International Research Journal of Education and Technology

Agriculture, being one of the primary sectors that sustains human life, faces numerous challenges, including the need for higher crop yields, better pest management, and efficient fertilizer use. With the advent of technology, these challenges can be addressed through datadriven solutions. This project introduces a comprehensive crop and pest prediction system built using Flask, which integrates advanced machine learning and deep learning techniques to provide actionable recommendations. The system employs a Convolutional Neural Network (CNN) for pest identification, enabling farmers to accurately identify pest species from images, reducing the need for manual inspection and improving pest management. Alongside, the Random Forest model processes agricultural data such as soil characteristics, climate conditions, and historical crop performance to recommend the most suitable crops, optimizing production based on location-specific factors.

 Furthermore, the system leverages NPK analysis to suggest the right fertilizer mix, ensuring the soil’s nutritional needs are met for each crop. The Flask framework acts as the backend of the system, efficiently handling user inputs and delivering predictions through an interactive web interface. The user-friendly platform allows farmers to input soil data, crop images, and other relevant details to receive personalized recommendations, helping them make informed decisions for sustainable farming practices. By combining these technologies, the system not only improves the productivity of agricultural practices but also promotes environmental sustainability by minimizing pesticide use and optimizing fertilizer consumption, ultimately contributing to a more efficient and eco- friendly agricultural ecosystem.

Index Terms - Smart Agriculture, Insect Detection, Pest Control, Machine Learning, Precision Farming, Image Processing

Crop Protection, Agricultural Technology.