From New York Times:
In the city of Shenzhen (12.4 million people), the Chinese government is rolling out 20,000 surveillance cameras that will be smart cameras. Their goal is to use facial recognition to identify police suspects as well as to detect unusual activity. Additionally, there are 180,000 indoor and outdoor closed-circuit television cameras owned by businesses and government agencies that the police will have the right to integrate. (There are also smart cards given to citizens that contain a lot of information). They are also using cell phone signals to track the location of police officers (chicago is testing this idea).
The article states:
Security experts describe China’s plans as the world’s largest effort to meld cutting-edge computer technology with police work to track the activities of a population and fight crime. But they say the technology can be used to violate civil rights. . . . Both steps are officially aimed at fighting crime and developing better controls on an increasingly mobile population, including the nearly 10 million peasants who move to big cities each year. But they could also help the Communist Party retain power by maintaining tight controls on an increasingly prosperous population at a time when street protests are becoming more common.
rshah Facial Recognition, Other Cities
From IP Mailing List:
Christian Kuhtz sent in a German story from Spiegel on how a facial recognition system operating under every day conditions had a matching rate of 30%. Its another example of the severe limitations for facial recognition. Here is the message:
Apparently the BKA (German equivalent of the FBI) tested face recognition, spent 200K euros to test the system in a rail terminal in the city of Mainz and basically declared it worthless in terms of being an investigative tool. Apparently (per the article) this is the first public trial under normal, every day conditions (rather than having the conditions manipulated for a good showing) and only matched 30%. Even when the lighting was modified to be ideal, it only reached 60%. The BKA considers the system only useful if the success rate is very near 100%.
The sample size was approximately 23,000 travelers per day over a period of roughly 3 months. The targets were 200 commuters who had volunteered for the trial and travel through this rail terminal at least once per day.
BKA recommended that this is not a suitable system for surveillance and facial recognition to try to match suspects in a manhunt etc.
Update: Another blog, CyTrap Labs claims the reporter got it wrong the the actual false positive rate was 0.1% and BKA’s own president Jörg Ziercke said these tests were a success. Since I can’t read German, I can’t figure this out.
rshah Applications, Facial Recognition
From Newswise PR:
This is a PR on gait analysis research, something that hasn’t really been covered here. Its based on research by Rama Chellappa, a professor in the department of electrical and computer engineering at the University of Maryland. He is also combining his gait technology with facial recognition techniques. Here is a snippet:
When a person’s limbs are unencumbered, gait movements are symmetrical. Represented graphically, these movements form a twisted helical pattern resembling a “figure 8” called a double helical signature. Chellappa and his team call this pattern, which is slightly different in each individual, “human gait DNA.” An individual’s gait pattern is changed by any activity that changes the symmetry of the movements, such as carrying a package. By defining these signatures, the system can recognize unique patterns in human gait and automatically detect asymmetric movements like an individual walking with a hidden object tied to an ankle or wrist. Hidden objects secured to the body in ways that don’t affect movement symmetry, for example, a fanny pack that is belted around the waist, aren’t currently detected by this technology.
Chellappa and his team have integrated human gait DNA into a real-time video surveillance system and used it to study and locate pedestrians. The experimental results have demonstrated the effectiveness of the system under lighting changes, shadows, camera motion, various viewing angles, as well as significant obstacles in the cameras’ viewing angles. The results also indicate that the approach is superior to many existing methods in terms of accuracy and reliability.
His research team is also “teaching” their gait recognition system to identify individuals by their unique gait. This is a much more difficult task, since subjects may deliberately attempt to walk in an uncharacteristic manner in order to try and cheat the system and avoid detection. If the suspect is unaware of the surveillance system, their normal walking style is more easily identified.
rshah Applications, Facial Recognition
From Technology Review:
The uses for smart cameras are often focused on security, but there are many ways smart cameras can contribute to our society. This article focuses on research by Sheryl Brahman that uses facial recognition techniques to identify when babies are in pain. This is something that is very hard for humans to detect, but her preliminary research has shown that video analytics were more than 90% accurate.
rshah Applications, Facial Recognition
Recent Comments