SIH2025FINALIIITBHOPALBTECHCSE5SEMS.pptx

Smart india hackathon internal round winning ppt

1. TITLE PAGE SMART INDIA HACKATHON 2025 Problem Statement ID - SIH25044 Problem Statement Title- Al-Powered Crop Yield Prediction and Optimization Theme- Agriculture, FoodTech & Rural Development PS Category- Software Team ID- 41 Team Name - Rudransh

2. IDEA TITLE 2 @SIH Idea submission- Template Rudransh Problem Discription : AI-driven yield prediction using historical data, weather, and soil health metrics. Smart recommendations for irrigation, fertilization, and pest control. Farmer-friendly app with regional language support & real-time alerts. Impact → Higher productivity, reduced costs, sustainable farming. Solution : Uses advanced AI & ML models to provide highly accurate crop yield predictions for different regions and crops , ensures better decision-making by integrating satellite imagery, detailed soil health metrics, and real-time weather data. Delivers field-specific recommendations for irrigation, fertilization, and pest management. Provides a farmer-friendly app with local language support, dashboards, and alerts. Ensures optimized resources, higher productivity, reduced risks, and sustainable farming.

3. TECHNICAL APPROACH 3 @SIH Idea submission- Template Rudransh Tech Stack : AI/ML : Python, ScikitLearn , TensorFlow, Matplotlib, Seaborn,Recurrent Neural Network, LSTM, BERT Backend : MongoDB,Flask Frontend : ReactJS, Tailwind CSS Data : Google Earth Engine, Weather APIs Process : Fetch satellite data Preprocess & train AI models Detect anomalies in crop health Generate real-time alerts + dashboard view

4. FEASIBILITY AND VIABILITY Feasibility : Built on open-source APIs (satellite, weather, soil) and cloud infrastructure → keeps costs low. Highly scalable to support multiple crops, regions, and farmer networks. Compatible with mobile & web platforms, ensuring wide accessibility. Challenges : Internet gaps in rural areas limit real-time data access. Data accuracy & availability issues due to fragmented sources. Farmers’ limited digital literacy may hinder adoption. Solutions : Offline mobile mode with periodic data sync when internet is available. Partnerships with government agencies, NGOs, and agri -universities for authentic datasets. Farmer training programs and simple UI/UX in regional languages to boost adoption. Helps those using ancient methods by giving simple, mobile/web-based insights → making modern techniques easy to adopt without needing deep technical knowledge. 4 @SIH Idea submission- Template Rudransh

5. IMPACT AND BENEFITS 5 @SIH Idea submission- Template Rudransh Target Audience : Farmers, NGOs, government agri bodies. Impact : Promotes sustainable farming with efficient use of water, fertilizer & pesticides. Strengthens food security & stabilizes agri -market supply chains. Boosts crop productivity & reduces losses through AI-driven predictions. Benefits : 20% increase in crop yield (hypothetical) Reduced losses from pests/disease. Cost-effective & easy-to-use system. Wider Impact : Supports SDG 2: Zero Hunger. Aligned with Atmanirbhar Bharat .

6. RESEARCH AND REFERENCES 6 @SIH Idea submission- Template Rudransh Artificial intelligence in agriculture: Advancing crop monitoring and yield improvements https://www.sciencedirect.com/science/article/pii/S2666154325001334 AI For Remote Crop Sensing & Yield Optimization https://farmonaut.com/precision-farming/revolutionizing-precision-agriculture-how-ai- remote-sensing-are-optimizing-crop-yields-and-farm-profitability AI for Crop Yield Prediction: Future of Agriculture 2025 https://www.omdena.com/blog/ai-for-crop-yield-prediction-future-agriculture-2025 React image cropping tutorials and libraries: https://blog.logrocket.com/top-react-image-cropping-libraries/ Java smart farming backend guide: https://moldstud.com/articles/p-exploring-agriculture-technology-solutions-with-java Crop prediction system example with Java backend: https://github.com/COS301-SE-2024/Crop-Prediction-System