surveillance
iSpy: long-distance shoulder-surfing
New Scientist reports on research on automated shoulder-surfing presented at the Conference on Computer and Communications Security in Chicago. A University of North Carolina at Chapel Hill research team, led by Dr. Jan-Michael Frahm, developed computer vision software that detects the visual feedback signals provided by touchscreen keyboards; e.g., the momentary magnification of each letter as it is typed. The news article suggests that it would be possible to to read the words typed by a user from 60 meters away, using HD video captured by an SLR camera, and from 3 meters using a cell phone camera. Click here for the paper.
Detecting people who return to the scene of the crime
A team of researchers from the University of Notre Dame are developing computer vision software to detect people who repeatedly visit the scene of a crime or attack. It is known that some types of criminals, such as arsonists, tend to visit the scene of their crime, and the U.S. military believes that bomb-makers visit explosion sites to gather feedback for improving their improvised explosive devices (IEDs). Jeremiah Barr, Patrick Flynn, and Kevin Bowyer are developing the Questionable Observer Detector (QuOD) biometrics system, which clusters faces of people present in the area in order to report which people appear more frequently than normal. More information is available from the Notre Dame press release.
U.S. funding research on 'shoulder surfing' attack mitigation
The U.S. National Science Foundation (NSF) has awarded $150,000 to Jan-Michael Frahm and Fabian Monrose of the University of North Carolina at Chapel Hill to investigate computer security attacks based on viewing mobile device screens and keyboards. (See "Thinkst warns of computer vision shoulder surfing attack on tablets" for an example of this kind of capability.) The funded research will characterize vulnerabilities and propose mitigation mechanisms.
Title: Automatic Reconstruction of Typed Input from Compromising Reflections
Abstract:
This project explores computer vision techniques aimed at exploiting compromising reflections associated with data input in mobile electronic devices such as smart phones. The ubiquity of these personal communication devices and their growing roles in data manipulation tasks, make unintended visual emanations an exploitable liability to data security. Nevertheless, there is still a gap in understanding of both the limitations of these techniques as well as the availability of effective mitigation mechanisms. It is the goal of this work to contribute to filling this conceptual gap.
The study builds upon recent state of the art techniques for automatic reconstruction of typed input from compromising reflections, comprising of robust keystroke event detection and classification mechanisms coupled to natural language processing modules. Such paradigm is both effective and amenable to low cost implementation in commodity devices. Based on these new developments, threat scenarios are no longer restricted to controlled scenarios using specialized equipment, but rather consist of highly flexible and possibly impromptu attacks. The project develops advanced cross-platform data input transcription prototypes used within a threat validation framework. This framework provides a characterization of both threat scenario operational limitations (e.g., imaging resolution, scene illumination, computational requirements) as well as the performance characteristics (e.g., robustness, accuracy) of the different vulnerability exploitation mechanisms. Moreover, the results of the analysis of diverse threat scenarios are being used to identify and develop appropriate mitigation mechanisms when possible.
QUT researchers develop soft biometric surveillance
Researchers from Queensland University of Technology (QUT), led by professor Clinton Fookes, have developed surveillance techniques for using soft biometrics. The techniques can use physical descriptions such as height, weight, skin, and hair color to identify people in video footage. Using radio-frequency tags, the team developed a system to improve airport baggage reconciliation, and to ensure the right passengers are carrying the right baggage. The accuracy and robustness of the system is currently being improved with testing in live environments. More details are available in an article in The Australian.
Computer vision startup SceneTap counts people in bars
Forbes reports that computer vision startup SceneTap has developed a system that uses face detection and gender recognition to automatically count the number of people of each gender in a bar. Users of SceneTap's app can be notified of the crowding and gender ratios of bars to help them decide which bar to go to. SceneTap, which has received angel investment funding, has placed cameras in 50 Chicago bars, and in 200 bars across the United States.
Video analytics detects potential riot incidents
Researchers at Kingston University, London, led by Dr James Orwell, are developing video analytic software to detect potential incidents such as riots, fights, loitering or robbery. The intent is to help the police analyze CCTV footage to find video segments relevant to potential police investigations. These new methods would allow video footage to be indexed, stored, and managed based on relevance, while reducing privacy infringements. The EU funded project will last 36 months. More information is available in a press release.
Facial recognition provides tools for police
Facial recognition software is being used by police in the UK to track down riot suspects. However, according to a New Scientist article, the CCTV footage suffers from poor quality. The article mentions a new system called Photoface, which uses multiple 2D face images to recreate a 3D model that produces more robust recognition under different lighting and viewing angle. Photoface is being developed by the University of the West of England, under Lyndon Smith.
More information is available in a news article by Tucone.com. Also reported in the article is research at Carnegie Mellon by Alessandro Acquisti, in which facial recognition software automatically links student volunteers directly to their Facebook profiles.
Recent news on Computer Vision Central:
Face recognition may be used in London 2012 Olympic Games
According to the Associated Press, facial recognition technology is being considered for London's 2012 Olympic Games. The technology is already being implemented along with other methods during the current riots in the UK. The face recognition technology is being improved, with the "promise in identifying people from high-quality, face-on shots taken off of surveillance photographs, mobile phones, passports or the Internet".
Los Angeles to shut down red light cameras
ABC News reports that one of the largest cities in the United States, Los Angeles, is shutting down its red light cameras that use computer vision software to detect drivers who disregard traffic lights. The issue at hand is not technical but legal: the courts ruled that a traffic fine disputed by the recipient of the citation could not be proven against that person without a live witness in court. The article states that nine U.S. states have banned the cameras, and others are placing limits on their use.
Related articles on Computer Vision Central:
Massachusetts revokes drivers' licenses based on face recognition false positives
IEEE Spectrum reports that the Massachusetts Registry of Motor Vehicles (RMV) uses face recognition software to identify people whose driver's license image appears similar to another person. This results in a revocation of the license. However, the article suggests that the only recourse to a false positive result is for both people to appear at a government office and prove that that no fraud has taken place. As the number of police and governmental bodies using face recognition software to flag similar faces grows, the number of people adversely affected by false positives is expected to grow. Massachusetts is currently being sued for damages and an injunction by one victim of a false positive identification by face recognition software.
Related articles on Computer Vision Central: