REVIEW ON SMART HARVEST: UNLEASHING AI AND ML TECHNIQUES FOR DISEASE IDENTIFICATION IN AGRICULTURAL PLANTATIONS

Abstract

One of the ongoing issues for farmers is plant disease, which puts their livelihood and access to food at risk. Finding diseases in plants or crops is problematic since analyzing each crop in significant areas requires too much time, effort, labor, and knowledge. The concept of agriculture is wholly altered by intelligent farming, which also helps to improve the quality and output of food products. It also aids in efficiently using the labor force needed for production. It is imperative in such a situation to guarantee that the crops are healthy and free from diseases. This paper uses convolutional neural network architectures to review contemporary deep learning-based strategies for various plants. This article will examine several machine-learning and deep neural network technologies. These techniques are used to recognize plant diseases from the images of infected plants. We conducted a thorough analysis of the volume of papers covering various plant diseases and other plants and fruits, and we evaluated these papers following crucial criteria. These factors include the number of classes (diseases), pretreatment, segmentation approach, classification type, classification accuracy, and size of the picture data set. This study aims to review several Internet of Things (IoT), deep learning, machine learning, and artificial intelligence (AI) methods and models to provide disease detection solutions and find the most appropriate model to achieve higher accuracy and precision.