Governmental infringement of our privacy has been a hot-button issue in recent months, with the public discovery and outcry over the U.S. National Security Agency’s mining of private citizen data. The government claims that by mining all of this information we are safer, although it is hard to evaluate those claims given the cloak of secrecy. Whether we like it or not, surveillance technologies are proliferating due to the dangerous world in which we live.
Last Sunday we saw one of the most scrutinized public events of the entire year: the National Football League’s Super Bowl. Unfortunately, the game wasn’t as entertaining as it could have been, but there was certainly much attention being paid to security for the event, both not he day of the game and in the days and weeks leading up to the game. Effective crowd surveillance is an important part of any large scale event such as the Super Bowl, but was perhaps even more important this year as the Super Bowl was played in New Jersey not far from New York City.
The security measures at MetLife Stadium for the Super Bowl included a 2.5-mile chain link fence surrounding the stadium, helicopter and boat patrols, thermal cameras and more to secure the stadium and prevent any threats that might try to use the surrounding marshlands to access the game. Video feeds recorded by cameras at the event will be processed by an artificial intelligence system capable of spotting unusual or suspicious behaviors.
We’re always interested in profiling new and interesting technologies, so we thought it might be interesting to take a look at some surveillance technologies that could be utilized to keep a venue like MetLife Stadium secure on a day when the entire world is watching.
Crowd Surveillance Technologies in General
Various governmental agencies have acknowledged that better methods of identifying risks in a crowd environment would be beneficial in preventing criminal activity. Security is always heightened at an event like the Super Bowl, but spectator security has come under increased focus during a year which saw the horrific events of the 2013 Boston Marathon. Thousands of people commuted to Super Bowl XLVIII through public transport, and MetLife Stadium can hold more than 80,000, making public security in that confined area a major concern.
Security concerns are important to address, but many worry about technology’s ability to access sensitive private information about individuals. The U.S. Department of Homeland Security, for example, has been developing a crowd-scanning technology known as Biometric Optical Surveillance System (BOSS) which uses video and facial recognition technology to identify individuals on a watch list. Privacy advocates interviewed for the above New York Times story argued that, although these systems haven’t been introduced on a wide scale, that discussions about limiting their use for privacy concerns should be taking place.
Some surveillance technologies are being increasingly used in metropolitan areas, but on-the-ground police officers still have the greatest presence in public streets. This 2012 case study about investigative surveillance in Los Angeles discusses networked camera systems capable of observing large crowds. These recent technologies give officers an idea of crowd size and movement trends for events, allowing them to disperse themselves more effectively among the crowd.
For more on this topic please see Facial Recognition Technology Raises Privacy Concerns.
Without further ado, here is a sampling of some recent and perhaps interesting security and surveillance technologies.
U.S. Patent No. 8,588,470, Methods and System for Improved License Plate Signature Matching by Similarity Learning on Synthetic Images
This patent protects a method of automatic license plate recognition that involves better optical character recognition technology. Assigned to the Xerox Corporation of Norwalk, CT, this system can create synthetic images of various license plate regions to build an image learning database. Once this system identifies the license plate, it can communicate with other traffic monitoring systems to determine if there is any information associated with that license plate.
This innovation is designed specifically to aid law enforcement who have the need to identify individual vehicles in video streams or still images. This patent, issued in November 2013, protects a method that improves the speed at which law enforcement officials can understand security risks by gaining access to police reports.
U.S. Patent Application No. 20130201330, Assisted Video Surveillance of Persons of Interest
This is an interesting invention that seems to be the type of real-life computer assisted surveillance tool that you might see while watching a movie or TV drama. The invention leverages computer systems and associated methods in order to determine whether the person captured on the video is a person of interest.
The methods support detection of moving persons in video frames, extraction of features of the detected persons and identification of which detected moving persons are likely to be matches to an individual that the authorities have deemed to be a person of interest. Identification of the likely matches can be determined using an attribute-based search, and/or using a specific person-based search.
The video surveillance system includes a graphical user interface for receiving search criteria from a user and displaying search results to a user, a processor for executing computer-executable instructions. The system also includes a storage for storing records including images of detected moving persons in video data, for storing features extracted from the records and for storing computer-executable code. The computer-executable code also includes computer-executable instructions for a module that receives a user input regarding a person of interest and calculates a match score for record based on a comparison of the user input regarding the person of interest and the extracted features for the record.
U.S. Patent No. 7,683,929, System and Method for Video Content Analysis-Based Detection, Surveillance and Alarm Management
This patent protects a surveillance system that can analyze video for the detection of an alarm situation. The system uses at least two video streams to provide surveillance of an area, and the video streams are analyzed by an object detection module to determine a series of events occurring to an object detected by the system. The system could create an alarm signal if certain event parameters are recognized by the object detection module
Patented by Nice Systems, Ltd., of Ra’Anana, Israel, this system is specifically designed for the detection and alarm management of terror-related threats. The invention addresses various drawbacks in human detection, including fatigue and lack of concentration, as well as surveillance video inabilities to detect suspicious actions. The system, patented in March 2010, has been developed to better detect situations where explosive devices are hidden in everyday objects.
U.S. Patent No. 8,345,939, System and Method for Fingerprint Recognition and Collection at Points-of-Sale and Points-of-Entry
This patent protects a fingerprint recognition and collection device capable of collecting prints from a plurality of contact areas. A transmitting unit on the device communicates the fingerprint data to a receiver which analyzes the fingerprints for law enforcement. The innovation is designed to be able to get prints for criminal identification while at a point-of-sale within a business or a point-of-entry for an event.
Fingerprints have long been used as a way of providing biometric identification of a criminal subject for law enforcement. However, this system, patented by solo inventor Theodosios Kountotsis of East Elmhurst, NY, seems to work by gathering the fingerprint data and then checking collected fingerprints against a fingerprint database. This facilitates in tracking down criminals while out in public, and one could envision this type of distributed system being employed within a stadium or at an event to alert police to the presence of individuals believed to pose a threat. Indeed, one of the images attached to this patent suggests that this system could be used by ticket scanners to collect fingerprints at sporting events.