🕵️‍♂️ Secure LSB Steganography Project

A Python + Flask web application for Steganography Embedding and Detection using the Least Significant Bit (LSB) technique — now enhanced with AES-GCM encryption, RSA public-key wrapping, and random pixel selection for improved security and confidentiality.

Developed for the Multimedia Forensics Lab Final Evaluation, this project demonstrates how sensitive data can be hidden securely within digital images while maintaining high visual fidelity.

---

✨ Features

• 🧩 Secure Embedding with AES-GCM:
  Secret messages are encrypted using AES before embedding to ensure confidentiality.

• 🔐 RSA Key Wrapping:
  AES + randomization keys are encrypted using the receiver’s RSA public key, so only the receiver’s private key can decrypt them.

• 🎲 Random Pixel Selection:
  Pixel embedding positions are randomized using a secret key shared only between sender and receiver.

• 🖼️ Automatic JPEG-to-PNG Conversion:
  JPEG/JPG images are automatically converted to PNG to prevent data loss due to compression.

• 📊 Steganalysis Dashboard:
  Compare cover and stego images using Histogram, MSE, and PSNR metrics.

• 🧮 Dynamic Capacity Display:
  Shows how many characters can safely be hidden in the selected image before embedding.

• 🌐 Modern Web Frontend:
  Built with Flask + HTML/CSS for simplicity, clarity, and accessibility.

---

🛠️ Tech Stack

Component	Technology
Language	Python 3.10+
Framework	Flask
Libraries	OpenCV, NumPy, Pillow, Matplotlib, PyCryptodome
Frontend	HTML, CSS (custom minimalist UI)
Platform	Localhost (can deploy to Render / PythonAnywhere)

---

📂 Folder Structure

LSB_Steganography_Project/
│
├── backend/
│ ├── encode.py # LSB embedding logic (supports random pixel selection)
│ ├── decode.py # LSB extraction logic
│ ├── crypto_utils.py # AES-GCM + RSA key wrapping/unwrapping
│ ├── analysis.py # Histogram, PSNR, MSE
│ └── utils.py # Common helper functions
│
├── webapp/
│ ├── app.py # Flask server & routes
│ ├── templates/ # HTML pages (index, result, extract, analysis)
│ └── static/ # CSS & assets
│
├── input/ # Uploaded cover images
├── output/ # Generated stego images
├── key_generation.py # code for generating keys
├── requirements.txt
├── .gitignore
└── README.md

🚀 Run Locally

🧱 1. Clone Repository

git clone https://github.com/<your-username>/LSB_Steganography_Project.git
cd LSB_Steganography_Project

💻 2. Create Virtual Environment

python -m venv .venv
.venv\Scripts\activate    # (Windows)
source .venv/bin/activate # (Linux/Mac)

⚙️ 3. Install Dependencies

pip install -r requirements.txt

🌐 4. Run Flask App

cd webapp
python app.py

Then open your browser → http://127.0.0.1:5000

---

🧠 How It Works

Stage	Description
1️⃣ Embedding	Converts the secret message to binary, encrypts it using AES-GCM, and embeds it in the least significant bits of random pixel channels.
2️⃣ RSA Key Wrapping	The AES and randomization keys are wrapped (encrypted) using the receiver’s RSA public key and shared as a base64 “wrapped blob.”
3️⃣ Extraction	The receiver uploads the stego image, private key, and wrapped blob. The system unwraps and decrypts the hidden message.
4️⃣ Steganalysis	Computes MSE, PSNR, and histogram similarity to evaluate imperceptibility.

🔑 Secure Workflow

---

🧾 Receiver (Generate RSA Keys)

from Crypto.PublicKey import RSA

# Generate RSA key pair
key = RSA.generate(2048)

# Save private and public keys
open("receiver_private.pem", "wb").write(key.export_key())
open("receiver_public.pem", "wb").write(key.publickey().export_key())

same code is in key_generation.py, run it to get the files

• Share receiver_public.pem with the sender.
• Keep receiver_private.pem safe and private.

💻 Sender (Embed Securely)

• Upload a cover image (.png, .jpg, .jpeg, .bmp)
• Enter the secret message
• Upload the receiver’s public key file (receiver_public.pem)
• Click Embed Securely
• Download the stego image and copy the wrapped key blob

🕵️‍♂️ Receiver (Extract Message)

• Upload the stego image
• Upload your private key (.pem)
• Paste the wrapped key blob
• Click Extract & Decrypt

---

👥 Team Members

• Maithil Mishra
• Piyush Deshpande
• Shresth Raj

---

💬 Acknowledgements

This project was developed as part of the Multimedia Forensics coursework under guidance from the faculty at IIIT Kottayam.

---

🧩 Future Enhancements

• 🧠 Password-based AES key generation (PBKDF2)
• 📈 PDF report generation for steganalysis results
• 🧮 Advanced methods – DCT/DWT-based embedding
• 🎨 Enhanced UI with progress bar and visual comparison slider
• ☁️ Online sharing with receiver key auto-transfer

---

📜 License

This project is released under the MIT License – free to use, modify, and share for academic purposes.

---