May 08, 2008

Challenge of False Positives

[From The Key Challenge for Video Analytics : IP Video Market Info]

John Honovich points out the importance of false positives. A false positive arises when a smart camera system warns operators of a situation, when the situation is normal. This problem arises when cameras move out of laboratories into real world conditions where weather and external stimuli, such as birds, make video analytics difficult.

Here are some interesting snippets:

Making sure you only trigger when someone is violating is hard because there are so many factors that might set off the analytic besides a valid adversary. To a computer, rain, dust and snow can all generate a similar form to a human being. Quick changes in light or the movement of water (waves) can also generate such forms. A camera that shakes because of the wind or issues with the mounting or installation also can trigger such alerts. The hard part in such analytics is to make sure that these alerts can be eliminated. This is a key metric in testing and differentiating between analytics.

False positives drive up the cost of systems. . . . While it is better operationally to centrally manage alerts, if the system generates dozens or hundreds of false alerts per day, the costs can become prohibitive. Let's say an 'intelligent' camera generates 5 false alerts a day at the cost of $1 per alert (the unit cost to pay a monitor to assess). That's $5 per day, over $1,800 per year and about $10,000 for a projected 5 year lifecycle. If you dozens or hundreds of cameras, this hidden operational cost can be in the millions. And this is not theoretical. This is the feedback you will hear time and again from real world deployments. It's widely accepted that this is improving but it is still the major factor in duing your due diligence in analytics.

Posted by rshah at 01:21 PM | Comments (0)

April 25, 2008

Privacy Protected Surveillance Using Secure Visual Object Coding

[From Intergovworld.com More privacy-boosting technology begets more video surveillance]

Developed by Karl Martin and Kostas Plataniotis, researchers at the faculty of engineering, their secure visual object coding application uses cryptography techniques to encrypt "objects of interest" within video frames -B faces or other features that may be used to identify a person - and store them separately. In order to view the original complete image, a decryption key is needed to restore the object of interest.

The project web site by Karl Martin shows a number of pictures of the process at work. The idea here is to encrypt either faces or complete bodies as a way to protect privacy. For example, surveillance could be monitored in real-time but faces would be obscured. However, this is reversible if a decryption key is used.

The PET application is an add-on designed to work with a front-end video surveillance system that has object detection capabilities. "Facial detection has been around for a long time to varying degrees of performance. Many vendors already offer it in consumer applications like digital cameras so you can focus on faces when taking photos."

They are currently working on implementing their system with the Toronto Transit Commission (TTC).

Posted by rshah at 07:36 PM | Comments (0)

April 15, 2008

Sensors in Buildings

[From 'Big brother' buildings offer less invasive security - tech - 09 April 2008 - New Scientist Tech]

An interesting article on the use of small cheap motion detectors within buildings. They can be used as an alternative to cameras in some areas, because they are cheaper and less invasive from a privacy standpoint. It is suggested that this technique could be used for security monitoring.

Here is a nice video of the system, Ambient Intelligence for Better Buildings (link to Journal article)

Posted by rshah at 01:54 PM | Comments (2)

March 13, 2008

Thruvision's Terahertz Camera

[From New security camera promises to protect, not reveal your body parts | NetworkWorld.com Community]

Thruvision offers a camera that relies upon Terahertz electromagnetic waves. The promise of terahertz is that these waves emit from people, but pass through solid materials. By passively collecting them, it is possible to identify if someone is concealing an object. The above story has lots of links for more information.

Posted by rshah at 01:16 PM | Comments (0)

March 10, 2008

Facial Recognition at the DMV

[From Biometric technology catches on with DMVs but privacy concerns slow broader reach]

A nice article on the growing use of facial recognition. A couple of interesting points:

After a driver sits for a photo at the Illinois Secretary of State office to renew a license, officials use facial-recognition technology to give the resulting image a close look. First, state officials verify that the face matches the images portrayed on previous licenses issued under the driver’s name. The second, more extensive run-through determines if the same face appears on other Illinois driver’s licenses with different names.

Since starting the program in 1999, the state has uncovered more than 5,000 cases of multiple identity fraud, said Beth Langen, policy and program division administrator at the Illinois Secretary of State office. The state pays Digimarc Corp. about 25 cents per license for the service, she said. . . . About 40 percent of the nation’s drivers are set to undergo such facial-recognition database checks when they renew their licenses in 20 states. . . . .

Police departments, eager for more investigative tools, are pressing for access to the millions of photographs in the motor vehicle databases. A few states prohibit such sharing, but many allow it.

Posted by rshah at 12:52 PM | Comments (0)

February 26, 2008

Gait Analysis via Plantiga Technologies

[From :: welcome to plantiga ::]

Gait analysis and recognition is another useful application of smart camera systems. There are several problems with reliable gait analysis including the collection of data and that the choice of footwear can affect how people walk. Plantiga offers a new shoe that can provide more gait and movement information and then transmit it wirelessly. As a result, it offers a unique method of biometric identification. For more, see this page on their web site.

I have no idea how well this works, but it is a novel and interesting approach.

Posted by rshah at 01:13 PM | Comments (0)

February 18, 2008

Smoke Detection

Smoke and fire detection is another application for smart cameras. A nice introductory video by Fire Sentry is online at youtube. For more background, see these academic articles, 1 and 2. A quick search also found this a camera with embedded smoke detection capability by axonX, known as SigniFire.

Posted by rshah at 09:31 PM | Comments (0)

February 10, 2008

Facial Recognition Upate

I was looking up the latest on facial recognition and I found a few interesting sites and stories.

First, the Metro Nashville, TN school system is installing face-recognition technology, which is suppose to be operational by December 2007. Its a bit strange that this is installed in an elementary school, but I will be awaiting the results of the test.

