Watermark Detector and Cleaner

Detect and sanitize imperceptible digital watermarks and steganographic payloads across spatial and frequency domains. 100% offline & local processing.

✨ What's New

• Modern UI & Asynchronous Engine: Fully redesigned with CustomTkinter. Features True Dark/Light modes, frameless dialogs, and a non-blocking asynchronous multi-threaded backend.
• Advanced 4-Layer Detection:
  • LSB (Spatial): Scans the least significant bits.
  • FFT, DCT, DWT (Frequency): Analyzes geometric patterns in Fourier and Wavelet spectrums.
• Adaptive Clearance Modes: Aggressive (maximum removal), Balanced (recommended), Conservative (quality preservation).
• Privacy Engine:
  • Rand Name: Renames the output file with a cryptographically random 12-character alphanumeric string.
  • Rand Size: Invokes an imperceptible micro-crop (±2px) to neutralize resolution-dependent fingerprinting.
  • Force JPEG: Compresses lossless images to strip container-level steganography.

🛠️ Usage

Quick Start (No Installation)

1. Clone the repo and install dependencies:

git clone https://github.com/AndrewWangDev/remove_watermark.git
cd remove_watermark
pip install -r requirements.txt

2. Launch the app:

python src/main.py

3. Workflow: Browse for an input image -> Select your detection/privacy parameters -> Hit Detect Watermarks (to scan) or Clean & Save (to output the neutralized image).

📊 Scoring & Verification

The Detection Engine runs an empirical mathematical analysis across 4 frequency and spatial dimensions to compute anomaly scores and determine a global confidence probability:

--- Starting DETECT ---
Method:     | LSB       | FFT       | DCT       | DWT       | 
Score:      | 0.32      | 28.50     | 142.10    | 12.40     | 
Result:     | Clean     | Clean     | Clean     | Clean     | 
Conclusion: The image is clean / unlikely to be watermarked (0% probability).
            No significant watermark signatures detected.
--- DETECT Finished ---

• LSB (Spatial): Measures spatial transition probability and difference energy in lowest bit-planes. Natural images have smooth neighborhoods (< 0.45); pseudo-random or encrypted steganographic payloads converge towards Bernoulli white noise (~0.500).
• FFT / DCT / DWT: Identifies geometric spikes, mid-frequency AC clusters, and wavelet high-frequency anomalies via fully vectorized transforms.
• Global Probability: A deterministic, multi-feature weighted confidence model (Weighted Confidence Evaluation). If no watermarks are detected, the score deterministically resolves to 0%.

🧠 Core Algorithm Architecture

This application employs a dual-domain analysis to detect and sanitize steganographic artifacts:

    [ Original Suspicious Image ]
                  │
                  ▼
 ┌─────────────────────────────────┐
 │   Spatial Domain Analysis       │
 │   (Extracting LSB Bit-Planes)   │ ──► LSB Detection
 └─────────────────────────────────┘
                  │
       (FFT / DCT / DWT Transform)
                  │
                  ▼
 ┌─────────────────────────────────┐
 │  Frequency Domain Manipulation  │
 │  * Target High-Frequency Spikes │ ──► FFT/DCT/DWT Detection
 │  * Apply Low-Pass / Masking     │
 └─────────────────────────────────┘
                  │
       (Inverse Transformation)
                  │
                  ▼
      [ Sanitized Clean Image ]

🏗️ Tech Stack

• Backend Algorithms: OpenCV (cv2), Numpy, PyWavelets
• Frontend Engine: CustomTkinter (Tkinter Wrapper)
• Packaging: PyInstaller + Pillow

⚖️ Disclaimer & Educational Purpose

This software is developed strictly for academic research, digital image forensics, and privacy preservation (PETs).

• It demonstrates mathematical signal-filtering methodologies (Fourier, Cosine, Wavelet, and Spatial LSB analyses) and evaluates the robustness of digital steganography schemes.
• The authors and maintainers do NOT condone, support, or encourage the use of this tool to bypass lawful corporate security controls, violate Non-Disclosure Agreements (NDAs), infringe intellectual property rights, or commit unauthorized data exfiltration.
• Users are solely responsible for ensuring that their use of this software complies with all applicable local laws, organizational confidentiality policies, and terms of service.