Slide 1: Title Slide
Script:
Good morning/afternoon everyone.
We are here to present our Stage II project titled "Agri Comprehensive Management Platform
using Machine Learning" under the guidance of Mrs. J. P. Kakad.
Our team includes:
Bhavarth Manik Khairnar
Ganesh Vijay Kolhe
Sanket Nandkishor Malode
Pankaj Vijay Sonwane
from Matoshri College of Engineering & Research Centre, Nashik under the AI&DS
Department.
Slide 2: Introduction
Script:
Our platform is designed to improve agriculture through AI.
It predicts crop yields based on data inputs, recommends fertilizers, detects diseases using
images, fetches real-time agricultural news, supports multi-language access, and connects
farmers with vendors to share real-time price updates.
Slide 3: Need
Script:
This system is needed because:
Agricultural productivity must improve.
Resources must be used efficiently.
Real-time disease monitoring is crucial.
Farmers need better access to market information.
Slide 4: Problem Definition
Script:
Farmers face problems like unpredictable yields, poor resource management, and lack of
decision-making tools.
We solve these using AI: crop prediction, disease detection, market forecasting, and resource
optimization – all in one platform.
Slide 5: Objectives
Script:
1. Study AI-based precision agriculture platforms.
2. Explore machine learning algorithms for predictions.
3. Design and implement the comprehensive platform.
4. Test and validate results.
Slide 6–7: Literature Survey
Script:
We studied existing platforms and research papers on:
Precision agriculture
Disease prediction using AI
Market linkage tools
This helped us identify suitable ML techniques.
Slide 8: System Architecture
Script:
This diagram shows how our system works:
Farmers input crop images, soil, and weather data.
Modules like crop prediction and disease detection process the inputs.
ML models (CNN & SVM) analyze data.
Outputs include suggestions, predictions, and news.
Slide 9: Details of Modules
Script:
The system has 5 main modules:
1. Crop Prediction – Suggests best crop to grow.
2. Fertilizer Recommendation – Advises on suitable fertilizers.
3. Disease Detection – Analyzes crop images for disease.
4. News API – Shares weather, schemes, and updates.
5. Vendor-Farmer Link – Ensures transparent pricing.
Slide 10: CNN Algorithm
Script:
CNN is used for disease detection:
1. Collect crop images.
2. Preprocess them.
3. Extract features using convolutional layers.
4. Use pooling, flattening, and classification.
5. Predict disease and recommend treatment.
Slide 11: SVM Algorithm
Script:
SVM is used for crop prediction:
1. Collect past data on soil and crops.
2. Extract numerical features.
3. Train SVM model.
4. Predict crop based on new inputs.
5. Recommend the best crop to plant.
Slide 12–13: Activity Diagrams
Script:
These activity diagrams show user flow:
Farmer logs in, submits data or image.
System processes the input through relevant module.
Output is displayed on the user interface.
Slide 14: Software/Hardware Requirements
Script:
Hardware: i3 or higher, 3 GB RAM, 250 GB HDD.
Software: Python, SQLite, Windows 11, VS Code or PyCharm.
Slide 15–16: Implementation
Script:
We implemented the system using:
Python for backend processing.
SQLite for data storage.
CNN and SVM models for prediction.
A user-friendly interface for farmers and vendors.
Slide 17–18: Testing
Script:
We tested:
Crop prediction using real datasets.
Disease detection using sample crop images.
Interface functionality for inputs and outputs.
Slide 19: Advantages
Script:
Benefits include:
Better crop planning
Early detection of diseases
Efficient use of resources
Market access
Real-time agri info
Slide 20: Disadvantages
Script:
Some limitations are:
Internet dependency
Poor image quality affects detection
Data accuracy is critical
Slide 21: Applications
Script:
Our platform can be applied for:
Crop selection
Disease monitoring
Fertilizer management
Connecting farmers to markets
Real-time information
Boosting yield and income
Slide 22: Conclusion
Script:
In the first semester, we developed a base model for crop prediction.
Next, we’ll expand to market insights and resource planning.
This platform will help farmers make smart, real-time decisions, reduce waste, and increase
profitability.
Slide 23: References
Script:
We referred to publications and research papers on AI in agriculture and blockchain integration.
These helped shape our platform’s functionality and direction.
Slide 24: Thank You
Script:
Thank you for your attention.
We welcome any questions or suggestions regarding our project.
