Face and the future

Identify online daters, better tools for law enforcement, face as currency, and more.

The future of facial recognition: 7 fascinating facts (TED Blog)

Building out the UID

UIDAI launches Aadhaar-enabled electronic know your customer service (Times of India)

Under the e-KYC process, one can authorise the UIDAI to release the KYC data to a service provider. The consent can be provided either in person (through biometric authentication) or online. The UIDAI will provide the details like name, address, date of birth, mobile number and email address to the service provider electronically.

The promise of UID is in the apps. It's nice to see them starting to trickle through.

BigID and the changing nature of national identity infrastructures

Nigeria's new ID has apps!

Credit card linked to Nigerian ID (Financial Mail)

In the programme's first phase, Nigerians aged 16 and older and all who have been resident there for more than two years will get the new multipurpose ID, which has 13 applications. It is expected that up to 13m Nigerians will use the product in the first phase.

Among the apps is MasterCard's prepaid technology, which will give cardholders the ability to make electronic payments. MasterCard says this will also have a positive impact on Nigerians who until now have not had access to mainstream financial services.

This one bears keeping an eye on.

In a couple of pioneering cases, the very concept of "The ID" is shifting

To most people, an ID looks a lot like a product — something useful that the government sells to an individual. Pay your fee; get your card. Lose your card; buy a new one.

India and Nigeria (South Africa is pretty bold, too) are pointing the way toward a future where ID isn't just a product, though no government is going to give up its ID card product line any time soon. The future as these countries see it is ID as a government-backed platform supporting an ID ecosystem. They have the bucket (database structure). Now it's being filled (populated). If they get the application programming interface/s (API) right, fasten your seat belt. Things will get really interesting really fast as all sorts of apps hooking into the ID infrastructure become available. Biometric technologies will be an integral part of this transition to "BigID."

UPDATE:
See also:
Brainstorming UID with Srikanth Nadhamuni
The video there is very informaative and extremely worthwhile.

UPDATE II:
I forgot to mention the UAE as another forward-thinking ID environment. The UAE ID is set to be deployed on smartphones.

Start with the applications

10 Big Data Trends From the GigaOM Structure Data Conference (eweek) Good observations having broad applicability in understanding how the recipe for organizational success is being rewritten. Read the whole thing.

Money quote:

While big data might be getting ahead of itself in enterprise promises, it is real in bringing new capabilities to business. You need to think about the skills you have in your company and developing the data skills to adapt to this new model. Open source, which often has a bit of a fringe reputation in the enterprise, will be part of your technology future. Established vendors are going to promise they can give you all the capabilities of the startups with added stability, but I haven't seen any evidence so far. Think about your applications from the outside in, instead of inside out.

The application, not the technology, is everything.

WVU's own Bojan Cukic holds forth on the state of biometric applications.

Tipping Point Unclear For Mass Market Adoption Of Biometrics

Cukic said that most biometric devices currently available are standalone solutions that do not have access to the Internet. Were this to change, Cukic believes the use of biometrics could receive a significant shot in the arm. "I would say that we have good components, but the question remains, who is going to be the developer responsible for offering these systems in which biometrics address some of the authorization and authentication problems that we face today?" he added.

Middleware and the application development it enables will be critical to moving these technologies out of the lab and niche government installations and into positive ROI applications for profit-making entities.
Or (more succinctly...

Very good article on the state of the industry. Middleware is the glue that pulls this together. #Biometrics @securlinx ow.ly/ccu1a
— Barry Hodge (@HodgeBarry) July 12, 2012

Mobile Devices and Biometric Modalities

Smartphones and tablets combine the most powerful attributes of the networked computer and the cell phone, extending the web into every nook and cranny of the globe.

In one awesomely tiny package they facilitate data collection, storage and access to data stored elsewhere.

As a platform for near field communication (NFC) and SMS One-time passwords, mobile devices are also increasingly being used to deliver identity management applications by using a person's known possession of the device as a way of verifying their identity. In access control lingo, mobile devices are being used as tokens.

Using mobile devices is a dream come true for businesses that rely upon tokens: Your customer already owns it; If they lose it, they will be aware of the loss very quickly and they will replace it at their own expense; People are disinclined to lend their phone/credential to someone else; Etc.

Now to the question of securing the device itself and biometric modalities.

Fingerprints are currently the most frequently used biometric for overtly identifying cooperative, habituated individuals. They have a lot of things going for them. Fingerprints are well-understood scientifically, durable, reliable, and fingerprint ID management techniques have been shown to deliver high return on investment in many applications.