Second, a facial recognition research group is online at http://www.face-rec.org/newsgroup/. There are lots of techie articles, but also an active google group (face-rec) discussing the issues.

Third, the National Institute of Standards and Technology has been evaluating and testing facial recognition technologies. The project is known as FRVT. Its too bad there haven't been other public tests for other features of smart cameras.

Posted by rshah at 01:26 PM | Comments (0)

DARPA Supporting Smart Cameras

[From DARPA Wants Supercharged Spy Cams | Danger Room from Wired.com]

DARPA is supporting various smart camera technologies in a number of ways. Here are snippets from the story:

Dynamic Multisensor Exploitation, or "DYME," aims to combine cameras, radar, and acoustic sensors to better find bad guys as they move through urban canyons, and along coastal waters.

The goal of "Building Labels for Urban Environments," or "BLUE" (you gotta love these acronyms), is to automatically label the structures seen in surveillance video.

The Combat Video Analysis Engine (no acronym, alas) would use "computer vision, machine learning and probabilistic models to detect and recognize complex threats and suspicious activities without identification of specific individuals."

Posted by rshah at 01:04 PM | Comments (0)

December 23, 2007

New Video Surveillance Cameras

[From New Video Surveillance Cameras - High-Tech Law Enforcement Spying Programs - Popular Mechanics]

Great article, here are some interesting snippets:

Liberty Island's video cameras all feed into a computer system. The park doesn't disclose details, but fully equipped, the system is capable of running software that analyzes the imagery and automatically alerts human overseers to any suspicious events. The software can spot when somebody abandons a bag or backpack. It has the ability to discern between ferryboats, which are allowed to approach the island, and private vessels, which are not. And it can count bodies, detecting if somebody is trying to stay on the island after closing, or assessing when people are grouped too tightly together, which might indicate a fight or gang activity. "

My heist had been condoned by Pedro Ramos, Pathmark's vice president of loss prevention, though he didn't know precisely when or where I was going to attempt it. The beer was identified by an object-recognition scanner at ankle level—a LaneHawk, manufactured by Evolution Robotics—which prompted the cashier's question. Overhead, a camera recorded the incident and an alert was triggered in Ramos's office miles away on Staten Island. He immediately pulled up digital video and later relayed what he saw. "You concealed a 12-pack of Coronas on the bottom of the cart by strategically placing newspaper circulars so as to obstruct the view of the cashier." (Using Storevision)

But examples abound. Take E-ZPass. Drivers signed up for the system to speed up toll collection. But 11 states now supply E-ZPass records—when and where a toll was paid, and by whom—in response to court orders in criminal cases. Seven of those states provide information in civil cases such as divorce, proving, for instance, that a husband who claimed he was at a meeting in Pennsylvania was actually heading to his lover's house in New Jersey. (New York divorce lawyer Jacalyn Barnett has called E-ZPass the "easy way to show you took the offramp to adultery.")

Posted by rshah at 09:47 PM | Comments (0)

November 25, 2007

Smart Camera Car

From ContraCostaTimes.com - L.A. tests array of police technology

LAPD's smart car or the Critical Incident Utility Vehicle is "equipped to sniff out stolen cars, recognize criminals, track fleeing suspects and conduct remote video surveillance in high-crime areas." In a six-month test, officers using ALPR found 200 "parked stolens" and 40 "rolling stolens". It also includes the following:

• A dart-gun-like projectile shooter that fires a global-positioning-system tracking device at a fleeing suspect's car. The electronic monitor clamps onto the rear of the car, allowing officers to discontinue the chase and safely follow the suspect using satellite tracking technology. Operated by compressed air, the "StarChase" system is mounted within the patrol car's front grille. Gomez said it can fire the projectile about 20 feet.

• Video receivers that can connect to closed-circuit TV cameras set up in high-crime areas. Officers responding to emergency calls in those areas can operate the cameras and observe the situation in real time before they arrive at the scene, Gomez said. So far, the cameras are set up around the Jordan Downs housing project in the Watts area, he said. Five patrol cars in that area also have the video capability.

• Specialized computers that allow suspects' fingerprints to be scanned and checked in the field. Wireless Bluetooth technology connects the cruiser's computer to a database that contains the fingerprints of those with outstanding arrest warrants.

• Facial recognition equipment that can identify known suspects. The system has been used in the department's Rampart Division and has resulted in more than 50 arrests, mostly for gang injunction warrants, Gomez said.

• In-car video cameras that record windshield views and activities in the patrol car's back seat. The system automatically uploads its images when the car returns to the police station. About 300 police cars in the department's South Bureau are being equipped with the camera gear.

  Gomez said the equipment adds as much as $100,000 to the standard $25,000 cost of a patrol car. Although the costs are expected to drop eventually, the timetable for citywide distribution of Smart Cars is uncertain, he said.

Posted by rshah at 09:09 PM | Comments (0)

November 20, 2007

Demystifying video analytics – is "intelligent" now a dirty word?

[From News - SourceSecurity.com]

An interesting position piece by ObjectVideo's Edward Troha defining intelligence. He argues that intelligent systems should only refer to systems using computer vision. As a result, ANPR and motion detection should not be considered intelligent systems.

I think Troha is pointing out an important technical difference in these systems, but he is obviously trying to position his technology in the marketpace. Still its a good argument and one that everyone should consider. Neverthess, I will continue following the developments of ANPR, because I consider them part of smart camera systems.

Posted by rshah at 07:48 AM | Comments (0)

November 17, 2007

Spray Can Sensors

From the San Jose Mercury News

A city near LA, Montebello, is spending $1 million on 120 cameras. 25 of them will include sensors that will detect the sound of paint spray. The technology costs $20,000 to $30,000 per camera. This is an effort to detect graffiti.

