Smart Farmers Equipment Rental and Pest Classification System: an Integrated Web Platform Using Machine Learning and Deep Learning

Small and marginal farmers face three recurring problems: the high cost of owning agricultural machinery, delayed or inaccurate identification of crop pests, and limited direct access to buyers for their produce. This paper presents the Smart Farmers Equipment Rental and Pest Classification System, an integrated web application that addresses these problems on a single platform. The system is developed using Python Flask, HTML, CSS and JavaScript with an SQLite3 database. It provides(i) an equipment rental module supported by a machine learning regression model that predicts rental price from equipment type, condition, usage duration and market demand; (ii) a pest classification module in which a Convolutional Neural Network (CNN) built with TensorFlow/Keras classifies uploaded crop images as harmful or non-harmful and returns precautions and preventive measures; and (iii) an agricultural marketplace that lets farmers list fruits and vegetables and lets registered companies browse and purchase them. Role-based access is provided for farmers, equipment owners, company/buyers and administrators. The paper describes the system architecture, UML and data flow design, module implementation, and testing strategy, with a target CNN test accuracy of at least 90%. The modular design allows the platform to be extended with additional crops, pest categories, equipment types, weather information and a mobile application