Tamara Wilhite is a technical writer, industrial engineer, mother of two, and published sci-fi and horror author.
What are the points in favor of using facial recognition with time and attendance software? What issues can facial recognition software have, and how can you mitigate them?
First we'll look at the points in favor of facial recognition software, such as speed, accuracy when implemented right and health benefits.
3D and Infrared Cameras Equal High Accuracy
3D images and infrared cameras mean that facial recognition software today is much more accurate than its predecessors. It won’t mistake one coworker for another. Infrared and 3D images also complement each other so that facial recognition processes much faster than it did even five years ago; individuals are properly identified even if they aren’t looking straight into the camera, put in colored contact lenses or lost twenty pounds.
Since the cameras have more data points to make the correlation without having to search through increasingly large databases, so they make the correct identification of someone in a split second. And logging in to work each day is faster because people don’t have to stop what they are doing to do so.
Another benefit of facial recognition software is that it can track users and verify they are still working without slowing them down by periodically prompting for a password.
Facial recognition is contactless, unlike fingerprint scanners. This seems like a minor difference at first. However, contactless identification of employees means that your verification of someone’s identity isn’t messed up by a paper cut on the index finger. Logging in employees as they enter the building is not slowed down as you wait for them to take off their gloves in the winter. Timecard logging doesn’t compromise the product’s quality when staff where gloves to protect patients from contamination, electrical components from electrostatic discharge or glass items from smudges. People using facial recognition to access the workplace or log their time don’t lose the ability to log in because they put lotion on their hands or just put down their lunch.
The lack of contact also improves workplace health. While many people are aware of the risks of touching door handles others touched without washing their hands after sneezing or using the restroom, the same problems exist with fingerprint scanners. In many regards, the health concerns are worse when fingerprint scanners are used because you don’t have the option of protecting your hands from contact with a grimy or dirty surface with a paper towel. Everyone has to touch it. Contactless facial recognition software eliminates this problem.
Facial recognition linked to time and attendance software is virtually immune to fraud. No one can log in with someone else’s user name and password and sign them in to work even if they are late that morning. Using facial recognition to track when employees are at their work stations ensures they don’t bill the company for a long lunch. No one can sign in on the manager’s computer and approve overtime when access to the system uses biometrics they cannot falsify.
Facial recognition software has one major benefit over using employee badges for identification purposes. Badges can be lost or damaged, but you’ll always be able to access the worksite and log into work as long as you show your face when facial recognition software is used. Nor will someone be able to access the worksite using a legitimate employee badge with high quality facial recognition systems.
Facial recognition software is not an either or proposition. Security to financial systems like time cards can be further enhanced by using a combination of facial recognition software and user credentials.
The Cons of Facial Recognition Software
No technology is perfect, and facial recognition software has known problems that can prevent it from working well. What are the downsides of using facial recognition as a way to track time and attendance?
Growing out facial hair can interfere with the proper recognition of an employee via facial recognition software. This issue can be mitigated by requiring staff to have their pictures for the personnel database retaken periodically or updated if they grow out facial hair. However, shaving off significant facial hair is more likely to cause a facial recognition system to not recognize someone.
Significant Weight Changes
Facial recognition software that uses infrared cameras isn’t as prone to mistaking one person for another if they lose or gain thirty pounds, but significant weight changes can cause misidentification of individuals. You can mitigate this problem by again updating employee images in the database periodically or informing people to update their pictures when they’ve had a significant change in their appearance since the last picture was taken.
Pregnancy doesn’t cause much of an issue with facial recognition systems unless there is dramatic weight gain or serious edema.
Significant Changes in Appearance
Using infrared to augment facial recognition software does not eliminate the potential problems significant changes in appearance can create. Getting a heavy tan over the weekend may affect facial recognition software, but choosing to cut your hair much shorter or start wearing hijab probably will affect facial recognition software. Facial recognition software doesn’t work well at all when someone wears a face veil or continually wears high collars and sunglasses that partially obscure the face.
Image Quality Requirements
You should not install facial recognition as a form of biometric identification unless you have cameras that capture the quality images required for this to work well, as well as the computing infrastructure to process images quickly. Likewise, the images in your database need to be high quality or else one person could be mistaken for another by the facial recognition software.
© 2015 Tamara Wilhite