"So when somebody's spraying a wall or specific target, this sensor activates the camera ... then sends a signal to our dispatch ... dispatch zooms in and sends units to the area," said police Det. Ismael Navarro.

Posted by rshah at 09:03 AM | Comments (0)

October 21, 2007

ANPR Anecdote

From FOXNews.com:
ANPR or license plate recognition caught a bad guy and received very good publicity.
Here is what happened:

According to news accounts, Donald Bachmann stole a pickup truck in San Jose, ditched it at a 7-Eleven, and then stole a white Toyota sedan. The police were notified. Shortly thereafter, the 12-year-old girl was walking home with her 9-year-old sister. Police say Bachmann rammed the Toyota into the older girl intentionally, severely injuring her. Her little sister rushed to aid her, but Bachmann allegedly kidnapped the bleeding girl and drove away. The 9-year-old immediately reported the crime.

Less than 15 minutes later, police say Bachmann parked the car, punched his victim and tried to sexually assault her. Miraculously, she escaped and ran to a nearby home. The residents called the police. Bachmann drove a few miles and ditched the Toyota.

The police responded to the 7-Eleven, collected evidence from the stolen truck, and found Bachmann’s fingerprints inside.

Hours later, Officer Max Boyer was on patrol, with license-plate recognition technology mounted on his patrol car. It can scan the license plate of every vehicle it passes. The computer compares the plates to databases of stolen vehicles. Boyer drove by the Toyota and a computer voice in his car said, “Stolen car.” He stopped to investigate. Police found blood in the Toyota belonging to the 12-year-old girl.

Posted by rshah at 07:00 PM | Comments (0) | TrackBack

October 15, 2007

Airport millimeter wave scanning systems

An update to an older story on SmartCheck Airport X-Ray Machines:
TSA is now testing the system as an alternative to pat-downs. This story ran in InformationWeek:

"Privacy is ensured through the anonymity of the image: It will never be stored, transmitted, or printed, and it will be deleted immediately once viewed." Ensuring privacy, as the TSA describes it, involves having security officers view images from remote locations. Thus, the security officer cannot identify the passenger, visually or by some other means, but can send word to fellow officers if a threat is detected.

According to the TSA, the scanning system applies a security algorithm to further protect passenger privacy by obscuring the passenger's face.
. . .

"First, this technology produces strikingly graphic images of passengers' bodies," Steinhardt said. "Those images reveal not only our private body parts, but also intimate medical details like colostomy bags. That degree of examination amounts to a significant -- and for some people humiliating -- assault on the essential dignity of passengers that citizens in a free nation should not have to tolerate."

The TSA says that since February, when it began testing backscatter scanning -- a similar technology -- in Phoenix, some 79% of those selected for secondary screening opted to submit to a backscatter scan rather than a pat-down.

The TSA plans to perform further testing of these systems at New York's John F. Kennedy International Airport and at Los Angeles International Airport. The security agency plans to purchase eight millimeter wave units for a total of $1.7 million. Millimeter wave scanners trials are currently being conducted at airports in Japan, the Netherlands, Thailand, and the United Kingdom.

Posted by rshah at 09:54 PM | Comments (0) | TrackBack

October 02, 2007

Tracking the Police in Chicago

From the Chicagoist:
The police in Chicago Lawn will be tracked by cell phones with GPS tracking. This is a pilot project that the city hopes to expand. The goal here is to better supervise the police, which should increase their safety and address concerns about police misconduct. The question in the article is whether the officers will accept the tracking. Will the data be used for major crimes or will it be used for minor infractions such as picking up dry cleaning?

I am more interested in how this data is integrated into the overall systems. You could see many uses of knowing exactly where police officers are located for incident response. Also, over the long term, the data could be used to figure out areas that the police are focusing on too much, while overlooking other areas. It could also be used to see what officers are the best responders. Lots of potential uses . . .

Posted by rshah at 03:23 PM | Comments (0) | TrackBack

September 16, 2007

Sense Through The Wall Technology

From BBC NEWS:
A roundup of smart camera technology by the BBC has something new:

"Each individual has a characteristic profile," explained Ian, holding a green rectangular box that looked like a TV remote control. Using radio waves, you point it a wall and it tells you if anyone is on the other side. His company, Oceanit, is due to test it with the Hawaiian National Guard in Iraq next year, and it turns out that the human body gives off such sensitive radio signals, that it can even pick up breathing and heart rates.

"First, you can tell whether someone is dead or alive on the battlefield," said Ian. "But it will also show whether someone inside a house is looking to harm you, because if they are, their heart rate will be raised. And 10 years from now, the technology will be much smarter. We'll scan a person with one of these things and tell what they're actually thinking."

I found out a bit more on the Oceanit web site in another news story by KGMB9:

Local company Oceanit has a two-year, $1 million contract to develop technology to keep soldiers safe as they sweep through a town or go house to house. It will be a hand-held unit that looks something like the tricorders "Star Trek Next Generation" troops used. "You go out there and you hit your buttons, and you kind of wait, take a reading of the building and then based on that, you do something," Sullivan said.

The technology uses radio waves, "so it bounces off them and from that we can determine heart rate, respiration and where they are," he added.

The company used similar technology in an earlier project to develop a sort of "Star Trek" bed, where a patient's vital signs are taken, without even touching the patient. Firefighters could also use the technology in a rescue to determine if there's anyone alive in a burning building.

