NT Police cracks down on crime with facial recognition technology

Northern Territory Police have partnered with NEC Australia to implement facial recognition technology to help fight crime.

The Northern Territory government has deployed NEC's NeoFace Reveal solution, following the trial of the technology in early 2015. The system enables NT Police to search through their database of photos, closed-circuit television footage, and videos taken from body-worn cameras, drones, and phones to identify persons of interest or missing persons.

There are currently 190 cameras in the network monitored by the police department's CCTV unit, in addition to the recently deployed mobile CCTV units that can be moved on-demand to 'hot spots' and major public events. So far, 100,000 images have been copied into the system from existing police databases.

Police Minister Peter Chandler said like fingerprinting, facial recognition is a form of identification that allows a computer to quickly match similar faces based on facial features.

He added the footage or images captured on CCTV footage are submitted to NT Police's facial recognition team who then load it into the facial recognition system for analysis and comparison with existing images in the database.

During the trial phase, NT Police were able to identify over 300 individuals using the solution.

The broader roll out of the application will complement the NT Police investment in mobile technology across handheld devices and image capture equipment. The state government recently issued 1,330 iPads to police officers and installed satellite communications in 51 police vehicles in remote locations.

"The technology is helping reduce investigation times by enabling investigators to quickly identify or rule out suspects soon after a crime has been committed," Chandler said.

Last week, the Australian Attorney-General's Department announced it will be introducing the National Facial Biometric Matching Capability to allow Australian law-enforcement agencies to share facial images amongst themselves.

Expected to be up and running by mid-2016, the system will support a one-to-one matching functionality to confirm identities of known individuals. The department said will allow agencies to share still images, and ruled out the addition of directly feeding licence plate cameras or closed-circuit TV into the system. However, stills from such cameras could be used.

"The capability is designed to replace existing manual, ad hoc facial image sharing arrangements between agencies; providing an efficient, secure, and accountable mechanism through which images can be shared and matched," the department said.