Harvest Guard Crop Disease Identification using Leaf Images

Abstract

The growing demands of people and animals around the world are driving a tremendous increase in the rate of plant and agricultural cultivation. Numerous cutting-edge methods have been developed by agricultural research to increase cultivation’s production rates. Protecting their plants from different diseases and insects however is a major concern for farmers. This lowers production and results in huge financial losses. Agricultural scientists have been putting forth endless effort for decades to create effective methods for promptly identifying plant illnesses and administering prompt treatment. Regretfully, conventional techniques for detecting anomalies in plants are labor-intensive, manual, and prone to delays. Numerous innovative technologies have been incorporated into the cultivation process to address these problems and improve the effectiveness of plant disease identification. In this work, we describe a model that uses machine learning, specifically deep learning, to detect leaf illnesses in crops including rice, tomato, and potato, The approach integrates computer science and engineering principles. Rice and potatoes are staple foods in Bangladesh, while tomatoes are a favorite among people of all ages. Despite being less common in the nation, the market for cherries and strawberries is expanding quickly. Farmers that grow rice, tomatoes, cherries, potatoes, and strawberries frequently suffer large losses due to a variety of illnesses and pest infestations. This paper uses image processing and a Convolutional Neural Network (CNN) model to train a dataset in order to reduce these losses and offer quick fixes. Our technique obtains a remarkable 88% accuracy rate. By improving crop and fruit production rates, decreasing plant illnesses, and successfully managing pest infestations, this research can help farmers everywhere. Keywords: