Image Audit Toolkit

Hey there!

Welcome to the Image Audit Toolkit! I'm working on this tool to help you analyze and check the integrity of your images. Whether you're a photographer, into digital forensics, or just want to reality check, this toolkit has some cool features to help you spot manipulations and dig into image metadata.

What It Can Do

• View Images:

  • Open single images or whole folders.
  • Easily flip through your images. ( initially started as a slideshow app )
  • Zoom in and out to get a closer look.

• EXIF Data Viewer:

  • See all the EXIF metadata for your images.

• Spot Manipulations:

  • Basic Checks: Look at EXIF data, image format, and size.
  • Advanced Detection: Use DCT and noise analysis to find possible edits.
  • Error Level Analysis (ELA): Detect weird compression levels.
  • Clone Detection: Find duplicated parts in an image.
  • Histogram Analysis: Check color histograms for any odd patterns.
  • more will be added as fast as I learn german for my test

• Generate Reports:

  • Create detailed audit reports in JSON, including matched camera info. ( may become unresponsive for a lot of time just wait it out )

• Match Cameras:

  • Find cameras that match your image resolutions using a built-in database.

Getting Started

1. Clone the Repo

git clone https://github.com/sh1d0wg1m3r/image-audit-toolkit.git
cd image-audit-toolkit

2. Set Up a Virtual Environment (Optional but Recommended)

python3 -m venv venv
source venv/bin/activate  # On Windows: venv\Scripts\activate

3. Install Dependencies

pip install opencv-python pillow tkinter matplotlib scipy piexif chardet pyyaml

( because I hate requirements.txt )

If you come from the future and the dependencies are not installable or have changed ( Python 3.13.1 (tags/v3.13.1:0671451, Dec 3 2024, 19:06:28) [MSC v.1942 64 bit (AMD64)] on win32 ) -- my current version

4. Prepare the Dataset

• Make sure there's a dataset folder in the root directory.
• This folder has YAML files with camera specs.

How to Use It

1. Launch the App

python image_audit.py

2. Navigate the Interface

• Menu Bar:

  • File: Open images or folders and exit the app.
  • Audit: Access tools like the EXIF viewer, manipulation checks, ELA, clone detection, histogram analysis, and report generation.
  • Dataset: Find cameras that match your image resolutions.

• Buttons:

  • Open Folder: Choose a folder with multiple images.
  • Open Image: Select a single image to analyze.
  • Previous/Next: Move through your loaded images.
  • Zoom In/Out: Get a closer or wider view of the image.

3. Create Reports

• After running your analyses, go to Audit > Generate Report.
• Pick where to save your JSON report.
• The report will include stuff like EXIF data, manipulation checks, ELA results, clone detection summaries, histogram analysis, and matched camera info.

Where the Data Comes From

The Image Audit Toolkit uses a camera database from the Open Product Data - Digital Cameras repo. This database has detailed specs of various digital cameras, helping the toolkit match your image resolutions to possible camera models. ( I don't see the original licence and have no idea what is permitted if you are the original creator please contact me here: zari@duck.com and I will resolve it as fast as possible )

What’s Broken

• Performance: Since it's early beta, some features like camera matching and advanced manipulation detection might be slow or glitchy with large datasets.
• Incomplete Features: Not everything is fully built yet. Your feedback can help decide what to tackle next.
• Error Handling: I've tried to handle errors smoothly, but unexpected issues might still pop up.

Want to Help?

• Help is welcome

License

GNU General Public License v3.0

Shoutouts

• Open Product Data - Digital Cameras: For the camera specs database. GitHub Repository
• OpenCV, Pillow, Tkinter, Matplotlib, SciPy, piexif, chardet, PyYAML: The awesome libraries that make this toolkit work.

---

This README is here to help you understand and use the Image Audit Toolkit. As I keep developing the project, I'll update this document with new features, improvements, and changes.