Smart Cameras Blog

[From Philips camera pill easy to swallow | Crave - CNET and Pillcam from CNET]

Two interesting pill size cameras/sensors for use inside the body. The Pillcam Colon Capsule Endoscope is a large pill with a camera inside it. It can generate up to 144,000 images over a 10-hour period.

The other pill size sensor is from Philips. It can be programmed to deliver targeted doses of medicine to patients with digestive disorders. Besides delivering medicine to multiple locations, the pill can measure data such as local temperature, and report measurements wirelessly to an external receiver unit.

rshah Applications

[From Step right up, let the computer look at your face and tell you your age and Physorg.com]

A press release from the University of Illinois focuses on smart cameras that can identify the approximate age of a person. The age recognition algorithms can estimate ages from 1 year to 93 years. The software’s accuracy ranges from about 50 percent when estimating ages to within 5 years, to more than 80 percent when estimating ages to within 10 years. The software consists of three parts: face detection, discriminative manifold learning, and multiple linear regression. The software was trained on a database of 1,600 faces, but a larger database can add more accuracy. The team was led by Thomas Huang.

The article lists a number of possible applications for this technique:

For example, age-recognition algorithms could stop underage drinkers from entering bars, prevent minors from purchasing tobacco products from vending machines, and deny children access to adult Web sites . . . . In addition to performing tasks such as security control and surveillance monitoring, age-estimation software also could be used for electronic customer relationship management. For example, a camera snapping photos of customers could collect demographic data – such as how many adult men and women buy burgers, or what percentage of teenagers purchase a particular soft drink. Or, combined with algorithms that identify a person’s sex, age-estimation software could help target specific audiences for specific advertisements. For example, a store display might advertise a new automobile or boat as a man walks by, or new clothing or cosmetics as a woman walks by.

Another research project mentioned at Physorg.com recognized six emotion states based on audio and visual data. The system could recognize happiness, sadness, anger, fear, surprise, and disgust across cultures with a success rate of 82%.

The researchers’ system extracted a large number of vocal characteristics, such as “prosodic features,” which include the rhythm, intensity, rate, and frequency of speech. Facial features were extracted holistically. Then, the researchers trained the system on several short video samples of individuals showing different emotions, from which it connected certain features with emotions.

rshah Applications

[From theNewspaper.com Photo Ticket Cameras to Track Drivers Nationwide]

The vendors of red light cameras and speed cameras are planning to add more features that expand the ability to track motorists. Redflex is planning to add OCR to their cameras in the next few months, which is also known as automatic license plate recognition technology. This could allow them to keep tabs on every car that passes through a particular intersection. “Imagine if you had 1500 or 2000 cameras out there that could look out for the partial plate or full plate number across the 21 states where we do business today,” Elsadek said. “This is the next step for our technology.”

The article also points out that these technologies can be abused. A recent example the article mentions:

In the past, police databases have been used to intimidate innocent motorists. An Edmonton, Canada police sergeant, for example, found himself outraged after he read columnist Kerry Diotte criticize his city’s photo radar operation in the Edmonton Sun newspaper. The sergeant looked up Diotte’s personal information, and, without the assistance of electronic scanners, ordered his subordinates to “be on the lookout” for Diotte’s BMW. Eventually a team of officers followed Diotte to a local bar where they hoped to trap the journalist and accuse him of driving under the influence of alcohol. Diotte took a cab home and the officers’ plan was exposed after tapes of radio traffic were leaked to the press. Police later cleared themselves of any serious wrong-doing following an extensive investigation.

rshah ALPR, Applications, Red Light Cameras, Traffic Congestion

[From Google’s Photo Face Recognition is Wow Marketing - Bits Blog - NYTimes.com]

Google has introduced facial recognition software in its photo sharing site. With Picassa web albums, users can now tag photos by name. The software then looks at other pictures in your albums and finds matches using facial recognition software. Judging from the early feedback, it works pretty good. Here is some more info on the feature. The software was acquired from Neven Vision. Here is an explanation of the process:

As I was dutifully uploading my summer vacation photos last night, the Picasa site alerted me to this new feature, which was introduced Tuesday and lost in the noise about Google’s Chrome browser. I clicked a button that set some Google server farm looking closely at my stored photos (mainly of my two constantly preening daughters).

