File Format Identification
File Format Identification is the process of figuring out the format of a sequence of bytes. Operating systems typically do this by file extension or by embedded MIME information. Forensic applications need to identify file types by content.
- Written in C.
- Rules in /usr/share/file/magic and compiled at runtime.
- Powers the Unix “file” command, but you can also call the library directly from a C program.
Digital Preservation Efforts
PRONOM is a project of the National Archives of the United Kingdom to develop a registry of file types. A similar project was started by JSTOR and Harvard as the JSTOR/Harvard Object Validation Environment. Attempts are now underway to merge these two efforts in the Global Digital Format Registry and the Universal Digital Format Registry.
The UK National Archives developed the Digital Record Object Identification (DROID) tool, an "automatic file format identification tool." This tool is written in Java and can be downloaded from SourgeForge.
TrID - File Identifier
- Recognize over 10,000 file formats
- XML formats definitions, compiled to a single container
- New filetypes can be added in an automated way, simply scanning a group of files
- Win32, Linux/x86 & x86-64; closed source; free for non-commercial use
Forensic Innovations File Investigator TOOLS
- Proprietary, but free trial available.
- Available as consumer applications and OEM API.
- Identifies 3,000+ file types, using multiple methods to maintain high accuracy.
- Extracts metadata for many of the supported file types.
- Proprietary but free demo.
- Provides detection of password protected archives, some files of cryptographic programs, Pinch/Zeus binary reports, etc.
Toolsley File Identifier
- Runs in the browser
- HTML5/JS port of the UNIX "file" tool and libmagic
- Based on machine learning techniques, uses multiple file features
- Uses novel signatures computed from file format samples
- Identifications are linked to http://fileformats.archiveteam.org ontology
- Free (Apache v2.0 License)
- The Apache Tika™ toolkit detects and extracts metadata and text from over a thousand different file types
If you are working in the field of file format identification, please consider reporting the results of your algorithm with one of these publicly available data sets:
- NPS govdocs1m - a corpus of 1 million files that can be redistributed without concern of copyright or PII. Download from http://domex.nps.edu/corp/files/govdocs1/
- The NPS Disk Corpus - a corpus of realistic disk images that contain no PII. Information is at: http://digitalcorpora.org/?s=nps
- Comparison of Apache Tika, DROID and 'file' on 1TB, ~3 million files from govdocs1 and a sample of Common Crawl, April 2016. http://126.96.36.199/mimes/mime_comparisons.html
Current research papers on the file format identification problem. Most of these papers concern themselves with identifying file format of a few file sectors, rather than an entire file. Please note that this bibliography is in chronological order!
- M. McDaniel, Automatic File Type Detection Algorithm, Masters Thesis, James Madison University,2001
- Content Based File Type Detection Algorithms, M. McDaniel and M.H. Heydari, 36th Annual Hawaii International Conference on System Sciences (HICSS'03) - Track 9, 2003.
- Fileprints: identifying file types by n-gram analysis, LiWei-Jen, Wang Ke, Stolfo SJ, Herzog B.., Proceeding of the 2005 IEEE workshop on information assurance, 2005. (Presentation Slides) (PDF)
- D.J. Hickok, D.R.Lesniak, M.C. Rowe, File Type Detection Technology, 2005 Midwest Instruction and Computing Symposium.(PDF)
- K. Martin, N. Shahmehri File type identification of data fragments by their binary structure. , Proceedings of the IEEE workshop on information assurance, pp.140–147, 2006.(Presentation Slides)
- G.A. Hall, Sliding Window Measurement for File Type Identification, Computer Forensics and Intrusion Analysis Group, ManTech Security and Mission Assurance, 2006. (PDF)
- FORSIGS; Forensic Signature Analysis of the Hard Drive for Multimedia File Fingerprints, J. Haggerty and M. Taylor, IFIP TC11 International Information Security Conference, 2006, Sandton, South Africa.
- M.Karresand , N. Shahmehri, "Oscar -- Using Byte Pairs to Find File Type and Camera Make of Data Fragments," Annual Workshop on Digital Forensics and Incident Analysis, Pontypridd, Wales, UK, pp.85-94, Springer-Verlag, 2006.
- M. Karresand, N. Shahmehri, Oscar: File Type Identification of Binary Data in Disk Clusters and RAM Pages, Proceedings of IFIP International Information Security Conference: Security and Privacy in Dynamic Environments (SEC2006), Springer, ISBN 0-387-33405-x, pp.413-424, Karlstad, Sweden, May 2006.
- R.F. Erbacher and J. Mulholland, "Identification and Localization of Data Types within Large-Scale File Systems," Proceedings of the 2nd International Workshop on Systematic Approaches to Digital Forensic Engineering, Seattle, WA, April 2007.
- R.M. Harris, "Using Artificial Neural Networks for Forensic File Type Identification," Master's Thesis, Purdue University, May 2007. (PDF)
- S.J. Moody and R.F. Erbacher, SÁDI – Statistical Analysis for Data type Identification, 3rd International Workshop on Systematic Approaches to Digital Forensic Engineering, 2008.
- M.C. Amirani, M. Toorani, and A. Beheshti, A New Approach to Content-based File Type Detection, Proceedings of the 13th IEEE Symposium on Computers and Communications (ISCC'08), pp.1103-1108, July 2008. (PDF)
- V. Roussev, and S. Garfinkel, "File Classification Fragment-The Case for Specialized Approaches," Systematic Approaches to Digital Forensics Engineering (IEEE/SADFE 2009), Oakland, California. (PDF)
- I. Ahmed, K.-S. Lhee, H. Shin and M. Hong, On Improving the Accuracy and Performance of Content-based File Type Identification, Proceedings of the 14th Australasian Conference on Information Security and Privacy (ACISP 2009), pp.44-59, LNCS (Springer), Brisbane, Australia, July 2009.
- I. Ahmed, K.-s. Lhee, H. Shin and M. Hong, Fast File-type Identification, Proceedings of the 25th ACM Symposium on Applied Computing (ACM SAC 2010), ACM, Sierre, Switzerland, March 2010.
- I. Ahmed, K.-s. Lhee, H. Shin and M. Hong, Content-based File-type Identification Using Cosine Similarity and a Divide-and-Conquer Approach, IETE Technical Review, 27(6), 2010.
- I. Ahmed, K.-S. Lhee, H.-J. Shin, M.-P. Hong, Fast Content-Based File Type Identification, Proceedings of the 7th Annual IFIP WG 11.9 International Conference on Digital Forensics, Orlando, FL, USA, February, 2011
- C. Mattmann and J. Zitting, Tika in Action, Manning Publications. December, 2011.
- M.C. Amirani, M. Toorani, and S. Mihandoost, Feature‐based Type Identification of File Fragments, Journal of Security and Communications Networks, Vol.6, No.1, pp. 115–128, 2013. (PDF)