Introducing RAISE dataset
News: Flat-field images for the NikonD7000 camera are now available!
Such high-luminance, not saturated, uniform-content pictures are particularly indicated for PRNU extraction applications. Download link (~5 GB)
RAISE is a challenging real-world image dataset, primarily designed for the evaluation of digital forgery detection algorithms. It consists of 8156 high-resolution RAW images, uncompressed and guaranteed to be camera-native (i.e., never touched or processed). All the images have been collected from 4 photographers over a period of 3 years (2011- 2014), capturing different scenes and moments in over 80 places in Europe employing 3 different cameras.
The RAISE dataset is intended to become a powerful and shared resource for forensic researchers and investigators in order to allow for a fair comparison of all the existing and forthcoming algorithms. Since the need of a large and diverse dataset (with no copyright and privacy issues as commonly available images downloaded from the Web) is common to many other domains of investigation in multimedia, we expect the RAISE dataset to constitute a valuable resource also within other fields of research dealing with image analysis, processing and understanding.
RAISE dataset can be easily and fast downloaded together with accompanying metadata. The user has the possibility to choose to download either the entire RAISE dataset or a subset of it. RAISE dataset supplies all the images with tags: "outdoor", "indoor", "landscape", "nature", "people", "objects" and "buildings”. As an option, the user could also select a specific image category of interest. Metadata are also provided, together with the embedded annotations.
Sample images

In all documents and papers that report experimental results based on RAISE, a reference to this study should be included:
"D.-T. Dang-Nguyen, C. Pasquini, V. Conotter, G. Boato, RAISE – A Raw Images Dataset for Digital Image Forensics, ACM Multimedia Systems, Portland, Oregon, March 18-20, 2015" .