Posted by rshah at 08:58 PM | Comments (0) | TrackBack

July 30, 2007

Airport Video Surveillance

From Inside Bay Area:

A nice anecdote about airport video surveillance by Duane McGray, who recently retired as head of security for the Nashville International Airport. The airport has video surveillance throughout its two terminals in case someone bypasses security. One thing they could do was:

"They would capture the individual's image, then send that digitized photograph to an e-mail address (on personal digital assistants carried by airport police officers) instantly, so that each of the officers can pull it up and have it in their hand as they look for the individual," McGray said. He added, "Very rarely has someone gotten through; probably less than one every year or 18 months."
There was one time when the digital images came in handy, however, said McGray, who is now executive director of the Airport Law Enforcement Agencies Network. "What they did do once is they caught a shoplifter."

Posted by rshah at 09:52 PM | Comments (0) | TrackBack

July 13, 2007

Face Recognition Flop

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.

Posted by rshah at 01:40 PM | Comments (0) | TrackBack

June 22, 2007

RFIDs and Cameras

From RFID Update:
An interesting use of RFID, bar code readers, and cameras. These systems are linked together so:

When the bar code label on freight is scanned to record a move, the RTLS tag sends a signal over the legacy WiFi network to the video surveillance system. System software quickly calculates the tag location and directs an appropriate camera to record the activity occurring there. Shipment information from the bar code scan is recorded with the transaction location, with a marker to enable easy retrieval of the video record.

This integration allows a warehouse to easily display camera footage of a package. The old approach (which required viewing multiple camera footage by hand) would take days to complete.

Posted by rshah at 01:38 PM | Comments (0) | TrackBack

June 11, 2007

Boeing to flip the switch on SBI-Net

From Washington Technology:
Boeing is starting to test a camera system this week as part of the Secure Border Initiative Network (SBI-Net). A 28 mile section that is a high traffic area now has:

nine mobile towers, each with radar, day/night cameras and other sensors. Each tower also has data processing and communications equipment to effectively distribute information to control centers, mobile units, agent vehicles and other law enforcement personnel. The information is processed into a common operating picture, computer application that provides a near real-time view of the location of suspected intrusions as well as the location of department personnel, vehicles and other assets.

. . .

Boeing sponsored an industry meeting day on May 30 attended by nearly 900 people representing more than 400 companies, McElwee said. The company has 55 requests for proposal for additional technologies to be deployed along the border.

“Project 28 gives us a demonstration of our approach and a test bed for incorporating improvements. The expanded team refreshes out technology, ensures low prices and gives us the capacity to execute on the much larger task orders that lie ahead,” McElwee said.

It will be interesting to see what useful technologies come out of this project.

Posted by rshah at 10:35 PM | Comments (0) | TrackBack

May 31, 2007

Amateur Facial Recognition?

From michaelzimmer.org:
An interesting post by Zimmer on how its possible to search on a Google Image Search for just faces (e.g., “&imgtype=face” appended to the query). Google is essentially providing a database of faces, which could be useful for facial recognition. While there are potentially scary applications, for now, its just something interesting. I'll be curious if anyone is using this with smart cameras.

Posted by rshah at 03:15 PM | Comments (0) | TrackBack

May 26, 2007

Amsterdam airport deploys body-scanning machines

From Reuters:

Amsterdam's Schiphol airport began using new body-scanning machines at security checkpoints on Tuesday, the first major airport to use the technology to find metals and explosives hidden under clothing. The "security scan" system, which uses harmless radio waves to display head-to-toe images of people, is also being used by other airports on a trial basis, but Schiphol is the only one to deploy the technology for regular use at its checkpoints.

Going through the scanner takes about three seconds, allowing users to avoid metal detectors or body searches. For privacy, the digital images are viewed by security personnel in another room and deleted after they are viewed.

. . .

But the alternative, being hand-frisked, is "never a happy story," Rutten says.

Posted by rshah at 06:23 PM | Comments (0) | TrackBack

May 15, 2007

Human Relationships with Intelligent Machines

From washingtonpost.com:
A fascinating article on how humans build relationships with the robots they use in Iraq and Afghanistan. There are currently thousands of various kinds of robots used by the military. Just as humans have grown attached to their cars, ships, and planes, the same is true for robots.
Its an interesting phenomenon and makes you wonder whether designers should design in features to make robots with personality. I doubt smart cameras inspire this kind of attachment, but why not?

Posted by rshah at 10:14 AM | Comments (0) | TrackBack

May 09, 2007

Stop Sign Cameras

From the newspaper.com:
Redflex, a major developer of red light cameras, has developed a stop sign camera, read their announcement.

The stop sign devices are based on red light camera platforms, but they differ greatly in use. The more familiar stoplight cameras typically photograph a vehicle entering an intersection if a signal light changes to red as little as 0.1 seconds after the car crosses the stop bar line. In most cases, the resulting ticket photograph will show both the vehicle in the intersection and a visible red light, offering visual documentation of the technical violation. With the new Redflex stop sign cameras, a machine will make calculations to determine whether a vehicle did not come to a complete stop and deserves a ticket.

While I think this isn't as important as red light cameras from a traffic safety perspective, it does show the growth in this industry. I am sure over the next several years, we will see lots of these ticketing devices. They appear effective at enforcing laws and generating revenue. Moreover, the public outcry over them hasn't sufficiently coalesced to protest these devices.

Posted by rshah at 10:46 AM | Comments (0) | TrackBack

May 02, 2007

Next-Gen Binoculars Merge With Soldiers' Brains

From Wired:
A story on a technology DARPA is developing that would merge binoculars with our brains. The idea is to use human brains for target recognition and pattern recognition. Its still pretty far out, but the article makes the technology foreseeable (the first prototypes won't be ready for several years).

