One of the latest technologies to enter the biometrics market is facial recognition. The earliest versions were used in law enforcement and airport security and in general they were a complete failure. In essence, this was because facial recognition only works well for those who want to be recognized.
When you understand the principles used in facial recognition it become very apparent why the technology is unreliable for clandestine identification. The facial recognition technology uses the geometry of the individual’s face to create a representative algorithm. A camera, and a light source are used to map the size, and distance between the most prominent features of the face (generally the eyes, nose and mouth) and these are used to construct this mathematical representation. To make yourself unidentifiable all you need to do is to wear sunglasses or a scarf.
More recently the technology has shown considerable promise in less security focused applications such as time and attendance. In time and attendance applications time clocks fitted with facial recognition technology are proving to be quite reliable and this is essentially because the operating environment is controlled and the individual cooperates in the process. The use of facial recognition is becoming more widespread, and is a very effective way of verifying and identifying people.
Here’s what you need to know:
- Facial recognition technology is non-intrusive, so you don’t have to provide fingerprints, or stop while a machine checks your retina. The camera used in facial recognition technology is effective from about a meter away, and so the person doesn’t always know that they are being identified, in a similar way to the fact that people aren’t always aware that they are on CCTV.
- This method of identification is accurate, and is becoming even more accurate as the hardware and software improves, and the technology is used by more and more people, and improvements are made.
- Facial recognition systems often require the person to smile or blink, or produce another human movement, in order to prevent against imposters who might wear a mold or mask from trying to gain access, or impersonate somebody else.
- Official documents such as a passport or driving license can utilize face recognition technology, in order to prevent against all sorts of criminal activity stemming from the production and use of false documents. The use of multiple documents from the same person will also be reduced.
- Facial recognition can be used for many other applications, such as authorizing transactions, perhaps instead of using a PIN number or signature.
- Physical access can be granted or denied to a building, or department, using face recognition technology rather than by using a traditional access code, or swipe card.
- Time and attendance systems are another obvious choice to use face recognition systems, as a way to combat people either clocking in or out early or late, or clocking in or out for an absent colleague.
- By utilizing facial recognition, people are not disclosing any information that isn’t on public view anyway. You don’t need to provide lots of very personal data.
- Having your photograph taken is socially acceptable, and happens regularly for many people. Using a photograph for identification and verification purposes is much less likely to be met with objection.
- The image used for facial recognition can be captured from an approved photograph, perhaps a driving license or passport. This means that the person doesn’t have to be present. This is obviously of huge benefit for companies with a lot of employees, or for government departments, and in many other situations.
Now you know more about facial recognition, and how it can help you, why not see how you can incorporate it into your security systems? It is the best means present to verify identity and protect a premise from intruders. It is a highly developed system and functions by scanning the unique and permanent facial characteristics of the individual. High accuracy in identity verification is guaranteed by its vendors. Its working can be as given below:
- The facial recognition system is installed at any access point of the premise. The terminal unit and the camera are installed at the access point. The database can be kept in the same PC or in the central server.
- The authorized individuals are required to enroll themselves into the foremost system. It is believed that there are eighty nodal points on the face. Few of these nodal points are scanned by the system. E.g. the distance between the two eyes, nose and eyes, length of the lips, etc. They are distinct for every human being and remain the same throughout their life.
- This geometrical structure of the face is stored in the database as face templates. The structure of the face, in terms of nodal points, consists of points and is stored in the database as face templates of the person by the facial recognition device.
- Whenever the person appears before the camera, next time onwards, he will be recognized by the system. His current facial characteristics are scanned in terms of nodal points. These facial details are tallied with all the face templates stored in the database. If the person is enrolled, the facial structure is sure to match with one of the templates stored.
- When the facial structure is found matching with one of the templates stored, the face is considered identified. Details that are stored corresponding to the template flashes on the screen and the identity of the person making the transaction is established.
Body geometry identification is not new – hand scan technology has been available for nearly 20 years. It relies on the geometry of the hand rather than the geometry of the face and is one of the most reliable biometric products available. Incidentally, when we talk about reliability in biometric circles we mean, in simple terms, that it nearly always recognizes your first time and it never mistakes you for someone else.
The latest generation of facial recognition time clocks is quite reliable. Recognition speed is very quick – it is faster than a fingerprint reader and almost as fast as an RFID card. The pricing is very competitive in comparison with fingerprint and RFID and certainly much cheaper than hand scan, iris scan and vein scans. The downside at the moment is that facial recognition time clocks are in the main being exported from China and they reflect the immaturity of the product and its lack of penetration into western business environment. At this stage there are only one or two products which have been integrated into mainstream time and attendance products suitable for implementation in the more complex award structures found in the US, UK, Europe and Australia for example.
If you are looking for a reliable facial recognition time clock it would be well worth your effort to find a local supplier with a good track record in facial recognition products. Edit this text supplier will have all the integration and software options available to you. Don’t take this choice lightly as the greatest advantage for your business usually lies in what comes out of the time clock in labor reports.