The Human Body
For as long as we have been in existence (this could stir up some very interesting debates between the creationist’s and evolutionist’s), the human body has always marveled us. For as much as we take it for granted day in and day out, it is truly the most complex “thing” ever created. There are mysteries of the body which will we will never unlock, nor find answers for. A prime example of this is my own dad. He was a professor of biology at Purdue, and was heavily involved in spinal cord regeneration research. He studied the many facets and intracasies of the brain for many years, but just barely touched on the surface of its enormous complexity. So, it comes to no surprise, that we as a human race, try to build technology which mimicks to the best we can the human body.
Yet another good example of this are neural networks. With this, one is trying to imitate the thinking process of the human brain to make important decisions, such as in financial and commodities trading. Probably one of the most popular areas is that of computer vision. Here, scientists and engineers are trying to model the eye, and the visual processing mechanisms so that a computer, or even a robot, can literally “see”. So, it is with all of this in mind, we come to the topic of today’s posting, and what it means to Biometrics.
A biometrics/security vendor just announced a key initiative to create an open standard for computer vision: “The Khronos Group today announced a new initiative to create an open, royalty-free standard for cross platform acceleration of computer vision applications . . . Computer vision has become an essential component of many modern applications including gesture tracking, smart video surveillance, automatic driver assistance, biometrics, computational photography, augmented reality, visual inspection, robotics and more. Many modern consumer compute devices, from smartphones to desktop computers, can be capable computer vision systems but require hardware accelerated vision algorithms to work in real-time . . . The Khronos vision API will be able to accelerate high-level libraries, such as the popular OpenCV open source vision library, or be used by applications directly. A strong focus of the working group will be on providing computer vision on mobile and embedded systems and enabling acceleration on a wide variety of computing architectures . . .” (SOURCE: http://www.findbiometrics.com/industry-news/i/9428/).
MyTake
Now granted, this is really neither a technical nor a scientific breakthrough in the field of computer vision, but it is still a very important milestone, none the less. For example, it now shows that the business community recognizes the power and potential of computer vision, and how it can fit into various business schemes. But what I like most about it is that this new standard is meant to be open and royalty free. This stands in sharp contrast to other standards I know about, where they are often very closed ended, and the people who can only really benefit from them are the those that pay the high price for it.
With an open standard, all can benefit from it, but best of all, collaborative efforts can be undertaken to modify it or greatly improve it, just like open source software (such as the LAMP software architecture). Now how does all of this relate to Biometrics, you may be asking? Well, the answer can be seen as plain as day in the above quote. Computer vision will work great with behavioral biometrics, such as gesture recognition. This is an up and coming biometric technology in my view, and is used heavily used in real time market research. Well, perhaps now computer vision can be used to a technology which relates directly to it:
Namely, that of iris recognition. Perhaps the breakthroughs in computer vision can be applied here as well, in order to make iris recognition much more robust and application heavy than what it currently is (it should be noted that iris recognition has made some serious leaps and bounds in terms of development and market applications just within the last decade).
I am by no means an expert on computer vision, but this is an immediate thought of mine. But irregardless of Biometrics or not, the human body is truly to be marveled at. We probably only truly understand about 10%, the rest of the 90% will be truly a remarkable and breathtaking journey.
Comments