The most far-reaching component of the binocs has nothing to do with the optics: it's Darpa's aspirations to integrate EEG electrodes that monitor the wearer's neural signals, cueing soldiers to recognize targets faster than the unaided brain could on its own. The idea is that EEG can spot "neural signatures" for target detection before the conscious mind becomes aware of a potential threat or target.
Darpa's ambitions are grounded in solid research, says Dennis McBride, president of the Potomac Institute and an expert in the field. "This is all about target recognition and pattern recognition," says McBride, who previously worked for the Navy as an experimental psychologist and has consulted for Darpa. "It turns out that humans in particular have evolved over these many millions of years with a prominent prefrontal cortex."
That prefrontal cortex, he explains, allows the brain to pick up patterns quickly, but it also exercises a powerful impulse control, inhibiting false alarms. EEG would essentially allow the binoculars to bypass this inhibitory reaction and signal the wearer to a potential threat. In other words, like Spiderman's "spider sense," a soldier could be alerted to danger that his or her brain had sensed, but not yet had time to process.

Posted by rshah at 02:06 PM | Comments (0) | TrackBack

April 17, 2007

Intelligent Assistive Technology and Systems Lab (IATSL)

IATSL - Intelligent Assistive Technology and Systems Lab:
Smart camera systems are typically focused on security applications. However, there are lots of other applications these systems can be used for. IATSL is a university laboratory that has applied smart camera technology to issues around accessibility and disability. Two simple examples:

Alex Mihailidis has worked on a smart camera setup that provides voice prompts for people washing their hands (link to Efficacy of Visual Prompting Among Older Adults with Dementia). This is needed, because people with dementia often have difficulty doing simple tasks such as handwashing.

A smart camera setup is used to watch stairs and automatically detect any anomalous (unsafe) events that may occur. This system is useful "to help further the understanding of the causes and catalysts of injuries on stairs and prevent accidents on stairs. Furthermore, such a system can help promote independence for older adults as their inability to safely traverse stairs greatly impedes their ability to live independently at home." (link to Automated detection of anomalous events on stairs)

Posted by rshah at 09:37 PM | Comments (1) | TrackBack

Update on ANPR in San Franscisco

From Examiner.com:
An update on an earlier story about the use of ANPR in San Francisco. The city states that with ANPR they have gone from finding 250 people to 350 people every two weeks with unpaid parking fines. Additionally, in one month they found 38 stolen cars, which was something they hadn't done before. The technology is in four parking enforcement vans. They seem happy with the system and are considering expanding to determine if people have residential parking permits. This is another good example of how ANPR is being used.

Posted by rshah at 09:27 PM | Comments (0) | TrackBack

April 06, 2007

Talking Cameras

From the BBC and Spy Blog:

The UK is going to start adding loudspeakers to some cameras to tell people to stop littering or other anti-social behavior. The BBC has a clip of the system in action, where people who leave behind litter are asked to please pick up the litter. This is slightly different than other talking camera systems, because this system relies on a live camera operator.

There are several issues this technology raises. The first is how intrusive do we want government/camera systems to be? Do we want them to shout instructions to us? Does it matter this is for littering or should it be extended to other more dangerous situations, such as assaults? The second is how effective this system will be. This strikes me more as a novelty than as something that will have lasting consequences. Along these lines, an issue raised by Spy Blog is the vast difference between the optical power of a camera and the audible range of a loudspeaker. Loudspeakers can't zoom and pan like cameras, so the audible instructions may be difficult to hear or confuse other third parties.

Posted by rshah at 03:17 PM | Comments (0) | TrackBack

March 02, 2007

Smart Microphones

From Business 2.0 via CNN Money.com:
Another approach that computer processing (the smarts) can exploit is analyzing sound. It is an interesting idea and so far the most notable is the gunshot detection systems.
Sound has a number of advantages including its cheaper (a high quality camera is $500, a high quality microphone is $50) and audio takes less bandwidth. Here are a couple of potential uses:

Sure, intelligent software can examine hours of security video footage and, say, try to recognize intruders, but it's far easier to train a computer to listen for the sound of breaking glass.

A guard at a bank of screens can't always monitor the right one. Listening algorithms can filter out the air-conditioning and hear a door opening in an abandoned corridor. By triangulating the sounds of footsteps, the software can then tell security cameras where to point.

Listening devices can monitor how efficiently office space is put to use. If a spare conference room is filled with hubbub less than once a week, it might be time to turn it into a cube farm.

I think its a great idea and I wonder if more vendors will consider implementing smart microphones. While there are legal issues with government use of microphones in surveillance, there is no reason it can't be used by private parties for lots of different applications.

Posted by rshah at 11:26 AM | Comments (0) | TrackBack

February 28, 2007

Cameras in Teenagers' Cars

From WSJ.com:
American Family Mutual Insurance will offer some of its customers a camera system known as DriveCam as a way of improving teenagers' driving behavior. The DriveCam captures sights and sounds inside and outside the vehicle.

DriveCam's palm-sized, exception based video event recorder is mounted on the windshield behind the rearview mirror and captures sights and sounds inside and outside the vehicle. Forces (e.g. hard braking, swerving, collision, etc.) cause the recorder to save 20 seconds of audio and video footage – the 10 seconds immediately before and after the triggered event.

When the video event recorder is triggered a light blinks to alert the driver. This is intentional so the driver knows what he/she did to activate the video event recorder and can aim to avoid repeating that behavior.

DriveCam's Certified Driving Behavior Analysts take a cursory look at downloaded events to identify anything that would be critical for a customer to know immediately (e.g. a collision). Next, the experts closely review and assign a risk score to each event.

