Each trial examined a critical aspect of performance for an automated tattoo recognition solution. In the identification trials, the MorphoTrak algorithm successfully found different instances of the same tattoo on the same subject, collected over time. The evaluation found MorphoTrak also excelled at finding a small region of interest within a larger tattoo, as well as determining whether an image contained a tattoo.
Tattoo images have traditionally been regarded as a soft biometric – that is, visual information that can be used to narrow down the range of candidates for identification and investigation, but cannot be used to explicitly identify an individual.
Law enforcement organizations have been collecting tattoo images as long as they have been collecting mugshots, and while mugshots can now be submitted for automated searches using facial recognition algorithms, tattoos are still categorized by text, in broad categories such as "Dragon" and "Skull."
The team that developed MorphoTrak’s tattoo recognition algorithm wants to help law enforcement make the transition from keyword search to automated search of tattoo images, much in the same way we now search for fingerprints and faces.
NIST’s Tatt-C evaluation is structured around problems that are designed to challenge the commercial and academic community in advancing research and development into automated image-based tattoo recognition technology. While some research and commercial capability is available, tattoo recognition is not a mature industry.
There is no common test data and use cases to evaluate and develop systems for next generation government applications. To address this shortcoming, the Tatt-C dataset was developed as an initial tattoo test corpus that addresses use cases derived from operational scenarios provided by the FBI’s Biometric Center of Excellence.
The Tatt-C dataset consists of still images of tattoos captured operationally by law enforcement agencies. The operational nature of this corpus imposes challenges on traditional image retrieval methodologies given the large variation in capture environment/process and tattoo content/quality.
