Facial recognition has become one of the most significant biometric technologies in workforce management, particularly for time and attendance.<\/p>\n
At GT Clocks, our journey with facial recognition began before the pandemic, but accelerated when businesses needed touchless solutions.<\/p>\n
Along the way, we\u2019ve learned what works, what doesn\u2019t, and the real-world factors that determine successful deployment.<\/p>\n
Time is money, and time theft (aka time clock fraud, payroll fraud, and \u2018buddy punching\u2019) is still prevalent in many workplaces.<\/p>\n
A recent report suggested businesses globally lose around four-and-a-half hours per employee per week to this practice. In many roles, that equates to 10\u201315% of total working hours, or about 2.2% of payroll costs.<\/p>\n
Facial recognition offers a simple, effective way to tackle these issues. Today\u2019s solutions are highly reliable and remove the possibility of buddy punching, ensuring accurate, secure, and touchless clocking.<\/p>\n
For organisations looking to reduce fraud and improve efficiency, it\u2019s become an obvious choice.<\/p>\n
In early 2020, GT Clocks was immersed in engineering a robust product roadmap centered on our cutting-edge time clock<\/a> \u2013 the GT8.<\/p>\n Positioned as a flagship design, this timeclock aimed to be a powerhouse of features, ensuring top-notch security and performance.<\/p>\n The success of our initial GT10<\/a> Android 6 device propelled us into significant HCM opportunities, especially in the thriving North American market.<\/p>\n Simultaneously, we were laying the groundwork for GT Connect<\/a>, an innovative SaaS Cloud-based clock management platform designed to encompass middleware, PII\/Biometric data compliance, and seamless data distribution.<\/p>\n At that time, our biometric roadmap was in a steady state, fingerprint technology provided by recognized industry leaders such as HiD\u2019s Lumidigm and Suprema SF range were fully integrated into our timeclock solutions, along with advanced on clock capability for 1:N identification into the tens of thousands.<\/p>\n Tried and tested technology, ideal for the HCM industry and adopted as the Defacto solution for biometric identification or verification needs for HCM applications.<\/p>\n However, the tides turned swiftly with the onset of the pandemic in early 2020, forcing us to reevaluate our product roadmap amidst the uncertainties that businesses across the globe faced.<\/p>\n A notable shift during the pandemic was the emphasis on touchless operation, particularly impacting the use of fingerprint technology. This prompted us to revisit facial recognition technology, previously considered less in demand in the HCM market.<\/p>\n Our challenge was clear \u2013 develop a facial recognition solution that seamlessly integrated with our existing devices, offered swift and user-friendly operation, ensured the utmost security in handling facial data, and, of course, remained cost-effective.<\/p>\n GT Clocks<\/a> had, in fact, evaluated facial recognition prior to this, but conversations with our existing customers led us to believe there was insufficient demand in the HCM market.<\/p>\n Fingerprint technology was regarded as mature, reliable, and an accepted solution for both identification and verification applications.<\/p>\n And whilst it is fair to say facial recognition has been around for some time in the market, it is also fair to say it was mainly used in high-end security applications.<\/p>\n Entering a secure facility is one thing, but clocking in 100 people starting a shift where every second is either their time or your money is entirely another.<\/p>\n So that was the challenge we faced, and I wish I had known back then what we have learned as a company since that day.<\/p>\n To set the scene, we already had many thousands of devices in the field, so whatever solution we developed needed to operate where possible with our existing solutions. The operation had to be as fast and easy to use as the established solutions, and importantly, it had to be secure, as we were capturing and processing people\u2019s facial data.<\/p>\n Finally, and most technically challengingly, it had to be cost-effective, as the incremental cost of the technology could not move the overall solution outside of the market price for a timeclock.<\/p>\n After a round of rapid prototyping, we made the strategic decision to leverage the onboard cameras in our GT8 and GT10 clocks. Through the development of sophisticated algorithms, we achieved facial recognition capabilities, enabling both verification (Card or Pin + Face) and identification (Face only).<\/p>\n Beyond the core solution, we incorporated features like Mask Detection and thermal screening, directly addressing the unique demands arising from the pandemic.<\/p>\n The choice of Visible Light and IR Cameras in the GT8 proved to be a good one, especially in diverse lighting conditions.<\/p>\n On a bench, we rapidly developed a demonstrable prototype that was performant and user-friendly, able to recognize a face and match 1:10,000 faces in around a second on our GT8 Clock. So far, so good! Then, the real work started translating that to the real world, making it work for our partners in the many environments it needed to.<\/p>\n It became very quickly apparent that the environment and lighting were key factors, and luckily, the technology choice we had made of Visible Light and IR Cameras in the GT8 turned out to be the right one.<\/p>\n Our GT8 device is equipped with a combined Visible and IR capable camera, augmented with both LED (white) and IR illumination, both of which significantly improve the solution in poor lighting environments.<\/p>\n But technology only takes us so far, and whilst we continue to develop facial recognition software updates and features such as temperature screening, the reality is that the installation and environment really do count.<\/p>\n Also key here is the customer\u2019s perception of what face recognition can do \u2013 can it recognize the face of someone wearing a large mask, glasses, and a hat, for example? The simple answer is no.<\/p>\n This is impossible if there is not enough information (or, put another way, points of minutia available for the algorithm to work effectively). It may be possible to verify where it\u2019s a one-to-one match, but it’s generally not suitable for identification.<\/p>\n It\u2019s not just about the technology, as much as it is about effectively communicating the real-world capabilities of facial recognition technology and, importantly, how it can be successfully implemented in differing environmental conditions.<\/p>\n Successful implementation involved addressing various considerations, from compliance and terminal placement to user spacing, lighting conditions, enrollment processes, and even clock cleaning.<\/p>\n Despite the challenges, our facial recognition solution emerged as a powerful and reliable tool, efficiently processing staff in a touchless manner while safeguarding end-user information.<\/p>\n GT Clocks, along with our customers, has learned a huge amount about successfully implementing face recognition, and here are a few of the considerations to think about:<\/p>\n Based on our experience, successful implementation depends on more than just the technology itself. Here are some of the key factors to consider:<\/p>\nDeveloping a Practical Facial Recognition Solution<\/h2>\n
What to consider when implementing a face recognition solution?<\/h2>\n
Facial Recognition Compliance and Attestation<\/h3>\n