The story provides some anecdotes of teenagers who claim that the blinking light has led them to drive in a safer fashion. I think the feedback loop here is an important component for not only tracking behavior, but also trying to improve it. Read the article to get the full details on privacy, the role of parents, and how this will affect our insurance rates. Here is the WSJ video (the article is much better):

Posted by rshah at 10:25 AM | Comments (0) | TrackBack

February 17, 2007

Sound Cameras for Noise Pollution

From PhotoEnforced.com:
Alderman Natarus of Chicago is working on fighting noise pollution with sound cameras. According to his site:

The good news is that we are exploring a new technology that may give us the edge in our efforts to combat excessive motorcycle noise. There is a new device on the market called a sound camera that can take noise readings and automatically issue a ticket just like the red light cameras that are being installed throughout the City.

I am working with Mayor Daley to research the possibility of installing sound cameras in the locations where we get the most motorcycle traffic in our ward.

Its an interesting idea, I just wonder if this is that big of a problem. After all, red light cameras are trying to avert conduct that results in injuries and death. (More than 850 people die and and about 170,000 are injured each year in crashes caused by drivers running red lights according to the National Highway Traffic Safety Administration - USA Today).

Posted by rshah at 03:29 PM | Comments (0) | TrackBack

December 29, 2006

ANPR in San Francisco for Outstanding Parking Tickets

From San Francisco Chronicle:

San Francisco has outfitted a car with ANPR to scan more than 250 plates an hour. It looks for cars with at least five outstanding tickets and stolen cars. The technology costs $92,000 for one car. I don't understand why its so expensive. But the price is being paid by ACS Inc. (they hold the ticket-processing contract with the city). Its a test run right now. I would wager that ANPR will be widely used in San Francisco for these purposes, but it seems a bit expensive right now.

This application has been discussed here previously. Boston and New Haven for using ANPR for collecting on outstanding parking tickets.

Posted by rshah at 10:59 AM | Comments (0) | TrackBack

December 19, 2006

Cameras for Evidence Gathering

Another use for cameras is to collect evidence of incidents or accidents. For example, there have been calls for patrol cars to have video recording (e.g., Steve Chapman of the Chicago Tribune). In one example:

In a three-month experiment with 74 Oakland police officers, complaints were filed against 15 of the cops when they patrolled without video cameras. But when they were driving cars with cameras, there wasn't a single complaint filed against any of the 74.

Another example in the Tribune is to place cameras on the front of Metra trains to collect evidence when accidents occur. Metra plans to install more than 500 cameras at a cost of $3.5 million.

In its test along the Rock Island Line, Metra's cameras captured events that engineers witness every day but that would shock many people, especially parents, officials said.

Efforts such as the international Operation Lifesaver program attempt to educate people about taking risks at rail crossings, but the videos captured by the cameras may be more effective. Footage of youths trespassing on tracks might help pinpoint areas near playgrounds or schools.

They also may put pressure on police to better enforce laws against pedestrians and vehicles ignoring gates and signals, he said.

Posted by rshah at 09:49 AM | Comments (0) | TrackBack

December 15, 2006

Gait Analysis

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.

Posted by rshah at 06:36 AM | Comments (0) | TrackBack

November 18, 2006

Face-Recognition for Fighting Fraud

From Slashdot citing the Des Moines Register:

Digimark Corp. of Beaverton Oregon will be processing 2.1 million Iowan drivers license photos to prevent fraud. The idea is that the system will alert a DOT employee if a person renewing their picture is different than the previous photo. This is a nice example of how smart cameras can automate processing to allow for new ways of monitoring. The article notes that similar biometric checking systems are used in Alabama, Colorado, Kansas, Massachusetts, Oregon and Texas.

Posted by rshah at 11:05 PM | Comments (0) | TrackBack

October 25, 2006

Smart Cameras for Industrial Security

From SecurityWatchInfo:
An article on how smart cameras can be used for industrial security, such as scrap metal theft, industrial-plant theft, frivolous injury claims, and the theft of of toxic chemicals, biomedical waste, and flammable materials. Its a good article on the added value of smart cameras. Here is one snippet:

Consider some common analytics that can be applied to an industrial security setting: Intelligent video detection can distinguish between vehicles and people and even limit detection to a specific direction, so when a scrap-metal thief climbs over a fence it will be detected, but an employee car leaving the facility would be ignored. Automatic detection of object removal can be used to limit detection to items that are of interest, whether it is the taking of a warehouse computer or a pallet of copper piping from a storage yard. Automatic detection of unattended objects can reveal when a box of product is tossed over the fence by an employee for post-work retrieval or alarm when a suspicious package is placed next to liquid propane tanks. Automatic detection of stopped cars can notify security when a vehicle stops for too long of a period in a prohibited area. Autonomous pan-tilt-zoom tracking, unlike stationary cameras, can provide automatic close-ups of a moving target for better recognition or real-time tracking of a camouflaged intruder's whereabouts.

Posted by rshah at 07:24 AM | Comments (0) | TrackBack

October 19, 2006

Assessing Pain in Infants

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.

Posted by rshah at 07:41 AM | Comments (0) | TrackBack

October 05, 2006

Virtual Fences for Security

From Security Park:

A story on smart cameras at the Port Macquarie Airport in Australia. The airport is using smart camera technology for intrusion detection and to monitor human activity in areas such as the passenger terminal, runway, and parking areas. One of the main uses is for perimeter security, which traditionally uses fencing systems that can be expensive. According the the security manager, smart cameras are easier on the budget and very effective:

Hall said that the perimeter intrusion detection feature within the QOLiT system was of interest to the airport, as high security fencing was not an option for them. “Leveraging off the perimeter detection feature within the VI Activity Detection system costs less than a tenth of what perimeter fencing would have cost. Plus you don’t have the up-keeping costs of fencing repairs and maintenance.” says Hall.

