Design and Analysis of Automated Pesticide Sprayer with Detection of Grapes using Machine Learning

Abstract

in a developing country like India, agriculture is the most important occupation. By using advanced robotic technology to replace labors with intelligent robots, agriculture may increase its productivity and efficiency. The goal of the proposed project is to reduce the length of time that farmers are exposed to unhealthy chemicals by creating a low-cost, high-efficiency pesticide spraying robot. Instead of workers, a robot will apply fertilizers and insecticides. In agriculture, the proposed robot is utilized to spray pesticides and detect healthy and unhealthy grapes to spray at specific plants. The implementations related with this study are divided into four categories: sensing, control, detection and spraying modules. Grape diseases are the main reason behind the reduction in the quantity and quality of agricultural yield. Identifying and managing grape diseases presents significant challenges for farmers. Therefore, it is crucial to identify plant illnesses as soon as possible so that farmers may act appropriately and on time to prevent future losses. The study focuses on a method for identifying grape condition (Healthy or Unhealthy) that involves image processing. In this article, an Android app that enables farmers to control the pesticide sprayer and a Machine Learning model is developed to identify plant illness by sending a picture of a grape to the system. The user's input image is processed in many steps to identify the disease, and the results are delivered to the system to spray pesticide specifically.