A collection of algorithms to synthetically create scientific images for forensics and integrity analysis.
For an in-depth explanation of the algorithms and dataset, please refer to our research paper: Benchmarking Scientific Image Forgery Detectors: https://link.springer.com/article/10.1007/s11948-022-00391-4
The library implements various image tampering functions commonly found in scientific image manipulation:
- Image Duplication
- Retouching
- Cleaning
- Notebook: Tampering Simple Scientific Figures - Demonstrates how to apply each type of forgery.
The library also supports the creation of compound figures, mimicking the structure of images in scientific documents. Compound figures can have two types of forgeries:
- Intra-panel Forgeries: Forgeries within a single panel.
- Inter-panel Forgeries: Forgeries involving multiple panels.
Requirements
- Python 3.8
- Install the required modules from requirements.txt.
This dataset, created using the library, is designed for forensics and scientific integrity research.
- Dataset Link (Zenodo): https://zenodo.org/records/15095089 - Contains all images, metadata, and related files.
The dataset includes both simple and compound figures in the training and testing sets.
Simple Images
Compound figure
The dataset is distributed under the Attribution 4.0 International (CC BY 4.0) license.
If you use any content from this repository, please cite:
@article{cardenuto_2022,
title={Benchmarking scientific image forgery detectors},
volume={28}, DOI={10.1007/s11948-022-00391-4},
number={4},
journal={Science and Engineering Ethics},
author={Cardenuto, João P. and Rocha, Anderson}, year={2022}
}