Exhibit Hall Demonstration: Using Facial Recognition to Improve Detainee Tracking and Case Management
- Room: West Hall, Aisle 600
Tuesday, September 25, 2018: 4:30 PM - 4:45 PM
US Government detention facilities systems allow for biometric enrollment via fingerprints, face and iris; however, few—if any—detention facilities use biometrics to track movement of detainees throughout the detention facility. For example, detainees may have cell assignments, exercise schedules, work schedules, and medical needs requiring check-in and check-out to and from designated areas. Often, correlating detainees with their case information can be a manual or paper-based process. Facial matching can quickly identify the detainee and display relevant information in a Heads-Up Display, eliminating paper-based or manual verification processes and providing agents and officers with more complete information on detainees. This would allow agents and officers the ability to accurately know and track each prisoner’s location and the amount of time they are in an area; thereby enabling agents and officers to better and more safely manage the detention facility while also increasing the safety of detainees. Dev Technology will be demonstrating our approach to this solution at booth #115. By performing the enrollment of a detainee with all three modalities, the government can identify the detainee by any of the modalities for future verification. Using facial recognition is key to tracking detainees through a facility as to not impede or slow down the operations of detention system. Detainees present their face for quick capture by the camera for identification. For example, as each detainee leaves a cell location, a photo is captured upon exit with a high-resolution camera. Then as the detainee enters the assigned area, a photo is captured again verifying entry into the area. With the improvement in facial recognition over the past 10 years, it is now ready for this type of application. A distributed face gallery matching can be deployed throughout regions in the US Government in support for quicker face matching. The face image matching galleries can be divided by physical and geographic sections. From this we can gain accuracy from facial matching by limiting gallery sizes by tracking movement through physical locations and facilities.