will help them to reduce the amount of their loss in farming.