Using a 3D scaling tool within the VI Activity Detection system, users can easily create and preset invisible ‘trip wires’ and ‘barriers’ in the software. Should any human trespass the invisible barriers, even if the perimeter is a long way off in the distance, the algorithms can discern between the sizes and aspect ratios of objects regardless of how near or far they are from the camera. In this way, unimportant activities such as sudden changes in lighting conditions or movement of small animals can be ignored, regardless of where they are in the image.

It appears that the real power of video analytics is in its ability to keep what does need to be addressed and filter out what doesn’t, alerting only when someone or something seems suspicious. “What Port Macquarie particularly like so much about the QOLiT VI Activity Detection system, is that they are alerted to situations that of importance to them. They are getting about ninety eight percent plus accuracy on event alerting, which is what they wanted.” he said.

Posted by rshah at 10:59 PM | Comments (0) | TrackBack

September 11, 2006

Smart Cameras in Stores

From BusinessWeek:
An article on loss prevention in stores notes two different smart camera technologies. The first is called Video Investigator from Intellivid and has been discussed here. The article notes:

Some Macy's (FD ), CVS (CVS ), and Babies 'R' Us stores have installed a system called the Video Investigator, whose advanced surveillance software can compare a shopper's movements between video images and recognize unusual activity. Remove 10 items from a shelf at once, for instance, or open a case that's normally kept closed and locked, and the system alerts guards sitting in a back room -- or pacing the sales floor -- with a chime or flashing screen. The system can predict where a shoplifter is likely to hide (at the ends of aisles, behind floor displays). A search function spots sudden movement that might indicate a large spill, prompting workers to clean up before it leads to a slip-and-fall accident and a costly lawsuit. And if someone opens a back door at 2 a.m., the system will record who sneaked in and link it with snapshots of the previous and next persons to use the door.

The second category of smart software focused on syncing video with cash registers also discussed here:

With the press of a button, managers can highlight irregular register transactions on their computers and pull up corresponding video. This could enable them to catch cashiers who cut deals for their friends or pocket cash refunds themselves. It could also curtail fraudulent returns by tracking the route customers take to the customer service desk -- do they head straight there or meander through the store, picking up their "return" merchandise along the way?

Posted by rshah at 05:24 AM | Comments (0) | TrackBack

August 07, 2006

Smart Cameras for Marketing Research

From PC-Welt:

Smart cameras are also applicable to marketing research. The article provides a few ideas along these lines:

Companies are using video to measure footfalls in stores and within subsections of stores. They map traffic -- where someone walks, where they stop and for how long -- to see how people browse. Why do they tend to stop longer in one spot? Is there something there that's drawing them? Are there "blind spots" in the store where people are missing good products? There's any number of things you can look for. The visual data is used to reconfigure store interiors to maximize marketing and drive sales. Outside the store, surveillance of mall parking lots can spot where traffic backups occur, so trouble spots can be reconfigured in ways that make it easier to park -- and shop

Posted by rshah at 09:56 PM | Comments (0) | TrackBack

Doctors Using Cameras

From 10 Now:

A new study being conducted at Upstate Medical University will look into using images of crash scenes to help treat victims. Doctors will use traffic surveillance cameras to see the accidents and rescue operations to try to help the staff in the E.R. provide better care to crash victims. They will have access to 20 DOT cameras around Syracuse to see the accident scenes right after they happen. The idea is that if doctors know more about the accident, they can better treat the patient. The study is funded by Alfred Kahn, who survived a crash in 2003 and spent weeks recovering at University Hospital.

What an unusual use of traffic surveillance cameras! It will be interesting to see the results of the study.

Posted by rshah at 09:50 PM | Comments (0) | TrackBack

July 18, 2006

Smart Cameras for the Pool

Poseidon Systems offers smart camera technology for drowning detection. It relies on underwater cameras and a processor that analyzes/tracks swimmers in real-time. They have press releases stating that their system helped detect near drownings. Here is a screen shot from their web site:

Posted by rshah at 08:02 PM | Comments (0) | TrackBack

June 15, 2006

NIce Systems

From The Register:

A story on Nice Systems which makes smart software. Nice is a leader in smart phone surveillance technology. This software is used by the police to identify people as well as when keywords are used. It is also used by corporate call centers to detect "emotions." For example:

FedEx's Custom Critical service, for example, uses Nice software to notice when someone says "wow," explains Eyal Danon, Nice's vice president for global marketing. "Whenever a customer says 'wow' on a call, that call immediately is being recorded and then it's being sent to hundreds and hundreds of call center agents so they can actually learn from the interaction and see what prompted the customer to say 'wow,'" he says. From ASAP

They are now moving into smart software for cameras. Here is a PR blurb from the Register article:

"By employing software-based analytics on unstructured multimedia content, companies are able to detect customer intent, often through near real-time interactions where a customer may express concerns, desires or provide other signals of their intentions," said the statement. The same advantages of crime prediction will be conferred on the security services, it said: "Our solutions enable our public safety and security customers to identify threats as they occur, and analyze video footage to identify suspicious objects or behavior more quickly and effectively."

Posted by rshah at 03:01 PM | Comments (0) | TrackBack

May 30, 2006

Intelligent Scarecrow

From Engadget:
A fun application for smart cameras:

students and faculty at the University of South Florida in Tampa have built a computer-powered model that can detect incoming birds and employ non-lethal countermeasures to protect their wards. The Intelligent Scarecrow, as it's known, was developed to combat the problem of nervey birds trying to feed themselves at the expense of Florida's $42 million fish farming industry . . . uses a networked camera linked to image recognition software for identifying winged menaces in the vicinity, and attempts to repel them with a mix of annoying sounds and even more annoying blasts of water.

