​NICE Systems’ new video-analytics solution pinpoints suspects within large crowds

October 9, 2014 - 

NICE Systems has unveiled Suspect Search, a video-analytics solution that can scan through footage to create an accurate virtual police lineup of persons of interest from even very large crowds.

Suspect Search allows agencies to search for suspects by photo or a composite based on a witness description. A network of ordinary surveillance cameras are then used to search for matches based on clothing and general appearance, tracking each potential suspect on a live map.

Images, video and location data can then be shared between relevant authorities such as between a police department and a district attorney.

Suspect Search technology does not actually involve facial-recognition. The resolution and angle of surveillance cameras isn’t always suitable for facial recognition, and witnesses are prone to mis-remember facial characteristics more often than general details like what the suspect was wearing. The company also noted that these general characteristics can actually be more effective than facial-recognition technology.

It also reduces the burden on manually surveying video footage from different cameras and time frames by filtering out around 95 percent of irrelevant frames of video, making it more likely that individual attention is focused on footage that is relevant to an investigation and will produce results faster.

“Suspect Search is an important tool which allows for the practical application of video analytics in order to search for a specific person in real time,” NICE Security Group executive vice president Chris Wooten said in a statement.

This technology might come to the rescue when locating and retracing the movements of a suspect, lost child, or other person of interest.

​The Internet2 Technology Evaluation Center (ITEC) lab at Texas A&M is using Suspect Search in conjunction with other applications in testing its use in first-responder scenarios.​

The solution has ​also ​been used in four proof-of-concept deployments for the past four months.

Among them was a U.S. public transit system, which successfully used Suspect Search to track and identify an individual how stole an iPad based on a first-hand description of the culprit who was largely identified because of his orange shirt.

Suspect Search pricing is based around the number of video streams and how much work is required to install, properly configure it, and train staff.

Leave a Comment

comments

About Stephen Mayhew

Stephen Mayhew is the publisher and co-founder of Biometrics Research Group, Inc.. His experience includes a mix of entrepreneurship, brand development and publishing. Stephen attended Carleton University and lives in Toronto, Canada. Connect with Stephen on LinkindIn.