Just a few minutes later, it presented a cluster of photos of what the software assumed were of the same person. Indeed, they were all of my six-year-old, Daphne. I typed in her name. Then it presented another batch it thought were of Daphne. I could click to confirm. I could also uncheck some photos that were not her. It similarly offered clusters of photos of my other daughter, Clare, as well as my wife and other people who I tend to photograph.

rshah Applications, Facial Recognition

[From Homeland Security's 'virtual' border fence ends up, well, nonexistent | Politics and Law - CNET News]

The continuing saga of the Department of Homeland Security’s “virtual fence” along the U.S.-Mexico border, also known as the DHS Secure Border Initiative (SBInet), for past history take a look at: 1 and 2. The project is relying on smart cameras, sensors, and radar as a virtual fence. I always thought this project could really show how useful smart camera systems could be. Alas, the project is not going well (GAO testimony):

The Government Accountability Office reviewed the SBInet program from March to September of this year and testified about its findings in a hearing in front of the House Committee on Homeland Security on Wednesday.

“Are we making progress?” said Randolph Hite, director of IT architecture and systems issues for the GAO. “The answer is, we don’t know.” “I’ve never seen anything that answers that question of will the benefits exceed the cost,” Hite said.

The CBP has awarded Boeing, the main contractor for the SBInet program, $933.3 million in projects so far. The DHS has requested $75 million from Congress for operations and maintenance of what’s described as “tactical” infrastructure in place for 2009.

When the GAO visited in June the site of Project 28, a 28-mile strip of land at which a prototype for SBInet is under use by the border patrol, the system was hardly functioning, said Richard Stana, director of homeland security and justice for the GAO. “It took us 45 minutes just to get the system up and running,” he said. Additionally, radars were thrown off, camera range was limited, and the ability to classify items under surveillance was limited, Stana said. He said the prototype “did not meet expectations,” but that it was hard to hold the contractors accountable because any expectations in place were “loosely worded.”

Stana said that while Project 28 was intended as a model for security along the rest of the border, the project has essentially been scrapped, and the CBP will use different technologies. “The cameras, the radars, everything will change,” he said after the hearing.

Lets hope version two works better.

rshah Applications

[From Fraunhofer-Gesellschaft Research News 08-2008-Topic 5]

Smart camera aficionados may remember that IBM started its work on smart cameras to tell a cumquat from a rutabaga for grocery stores. But security emerged as a more profitable outlet and the rest of us now wait patiently for vegetables to be manually entered. But another group has taken on this task:

Working on behalf of the industrial weighing company Mettler-Toledo, researchers at the Fraunhofer Institute for Information and Data Processing IITB in Karlsruhe have developed a webcam module for self-service scales. “The scales automatically recognize which fruit or vegetables are to be weighed and ask the customer to choose between only those icons that are relevant – such as tomatoes, vine-ripened tomatoes and beefsteak tomatoes,” states IITB scientist Sascha Voth. Customers can confirm the correct variety on a touch screen.

But how do the scales know whether the customer has placed a pepper, a tomato or a kiwi fruit on them? “The goods are registered by a camera integrated in the scales. An image evaluation algorithm compares the image with stored data and thus automatically recognizes which type of fruit this is,” says Voth. Even the cloudy plastic bags in which the fruit may be packaged at the counter are no problem for the scales – the image evaluation system recognizes the various types of fruit and vegetable anyway.

rshah Applications

[From Shops track customers via mobile phone - Times Online ]

An article on Path Intelligence, which offers the ability to track shoppers through malls. They do this by measuring signals from a shopper’s cell phone with several monitoring units. This technology is installed in several malls in the UK. The benefits are:

A shopping mall could, for example, find out that 10,000 people were still in the store at 6pm, helping to make a case for longer opening hours, or that a majority of customers who visited Gap also went to Next, which could useful for marketing purposes. In the case of Gunwharf Quays, managers were surprised to discover that an unusually high percentage of visitors were German – the receivers can tell in which country each phone is registered – which led to the management translating the instructions in the car park.

