Table of contents
🤔 Problem Statement
In response to the growing demand for sustainable agricultural practices and the need for advanced decision support tools, our team initiated the Krishikala project to empower farmers with a comprehensive software tool. The goal was to aid farmers in making informed decisions about crop selection, fertilizer usage based on NPK values, and early diagnosis of plant diseases through image recognition.
📝 Solution Developed
My primary responsibility in the Krishikala project was to develop machine learning models for fertilizer prediction, crop recommendation, and image-based disease diagnosis. The objective was to enhance the decision-making process for farmers, enabling them to optimize agricultural practices for improved yield and sustainability.
I took the lead in designing and implementing machine learning algorithms for fertilizer prediction, considering the crucial NPK values. This involved extensive data analysis and model training using Python, scikit-learn, and TensorFlow. Simultaneously, I crafted a crop prediction model by incorporating environmental factors such as soil type, climate, and historical data. Additionally, I integrated a convolutional neural network (CNN) to enable image-based disease diagnosis, leveraging Python and image recognition technologies.
🚀 Results
The fertilizer prediction model achieved a 15% improvement in accuracy, leading to more precise recommendations and optimized resource utilization for farmers. The crop prediction model increased accuracy by 20%, empowering farmers to make informed decisions for maximum yield and profitability. The image-based disease diagnosis feature, with an 85% accuracy rate, provided an effective tool for early detection and prevention of plant diseases, contributing to sustainable agriculture practices.
All the project files and code for this project can be found right here.
If you made it this far, here's a virtual cookie for you! 🍪