Our Good ‘Ole Fingerprint
As I write constantly about, this is probably one of the most traditional ways (along with the Social Security number) for us to be identified and verified, not only in every day life, but also from the standpoint of law enforcement. Even in Biometrics, the fingerprint is the most widely used and accepted form of technology. And in the world of Fingerprint Recognition, the quality of the image, or the picture is what is most important. And this, in turn, is a function of the sensor that is used. Today, in regards with Fingerprint Recognition, there are three main types of sensor technologies used: 1) Optical; 2) Thermal; and 3) Ultrasound.
Of these three, it is optical which is used the most, all the way from the largest of the devices to the smallest SSO on your Smartphone. But lately, there is a newer type that is coming out, which is known as “multispectral imaging”. And today, I found an excellent description of it, which I wanted to share all of you.
Here we go: “Multispectral imaging uses multiple wavelengths of light to capture an image of both a person’s external fingerprint and their identical internal fingerprint, the foundational capillary bed. Even if the external print is damaged or obscured, multispectral fingerprinting uses the internal print data to ensure clear, clean images of anyone in any environment. High-quality images are captured by multispectral sensors even when fingerprint ridges are hard to distinguish due to genetics, age, dirt, finger placement, wetness, dryness and other environmental conditions. With multispectral imaging, the two to five percent of people with “problem fingerprints” that fail to enroll on readers with traditional technologies can enroll as easily as a young graduate student working in a clean room.” (SOURCE: http://www.biometricnewsportal.com/biometrics_news3.asp?articleid=266608&arttitle=Multispectral fingerprinting uses the internal print data to ensure clear, clean images of anyone in any environment).
My Take
About 95% of the time, or even higher than that, optical technology always works very well, and can enroll/verify/identify most end users of the system. But however, Biometrics is again, just another piece of technology, and there will be a certain percentage of the population which simply cannot have their fingerprint measured by the Biometric system. This could be a for a whole host of reasons, ranging from environmental considerations, to the physical ailments of the end user. As you can see from the above quote, is designed to capture the image of the fingerprint from both the outside and the inside of the finger.
The traditional optical technology, for the most part, just takes an image of the fingerprint from the outside. And in this regard, the environment and the finger must be clean and dry, in order to capture the best and most unique samples. So, this is where the benefit of multispectral imaging takes place-it acts as a fail safe, because the internal image is captured as well. In my view, the technological advancements that can be made in Fingerprint Recognition has probably reached a saturation point from a macro standpoint. Most of the advancements will be made in the lesser known areas, such as how to capture the best fingerprint image possible. A key competitor to this is even Vein Pattern Recognition, where an infrared beam light is shone in order to penetrate the skin and capture the image of the vein patters.
But, multispectral imaging also holds its business benefits as well. Remember that old 80-20 rule, where 20% of your customers bring in 80% of your revenue??? Well, the same holds true here as well. It is that 5% of the people who cannot be verified by optical technology which will give 95% of the headaches and hassles to the Biometrics Industry, and to Corporate America. If you are an SEM, this will not pose too much of a problem. But if you are Fortune 50o company, this statistic can be magnified by 10x. Hopefully, with the dual image taking of multispectral imaging and the other advancements which come out, this statistic will be totally eliminated, and thus imposing a much lowered Total Cost of Ownership (TCO) on Corporate America.