Clustering-based Hemileia Vastatrix Disease Prediction in Coffee Leaf using Deep Belief Network

Abstract

Leaf Rusting Disease in coffee plants is a severe agricultural issue that affects the growth of plants and the production of crops. Coffee farming is a highly organized agricultural industry. Coffee cultivation is an extremely structured branch of agriculture. India is one of the world's top eight coffee producers, owing to optimal growing conditions. At the same time, coffee is susceptible to a wide range of illnesses, which have a substantial impact on crop yield. An effective design based on newly established deep learning (DL) model based on Deep Belief Networks (DBN), which is a multilayer generative graphical model for detecting leaf rust disease in coffee plantations. Disease detection in coffee plants is inaccurate with current approaches. The present strategies for detecting illness in coffee plants are ineffective. This research describes a high-accuracy method for detecting Coffee Leaf Rusting Disease using a Deep Belief Network. To find the damaged regions in the leaf photograph, the proposed DBN strategy includes the creation of an appropriate region-growing segmentation. Deep Belief Networks are a sort of neural network that can learn from complicated information unsupervised, and feature extraction is often conducted by RBMs in the network's bottom level. A wide range of tests was carried out to demonstrate the positive performance of the DBN method, and the outcomes ensured that the DBN model was improved.