Image-based Secure Data Embedding and Compression Using Huffman Coding and Histogram Shifting

A comprehensive steganography system that securely embeds and compresses data within color images using Huffman coding for optimal compression and histogram shifting algorithms for invisible data hiding.

🎯 What This Project Does

This system allows you to:

• Hide secret text inside regular images without visible changes
• Compress data before hiding to fit more information
• Work with color images instead of just grayscale
• Analyze image quality after data embedding
• Check capacity before encoding to choose the best cover image

🚀 Key Features

Enhanced Data Hiding

• Multi-channel encoding: Uses all RGB channels instead of just one
• 3x capacity increase: Significantly more data can be hidden compared to grayscale methods
• Huffman compression: Secret data is compressed before embedding for efficiency
• Format flexibility: Works with various image formats (PNG recommended)

Security & Recovery

• Secure keys: Generates encryption keys (enc_data.pkl) for decoding
• Frequency tables: Acts as additional security layer for Huffman decoding
• Lossless recovery: Perfect reconstruction of hidden data

Quality Analysis

• PSNR calculation: Measures signal quality after encoding
• SSIM analysis: Structural similarity assessment
• MSE computation: Mean square error evaluation
• Capacity testing: Pre-encoding capacity analysis

📁 Project Structure

├── Code/                   # Core implementation
│   ├── huffman.py         # Text compression/decompression
│   ├── histo-shift.py     # Image steganography engine
│   └── requirements.txt   # Dependencies
├── Analysis/              # Quality assessment tools
│   ├── main.py           # Combined analysis runner
│   ├── psnr.py           # Peak Signal-to-Noise Ratio
│   ├── ssim.py           # Structural Similarity Index
│   └── pvd.py            # Mean Square Error
└── README.md

🛠️ Installation & Setup

Prerequisites

• Python 3.x
• pip package manager

Quick Install

# Clone the repository
git clone https://github.com/Sudeep72/Image-based-Secure-Data-Embedding-and-Compression-Using-Huffman-Coding-and-Histogram-Shifting.git
cd Image-based-SecureData-Embedding-and-Compression

# Install dependencies
pip install --upgrade pip
pip install -r requirements.txt

💻 Usage Guide

Step 1: Compress Your Secret Data

python huffman.py
# Choose option 1 (Encoding)
# Input: secret_message.txt
# Output: encoded_binary.txt + frequency_table.pkl

Step 2: Hide Data in Image

python histo-shift.py
# Choose option 1 (Encoding)
# Input: encoded_binary.txt + cover_image.png
# Output: encoded_image.png + enc_data.pkl

Step 3: Check Image Quality (Optional)

python main.py original_image.png encoded_image.png
# Output: PSNR, SSIM, and MSE values

To Recover Hidden Data:

# Step 1: Extract binary from image
python histo-shift.py
# Choose option 2 (Decoding)
# Input: encoded_image.png + enc_data.pkl
# Output: recovered_binary.txt

# Step 2: Decompress to original text
python huffman.py
# Choose option 2 (Decoding)  
# Input: recovered_binary.txt + frequency_table.pkl
# Output: recovered_message.txt

🔧 Tool Options

Huffman Encoder (huffman.py)

• Encoding: Compress text files into binary format
• Decoding: Recover original text from binary + frequency table

Histogram Shift Engine (histo-shift.py)

• Encoding: Embed binary data into cover images
• Decoding: Extract binary data from encoded images
• Capacity: Check maximum data capacity of any image

Quality Analyzer (main.py)

• PSNR: Peak Signal-to-Noise Ratio measurement
• SSIM: Structural Similarity Index calculation
• MSE: Mean Square Error analysis

📊 Performance Benefits

Feature	Original Method	Enhanced Method
Color Support	Grayscale only	Full RGB
Data Capacity	Single channel	Triple capacity
Compression	None	Huffman encoding
Quality Analysis	Basic	Comprehensive
Security	Basic hiding	Multi-layer keys

📈 Example Results

Capacity Test Output:

Testing image capacity...
cover1.png: 1,245,680 bits
cover2.png: 2,891,456 bits  
cover3.png: 856,320 bits

Quality Analysis Output:

Image Quality Analysis:
PSNR: 48.32 dB (Excellent)
SSIM: 0.9847 (Near Perfect)
MSE: 0.0023 (Very Low)

🔮 Future Enhancements

• Web-based user interface
• Advanced encryption integration
• Support for video steganography
• Real-time capacity optimization
• Batch processing capabilities

📝 Technical Notes

• Recommended format: PNG for optimal capacity and quality
• Key files: Always keep enc_data.pkl and frequency_table.pkl safe
• Image size: Larger images = higher data capacity
• Compression ratio: Varies based on text content repetition

🤝 Contributing

Feel free to fork this project and submit pull requests for improvements!

📄 License

This project is open source and available under the MIT License.

---

Project: Image-based Secure Data Embedding and Compression Using Huffman Coding and Histogram Shifting
Technology Stack: Python, OpenCV, NumPy, Matplotlib
Algorithms: Huffman Coding, Histogram Shifting, RGB Channel Processing