Insect infestations are a major challenge in modern agriculture, causing extensive damage to crops and significantly impacting food production and economic stability. Early and accurate detection of insect pests is critical for effective pest control and sustainable crop management. This project presents an automated system for insect classification and detection in field crops using Convolutional Neural Networks (CNN), a class of deep learning models known for their effectiveness in image recognition tasks.

The system is designed as a web-based application that enables users to upload images of crops. It then analyzes the images to detect and classify insect species in real time. The model was trained on a diverse dataset of insect images, and various CNN architectures, were evaluated for their accuracy and efficiency. The best-performing model achieved an accuracy of 92% in classifying insect species.

In addition to detection and classification, the system provides visual annotations of detected insects and offers educational content to help users understand and manage pest threats. The platform also allows performance comparison of different models, making it a valuable tool for both agricultural practitioners and researchers. This project demonstrates the potential of machine learning in advancing agricultural technologies and contributing to smarter, more sustainable farming practices.