Cotton Leaf Disease Detection

Abstract

In India, cotton is considered one of the most important cash crops, with many farmers cultivating it in large quantities. However, cotton plants are highly vulnerable to disorders caused bytemperature fluctuations, diseases, and pest attacks. Over the past few decades, these issues have led to significant losses in productivity. Globally, around 10% of cotton yield is lost annually due to leaf diseases, with India alone accounting for about 24% of the world's cotton-growing area. Annually, India experiences an estimated 18% loss in cotton production due to disease outbreaks, translating to financial losses of nearly 900,000 Indian rupees. Manual diagnosis of these diseases is difficult, as symptoms are often hard to detect with the naked eye, even for experts. This results in inaccurate identification and excessive pesticide usage, harming healthy crops. To address this, the proposed system leverages deep learning— specifically Convolutional Neural Networks (CNNs)—to automatically detect and diagnose cotton leaf diseases. The system focuses on extracting critical features such as color and texture from leaf images, using data collected from both primary sources and agricultural forums. Early detection through this method aims to enhance crop management, reduce yield loss, and promote precision agriculture.