Detecting mango leaf diseases using deep learning:

Abstract

Mango leaf diseases suffer from a multiple challenge in diagnosis due to different types of crops, changing agronomic disease indices and several environmental factors that influence them. It is still hard to detect them early since extant methods depend on data restricted by geography, thereby making them inefficient. This is essential for timely detection and control of such diseases in order to prevent huge financial losses that result among farmers. This research introduces an innovative strategy using deep learning and image processing technologies towards that end. Using CNN (Convolutional Neural Network) model, the study achieved a remarkably high 98.75% accuracy in separating healthy mango leaves from those with diseases. Rigorous tests were carried out over a range of leaf conditions to verify the effectiveness of the model. The technology has helped apply the algorithm onto 3000 leaf pictures which assist in identifying whether mango leaves are healthy or diseased when illness starts thus this is early disease detection. This innovation does not simply provide a timely and effective solution to enhance the management of diseases in mango farming, but also sets a benchmark for comparable advances in the broader agricultural milieu. The importance of this new technique in revolutionizing plant disease, identification and prevention highlights its significance in agricultural development. By overcoming geographical barriers and embracing state-of-the-art technology, this study begins a fresh chapter in agricultural practices; It is hoped that it will help farmers enable farmers to manage their diseases better and produce more sustainable crops