These are some of the reasons I lamented Motorola's announcement that it was leaving the fingerprint sensor out of the Atrix 2. The decision makes sense, though. The fingerprint sensor wouldn't be widely used until developers had written software using it, but including the sensor would drive up the cost of each unit for a thinly-used feature. The innovation chicken-and-egg problem is a real one and Motorola seems to have made the judgement that they weren't gaining enough of an advantage in the highly-competitive mobile device market by including it.

But that hasn't meant the end of mobile device biometrics. Just as businesses that issue tokens have been able to take advantage of the fact that their users are already carrying the necessary technology around with them, biometric identity management application developers are doing the same.

Mobile devices already contain the hardware required to deliver two biometric modalities: a camera for facial recognition and a microphone for voice. These modalities present challenges not usually associated with fingerprint biometrics — in the case of facial recognition challenges include lighting and the well-publicized photograph hack; for voice, background noise can be a problem — but they offer the advantage that the hardware is "free" and never going to be yanked out of mobile devices. That's quite an advantage, and it points to why face and voice biometrics are the front-runners for handset biometrics.

Nice and tidy, eh?

So, what to make of today's news that Fujitsu is set to compete more aggressively in the global handset market?

Fujitsu Aims for European Mobile Phone Market (Financial Times)

Fujitsu’s smartphones will certainly feature electronic money technology – enabling owners to use NFC, the mobile payment system – and biometric recognition to make their use as mobile wallets more secure.

Fujitsu, more than any other handset manufacturer, is deeply involved in biometric sensor hardware (finger, palm) that doesn't currently reside on stock mobile platforms. So stay tuned.

The Cloud & Biometric ID Management

Cloud-Based Application Development for Biometrics Data (M2SYS Blog)

More and more international governments and security-intensive companies are using biometric-enabled identity cards for their employees and professionals. As this technology becomes more widespread, the need to make this technology more mobile, and more accessible is becoming clear. Experts all over the globe are pushing for cloud-based biometrics for greater efficiency and mobility.

More at the link.

Biometric Applications on Android Tablets

You're going to start seeing a lot more of this, and not just with biometrics.

Android platform adapted for biometrics (UPI)

"EB's platform is cost-effective and allows fast turnaround, which helps government agencies stay current with their handheld devices. Rather than developing small batches of high-cost, highly specialized product, EB's platform allows us to integrate low-cost modular attachments to meet the needs of specific agencies, while simultaneously producing high volumes of core products at substantially lower cost per unit."

The combination of technologies would allow a broad range of stand-along and network-connected biometric applications for U.S. government agencies, the companies said.

A couple of points:
♦ Right now, Android is the only viable option for this type of product development.
♦ The diversity of mobile hardware running Android is really impressive, and amazingly inexpensive.


See Also: Motorola Atrix 2 Has No Fingerprint Reader. Motorola designed and (briefly) sold an Android device with an on-board fingerprint reader.

In other Atrix news: Motorola Mobility Announces Availability of Motorola ATRIX in Turkey.

Sensor Manufacturers Responding to Demand for Mobile Biometrics

Fingerprint sensors for mobile ID devices (Help Net Security)

The linked article focuses on the products one firm, DigitalPersona. The market that the sensors are built to address is also well described.

These devices are increasingly being used around the world in Civil ID applications such as voting, benefits-checking and micro-finance.

Many emerging countries are turning to biometrically-enabled mobile ID terminals in an effort to combat fraud in delivering essential services such as voting, benefits, pensions and banking. These devices were initially used for military and law enforcement, but the rapid growth of Civil ID programs is creating demand for smaller, more affordable versions that deliver high-quality identification.

In order for biometric technologies to reach their full potential, several things need to converge: Algorithms, Sensors, Applications, Deployment.

Algorithm - A limited sequence of instructions or steps that tells a computer system how to solve a particular problem. A biometric system will have multiple algorithms, for example: image processing, template generation, comparisons, etc.
Sensor - Hardware found on a biometric device that converts biometric input into a digital signal and conveys this information to the processing device.
Application - The software that allows sensors and algorithms to be used to meet a need.
Deployment - Installing, setting up, testing and running. This military term, which means the placement of troops and equipment in the field, is widely used with computers as an alternate to the word "implementation."

Software companies like SecurLinx need algorithms and sensors to make applications and deploy systems. There is a huge demand for mobile applications and it's good to see the manafacutrers developing the prerequisite hardware.