Scientific Image Manipulation Dataset
收藏arXiv2020-03-05 更新2024-08-06 收录
下载链接：
http://arxiv.org/abs/2001.07799v2
下载链接
链接失效反馈官方服务：
资源简介：
本研究创建了一个新的科学图像篡改检测数据集，主要包含western blot和显微镜图像，旨在模拟科学中的问题图像。数据集通过人工选择区域进行移除、拼接和润饰等操作，以创建不同类型的篡改图像。该数据集用于训练和测试基于噪声不一致性的图像篡改检测方法，以提高科学出版物中图像篡改的检测能力。
This study developed a novel scientific image forgery detection dataset, which primarily consists of western blot and microscopy images, and is designed to simulate problematic scientific images. The dataset is constructed by performing operations including region removal, splicing and retouching on manually selected areas to generate various types of forged images. This dataset is utilized to train and test noise inconsistency-based image forgery detection methods, so as to enhance the detection performance of image forgery in scientific publications.
提供机构：
Office of Research Integrity
创建时间：
2020-01-22
