A web based application for swift and accurate cotton leaf disease classification using deep learning approach

Abstract

This report focuses on advancements in cotton leaf disease detection, aiming to address agricultural challenges and promote sustainable farming practices. Leveraging deep learning models such as VGG16 (93.80% accuracy), ResNet152 (90.80% accuracy), and InceptionV3 (96.40% accuracy), this research introduces a web-based tool for accurate disease identification in cotton plants. Manual methods often lead to inconsistencies, highlighting the need for automated solutions. The integration of convolutional neural networks (CNNs) into a user-friendly application enables precise disease detection, contributing to improved crop management and yield optimization. The project's objective is to overcome limitations in traditional methods by harnessing the power of deep learning, offering benefits such as reduced labor, minimized errors, and increased productivity. Through the development of an accessible application, farmers can make informed decisions, leading to enhanced crop health and sustainable agricultural practices. Additionally, the project aims to provide educational resources and establish a feedback mechanism for continuous improvement, fostering collaboration and knowledge sharing within the agricultural community. This research pioneer’s transformative technology in agriculture, specifically targeting cotton leaf disease detection, with far-reaching implications for sustainable farming practices.