Abstract: This paper introduces and documents a novel image database specifically built for the purpose of development and bench-marking of camera-based digital forensic techniques. More than 14,000 images of various indoor and outdoor scenes have been acquired under controlled and thus widely comparable conditions from altogether 73 digital cameras. The cameras were drawn from only 25 different models to ensure that device-specific and model-specific characteristics can be disentangled and studied separately, as validated with results in this paper. In addition, auxiliary images for the estimation of device-specific sensor noise pattern were collected for each camera. Another subset of images to study model-specific JPEG compression algorithms has been compiled for each model. The 'Dresden Image Database' will be made freely available for scientific purposes when this accompanying paper is presented. The database is intended to become a useful resource for researchers and forensic investigators. Using a standard database as a benchmark not only makes results more comparable and reproducible, but it is also more economical and avoids potential copyright and privacy issues that go along with self-sampled benchmark sets from public photo communities on the Internet.

@inproceedings{gloe_bohme_acm10,
  url          = {http://portal.acm.org/citation.cfm?id=1774088.1774427},
  booktitle    = {ACM Symposium on Applied Computing},
  author       = {Thomas Gloe and Rainer B\"{o}hme},
  volume       = {2},
  location     = {Sierre, Switzerland},
  year         = {2010},
  title        = {The '{Dresden Image Database}' for benchmarking digital image forensics},
  pages        = {1584--1590},
}