Posted by rshah at 10:06 PM | Comments (0) | TrackBack

May 16, 2006

Reducing Wait Times with Smart Cameras

From SecurityInfoWatch.com:

A nice simple application for smart cameras from Vidient. Helsinki airport will use smart cameras to automatically alert security officers of long wait lines at security checkpoints, so they can open more lanes. A nice simple application that allows cameras to do the routine work and free up security personnel for more complex tasks.

Posted by rshah at 09:17 PM | Comments (0) | TrackBack

May 12, 2006

Airport Security

I ran across a couple of articles today on smart cameras and airport security. The first was at silicon.com and discussed the role of smart cameras at San Francisco airport. The article notes that:

Vidient claims its algorithms can monitor what those cameras are seeing and can flag up any suspicious events. Most commonly they are looking for individuals "tailgating" - following another person through an opened door - or static objects, either in secure areas or at kerbside. Foster said a "one swipe per person" system for passing through all doors is rigorously enforced. The cameras can detect if two people pass through an open door, and can even differentiate between one very large person and two smaller people. Even staff who are entitled to pass through that door must swipe, in order to prevent an alert being triggered. And those who do trigger an alert, Foster said, are "admonished" for failing to follow security best practice. "People now know we have a system in place," he said Foster, adding that alerts have fallen considerably after some re-education and admonishment.

Another article from the Wall Street Journal discusses various technologies for securing airports. The articles notes the privacy issues with backscatter xray technologies. This is an issue I posted about several times, 1, 2, and 3. In response to the privacy concerns:

Rapiscan Systems, a division of OSI Systems Inc., of Hawthorne, Calif., and American Science & Engineering Inc., Billerica, Mass, have worked to develop proxy images that mask private parts. The image created would show the outline of foreign objects on the body, but not the body itself. Pilot testing on these machines, which cost $70,000 to $100,000, should begin in early spring.

Posted by rshah at 10:42 PM | Comments (0) | TrackBack

May 08, 2006

VideoNEXT

From Newswire:
A press release for videoNEXT notes a few applications of smart cameras:

-- Airports currently employ guards to prevent people from entering the secured gate area by entering at the exit. Often this exit is a corridor that immediately adjoins the entrance, which is also manned by guards who are clearing passengers for entrance to the secured gate area. The videoNEXT object identification and tracking technology can be deployed to detect persons moving in the wrong direction. Upon detection, the system would alert guards stationed at the entrance. These guards could intercept the intruder. There are more than 400 airports in the United States. The exits are guarded seven days a week for 16 or more hours per day. The potential savings are considerable.

-- At border crossings, one problem for US Customs agents is that cars often ride up to a security point, then make U-turns. Law enforcement agents suspect that they are engaged in criminal activity, and run after seeing more security than expected. The object tracking technology can alert a person at a remote location that a U-turn has occurred and will record the make and model of the car, as well as the license plate.

-- The Army is also exploring ways to use the technology on the battlefield. If the system is monitoring an area, for instance, it can alert commanders when something or someone has moved in one of the areas being monitored.

Posted by rshah at 09:03 PM | Comments (0) | TrackBack

February 01, 2006

Talking Cameras

From The Chief Engineer:

Baltimore's Board of Estimates has approved five talking cameras - armed with motion detectors, a bright flash and a recorded warning - as part of an effort to curb quality-of-life crimes, especially illegal dumping.

When the solar-powered cameras detect motion, they will issue a scolding: "Stop. This is a restricted area. It is illegal to dump trash or spray graffiti here. We have just taken your photograph. We will use this photograph to prosecute you. Leave the area now."

While that is the default message, the city could choose to record a collective admonishment from nearby residents or even a personal threat by Mayor Martin O'Malley.

The cameras cost $5,000 apiece and will be added to an already expansive network of surveillance equipment in Baltimore. City officials would not say where, specifically, the cameras would be placed.

But the idea is to surprise litterers with a booming voice, most likely coming from a light pole. The camera will also snap a still photograph and save it to a storage card, which police could use to identify a suspect.

"It's quite startling," said Ken Anderson, president of California-based Q-Star Technology, which developed the camera. "It's generally going off in the middle of the night, (and) people generally aren't expecting it."

About 150 cities use the cameras to control graffiti, loitering and illegal dumping, Anderson said. Cincinnati has installed 20 cameras, mostly in residential areas and city parks.

Posted by rshah at 07:41 PM | Comments (0) | TrackBack

January 26, 2006

Uses for Cameras Other Than Security

From Security Park:
A pitch by ADT notes three different use of cameras other than crime prevention. This is a useful followup to an earlier post discussing regulating the use of cameras for crime prevention.

Employee Productivity Monitoring can significantly impact employee behavior and improve compliance to company policies and procedures. A popular fast food chain recently introduced video monitoring at a franchise, which helped identify and correct employee training issues and staffing inefficiencies. The result was a 35% improvement on customer time spent in line.

Customer Service Information gleaned from video surveillance often leads to a better customer experience. For example, retailers can review video and determine that additional registers are required or certain employees are not engaging customers in the friendliest manner.

Compliance Video monitoring technology can help improve compliance in the food industry, where quality assurance procedures are critical. From how raw food is handled to how final products are packaged, remote video monitoring integrated into a plant's systems can provide assurance that industry regulations, sanitation and safety criteria are being followed.

Update:
Peter over at Public Eye points out other uses for cameras including construction monitoring, marketing, and remote telepresence. Check his web site to see actual footage.

Posted by rshah at 05:04 PM | Comments (0) | TrackBack