https://github.com/killervardhan8/geolandml
The primary objective is to classify land areas based on their suitability for barren land, water, and vegetation. The approach involves advanced image processing and Neural Network (NN)-based classification to enhance land analysis accuracy, aiding in data-driven land use planning and resource management.
https://github.com/killervardhan8/geolandml
Last synced: 2 months ago
JSON representation
The primary objective is to classify land areas based on their suitability for barren land, water, and vegetation. The approach involves advanced image processing and Neural Network (NN)-based classification to enhance land analysis accuracy, aiding in data-driven land use planning and resource management.
- Host: GitHub
- URL: https://github.com/killervardhan8/geolandml
- Owner: KillerVardhan8
- Created: 2025-02-04T12:32:35.000Z (4 months ago)
- Default Branch: main
- Last Pushed: 2025-02-21T11:53:41.000Z (4 months ago)
- Last Synced: 2025-02-21T12:30:02.707Z (4 months ago)
- Language: Jupyter Notebook
- Homepage:
- Size: 5.86 KB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# Land Zoning Using Satellite Imagery and Machine Learning
## 📷 Tools Utilized
Python
TensorFlow
## 📌 Project Overview
This project focuses on developing a comprehensive system for **land zoning** using **satellite imagery** and **machine learning** techniques. The primary objective is to classify land areas based on their suitability for **barren land, water, and vegetation**. The approach involves advanced image processing and **Neural Network (NN)-based classification** to enhance land analysis accuracy, aiding in **data-driven land use planning** and **resource management**.
This project extends ongoing research work, which has been recognized and accepted by multiple international journal institutions for its innovative approach to land classification using satellite imagery and machine learning.
## 🚀 Key Highlights
- **Technology Stack**: TensorFlow, Python, Sentinel Hub, Kaggle Open-Source Dataset, OpenCV, Rasterio, NumPy, Matplotlib
- **Model Type**: Neural Network (NN) for multi-class classification
- **Methodology**: Data Acquisition → Preprocessing → Classification → Scoring → Visualization
## 📥 Data Collection
Satellite images are collected from:
- **Sentinel Hub**: Offers real-time satellite imagery.
- **Kaggle Open-Source Dataset**: A publicly available dataset used for training and evaluation. [Click here to access the Barren & Vegetaion dataset](https://www.kaggle.com/datasets/requiemonk/sentinel12-image-pairs-segregated-by-terrain) and [Click here to access the Water dataset](https://www.kaggle.com/datasets/franciscoescobar/satellite-images-of-water-bodies).
## 🛠 Data Preprocessing
The images were processed through the following steps:
- Extracted from respective file directories based on **barren land, water, and vegetation** classification.
- Resized to a uniform **256x256** pixel dimension for consistent input to the model.
## 🏗 Model Development
A **Neural Network (NN)** was developed using **TensorFlow**:
- Trained on **labeled images** with **data augmentation** for better accuracy.
- Uses **Neural Networks** for feature extraction.
- **Evaluation Metrics**: Accuracy, Loss Function Calculation.
## 📊 Scoring & Visualization
Each image is **scored on a scale of 0 to 1** for land suitability:
- **Barren Land**: Identified through surface texture analysis.
- **Vegetation**: Classified based on greenery levels.
- **Water**: Detected through edge analysis and color differentiation.
## 📑 Reporting
The final output includes:
- **Classified land maps** with **metric scores**.
- **Graphs & distribution plots** for visualization.
- **Comprehensive report** summarizing methodology, findings, and land use recommendations.
## 📚 Skills Demonstrated
✅ **Machine Learning & Deep Learning**: TensorFlow-based model training.
✅ **Data Processing**: Image segmentation, classification, and preprocessing.
✅ **Visualization**: Graphs, heatmaps, and suitability scores.
✅ **Project Management**: End-to-end ML pipeline implementation.
## ⚡ Future Enhancements
- Integrate **time-series analysis** for land use change detection.
- Deploy as a **web application** for real-time land assessment.
- Optimize the model with **Transfer Learning** for improved classification.
---
## 📜 License
This project is open-source under the **MIT License**.
## 🤝 Contributing
We welcome contributions! Feel free to open issues and submit PRs.
## 📬 Contact
For queries, reach out via **[LinkedIn](https://www.linkedin.com/in/suryavardhan8)** or through mail - **[email protected]**.