Skip to content

New CSAFE Data Portal Promotes Greater Objectivity in Evidence Analysis

Forensic scientists and researchers can now directly implement CSAFE data into their laboratory work. The new CSAFE Data Portal advances the development and validation of forensic evidence analysis techniques with free, publically available data.

Utilize CSAFE datasets to test current analysis techniques, design new methods, and increase confidence in forensic evidence conclusions. Visit the CSAFE data portal to access datasets in shoeprints, digital evidence, bloodstains and glass, along with NIST datasets in ballistics and software reference library.

“Open source data for the forensic community is important because it allows solutions that use the data to be benchmarked by not only the team providing the solution but other community members. This enables the community to benefit from thorough testing, to find weaknesses and strengths of the tools and also of the data itself,” CSAFE researcher Jennifer Newman said.

Public access to reference data plays an important role in criminal investigations.  CSAFE researcher Susan Vanderplas explains, “Results used in court can be fully audited by both the prosecution and defense if there are test data sets available and the methods are also open-source.”

Promote greater objectivity and stronger statistical bases in forensic science by incorporating CSAFE datasets into evidence analysis procedures. Review a summary of each CSAFE dataset below, and visit the CSAFE Data Portal to learn more and download the data.

Database Snapshots:

StegoAppDB Forensic Image Database:

  • The first database consisting of almost 1 million images generated using mobile steganography apps, indicating that several software tools currently used for labeling an image as stego or not, cannot detect stego images. This database will allow the development of new tools to detect stego from mobile apps.

Bloodstain Pattern Data Set:

  • This is a data set of 61 blood spatter patterns scanned at high resolution, generated by controlled impact events corresponding to forensic beating situations. The main parameters of the study were the impact velocity and the distance between the blood source and the target sheet.

Shoe Outsole Impression Database:

  • A curated data set for examination of Randomly Acquired Characteristic (RAC) development, individual wear pattern characterization, and benchmarking of statistical methods for matching outsole prints.

Glass Elemental Database:

  • The first glass data source with 31 panes with 24 fragments measured in each pane, 20 times repeated measured in each fragment, allowing researchers to investigate within fragment variations, within pane variations, between pane variations and between manufacturer variations.

CSAFE also provides access to open-source software for forensic evidence analysis. Our automatic matching algorithm tools for a variety of evidence types are publicly available in our website Resources section. Do you have a question about CSAFE datasets or tools? Contact our team to learn how these resources can benefit your work.