Cotton leaf analysis based early plant disease detection using Machine Learning

Abstract

The cultivation of cotton plays vital role for daily agricultural practices, as it is extensively cultivated by a significant number of farmers. Early identification of crop diseases is essential for farmers to not only minimize crop losses but also to enhance product quality. The main types of diseases affecting cotton crops include premature leaf abscission and foliar infections, which often result in increased pesticide application. Recent advancements in machine learning algorithms applied to cotton leaf imagery have transformed the agricultural sector by facilitating automated disease detection. This study introduces an innovative methodology that employs machine learning and deep learning techniques for the early identification of cotton plant diseases through the analysis of leaf images. The efficacy of these algorithms is evaluated using metrics such as accuracy, precision, F1-Score, and recall.