Neil Norman is the CEO of Human Recognition Systems Ltd. Here he presents four key principles behind purchasing a biometric system.
If I were to put to you the following multiple-choice question, what would your answer be: ‘You are about to embark on a journey, what form of transport do you want to take? Is it a) plane, b) a boat or c) a bicycle?’. Hopefully, you would challenge back and ask where you were going.
In the field of biometrics, we often find that our customers come to us with a firm decision in their mind as to which form of biometric they want to form the basis of their system, but without any appreciation of what use they intend to put it to. There are four key principles to consider:
Biometric technology comes in many forms and at its simplest level, measures your identity based on an image or measurement of some part of a person’s anatomy or a particular behaviour unique to them. Added to this is the fact that certain biometrics are inherently more detailed than others, for example an iris pattern is extraordinarily complex, whereas the shape of a face is less so, meaning that for matching purposes, iris recognition is significantly more accurate. And finally, people need to present their biometric to a device of some sort, so it is imperative that the environment, process and most notably the person are all conducive to providing the chosen biometric eg facial recognition through a hijab? People often think they understand the requirement, but you must explore aspects of the requirement that perhaps you would not normally consider when choosing the right form of biometric technology.
The biometric market is extremely complex with many confusing and often untrue claims being promoted by the marketing departments’ vendors and providers alike. Sadly and in common with any early-stage ‘buzz’ technology industry, it can often be difficult to differentiate between the authentic biometric companies and the snake-oil salesmen, many of which are large, well-known technology companies with over-eager marketing departments.
The simple fact is that if you want to implement a successful biometric system, you must select a biometric company with track record where you can see examples of implementations similar to the type, size and scale of your requirement. Only an experienced biometric company with upwards of 100 implementations will be able to help you work through your requirement, select the best technology solutions and most importantly, guarantee a successful integration and continued support. Ask to see examples of their work, go and visit their office to see what technology they can show you, ask who their technology partners are and so on.
Two things will change from the time you turn your biometric system on: the technology will evolve rapidly and the requirement will shift. Based on these two inevitabilities, it is important that you select a biometric system that is flexible and agile enough to adapt with time. For example, a key component of any biometric system is something we call the biometric algorithm, responsible for converting the biometric image into a searchable number or code called a template. In many cases, biometric systems do not store the biometric image and only record the template meaning that your system is not interoperable with a biometric system from another manufacturer which in turn means that you suffer proprietary lock-in.
You may not think this is a problem, but when a biometric vendor ends a product line or goes out of business, if you cannot migrate your biometric database to a new system, it becomes redundant. So, do not become dazzled by the shiny biometric device and ensure, in-keeping with principle two, that you understand other aspects of your system such as future proofing, redundancy and so forth.
As with any other field of technology, biometrics offers many options in terms of robustness and performance. For example, a Dyson may work wonders at home, but you will never see anyone using one to clean the floors in an airport. The same principle applies for biometrics. At Human Recognition Systems we have iris cameras ranging from £500 to £10,000 and like their price range, they are put to use in a range of different environments delivering completely different capabilities. We come across so many horror stories where systems have become redundant simply because a buying decision was made solely on unit price without placing focus on total cost of ownership.

Neil Norman is the CEO of Human Recognition Systems Ltd. (HRS), an award winning system integrator and solution provider specialising in biometrics, identity management and video analytics. Established in 2002, HRS has developed multi-biometric, smart surveillance and identity management solutions for companies including: ManchesterAirport, BNFL, The MoD, HMP, Carillion and LiverpoolJohnLennonAirport.
Please Login or Register.
Login or RegisterSign up for our regular newsletter containing all the best news and views from the site...
Article Comments
No reviews have been submitted
Please add a comment on this article from the Add Comment below.