Plant Disease Detection using Machine Learning

This research explores the use of machine learning approaches for accurate identification and detection of major diseases in pulse crops, targeting improvements in early diagnosis and management of infections such as downy mildew, Ascochyta blight, rust, powdery mildew, and anthracnose. These diseases trigger diverse signalling pathways. Specific genes and proteins, are upregulated in response to pathogen invasion, enabling plants to mount targeted defence responses. Understanding these mechanisms highlights opportunities for enhancing disease resistance through breeding and genetic research in legume crops.This paper outlines a procedure for identification of diseases in pulse crops using modern image processing technologies and machine learning techniques. The objective is to build a system that can recognize the common ailments of pulses which would otherwise lead to a huge loss in case of the weeds not being controlled. The system uses a data set of diseased and healthy plants and applies CNN’s to classify and predict disease. The findings achieved very encouraging Level of accuracy thus suggesting further work could extend to practical use by farmers in assessing the health of the crops more conveniently.