Design and Implementation of Agri Farm Application Using ResNet Deep Learning Model

Abstract

This project introduces an innovative website designed to enhance agricultural productivity and support farmers in India, where a significant portion of the population relies on agriculture for their livelihood. Leveraging the potential of machine learning and deep learning technologies, the website offers three applications: crop recommendation, fertilizer recommendation, and plant disease prediction. The crop recommendation application allows users to input soil data, based on which the system suggests the most suitable crops to cultivate, thereby aiding farmers in making informed decisions about their farming practices. In the fertilizer recommendation application, users specify their soil data along with the crop they are cultivating. In India, where a sizable section of the population depends on agriculture for a living, this project provides a cutting-edge website created to support farmers and increase agricultural output. Crop recommendation, fertilizer recommendation, and plant disease prediction are the three applications offered by the website, which make use of the capabilities of machine learning and deep learning technology. The datasets that have been tested and trained have been taken from Kaggle. In existing, they only offer crops that are healthy or harmful. They only provided 96% accuracy.  Crop disease treatment fertilizer recommendations are not made for that particular plant. They have no idea what to do next.  It will indicate whether the plant is healthy or unwell. It will indicate the cause of any illness in the plant. Why the plant was impacted and offers the plant a remedy. Additionally, it recommends fertilizer based on the plant's needs we have achieved 99% accuracy.