The technology relies on the IMEI code for a cell phone. IMEI is a unique number given to each cell phone. Supposedly, only the phone network can match a handset IMEI to the personal details of the customer. (I don’t know how this approach compares to cell phones used for monitoring traffic congestion.)

This is another example of a smart system used for marketing purposes.

rshah Applications

[From MATEUSA.NET]

A sensible application of smart cameras for traffic. I apologize for just using a press release, but this should provide a flavor of how they can be used.

MATE’s Behavior Watch™ intelligent video detection system performs analysis using specifically designed outdoor algorithms to detect unusual events that may cause a security hazard: stopped cars on a road or shoulder, boats lurking near bridges, people or animals crossing highways, accidents or fallen cargo. This automated incident detection system and real-time alarm notification solution helps traffic operators increase their awareness and response time to incidents that can cause traffic congestion and endanger public safety.

rshah Applications

[From Racist CCTV - facial recognition techniques used to classify people by race - thankfully this is only an art project ! - Spy Blog - SpyBlog.org.uk]

An interesting art project from Benjamin Wales. He has setup a smart camera system that has the ability to find and follow faces and then analyse and store their race data. His motivation was to critique our ideas about surveillance.

He also has another interesting project, Static Obesity Logging (SOLA), which analyzes BMI. The system is able to remotely calculate Body Mass Index and publish the data via wired and wireless networks.

Both of these applications are interesting ways to use the power of smart cameras.

rshah Applications, Facial Recognition

[From Video analytics: Video analytics experts point at computer vision as the key to accuracy in intelligent video surveillance]

A PR laden article that has a few useful insights into smart cameras. It notes limitations of smart cameras and then discusses their applications for retail stores, marketing research, and advertising.

Limitations

However, computer vision has yet to achieve the level of intelligence critical to video analytics systems. “I think one of the limitations of currently available video analytics systems is their accuracy level. Most applications are not 100 percent accurate because computers still lack the visual intelligence that humans possess,” noted Michelle Tecson, sales and marketing executive associate at Neugent. In people counting, for example, computers currently count blobs—blobs of people coming in, whether two people entering together side by side or one large person with a shopping cart. Both will be counted as one. When a person is standing directly behind another person sitting down, with his back on the sitting person, the computer will recognize them as one instead of two.

Applications:

Retail stores use video analytics to identify hotspots within a store. The term “hotspot” was coined by weather channels that show typhoon areas in color codes. In the retail arena, hotspots are places within the store that are most frequented by customers. By accumulating data on customers’ movements within the store over a period of time, the storeowner can create a color-coded map identifying hotspots. Based on the identified hotspots, the storeowner can plan the future layout of the store or the positioning of merchandise to boost visibility of items and increase profits. . . .

Video analytics in market research and advertising is a slightly more sophisticated system, but the benefits are real and tangible. Mariano observed that in some applications, face recognition technology is involved. “It’s a fairly new DVR application. Some companies in the US, such as Brickstream and Shoppertrak, are installing DVR solutions to analyze customer behavior in such places as banks, retail stores, grocery stores or gasoline stations,” he said. Mariano cited one project his former company had with McDonald’s in Philadelphia in the US, which involved counting cars going through drive-thru lanes, and detecting and classifying the gender of drivers based on their facial characteristics. They discovered that the DVR solution was 70 percent accurate, which was not bad for a machine. . . .

Advertising agencies have also introduced some video analytic solutions that measure the impact of an ad to a passerby or bystander. It detects a face (if it’s male or female) as well as the face’s movement, whether it is looking at the ad on the wall and for how long, or looking away from it.

rshah Applications