Rank One unveils forensic face biometrics suite for investigations at scale

Rank One Computing (ROC) has announced new biometric forensic face recognition suite ROC Explore to help law enforcement combat human trafficking, terrorism, and organized crime.

The new software integrates ROC’s lightweight software development kit (SDK) with next-generation user experience (UX) and design.

According to the company, ROC Explore can analyze terabytes of facial media on commercially available devices like laptops. This is important for forensics applications, where investigations into the crimes above often involve mountains of digital information that are too large for manual analysis, according to the announcement.

The company’s face recognition algorithms have been tested by the National Institute of Standards and Technology (NIST) with some impressive results. ROC’s facial recognition software’s accuracy was recently found to be 98.9 percent for Photo ID, 99.4 percent for controlled capture, and 96.8 percent for social media.

The average enrolling speed for the ROC SDK is 172 milliseconds, and the comparison speed 631 nanoseconds.

The software is also able to work with small templates of 260 bytes and has an extremely reduced memory footprint of just 41Mb.

Thanks to these features, the ROC Explore suite allows for effective search against tens of millions of facial templates in real-time.

The software then adjudicates matches using a one-to-one comparison tool and clusters ingested imagery into unique identities.

To facilitate monitoring of matches, the company says ROC Explore can also configure watchlist alerts from matches found between ingested images and previously processed galleries.

The suite is already available on Windows, Linux, and macOS, together with companion mobile apps for Android and iOS.

Article Topics

biometric identification  |  biometric matching  |  biometric software  |  biometrics  |  facial recognition  |  forensics  |  police  |  ROC  |  SDK