Join journalist Tommy Cummings for a weekly look at all things interesting, new, and noteworthy for design engineers.

The development of facial recognition continues as society debates the technology's merits and ethics. Are the safety and security benefits worth the threats to privacy?

We'll let those conversations rage. It doesn’t alter the fact that face recognition and the technology behind it aren't going away anytime soon. Developers can choose emerging biometric authentication products with increased computing speed and power in addition to improvements in imaging technology.

The concept of facial recognition formed in the 1960s with a system of measures to classify photos of faces, much like a fingerprinting database. Today, some smartphones, starting with the iPhone X, have facial recognition technology, which faces challenges while users wear masks prescribed in COVID-19 protocols.

As the technology evolves, this week's New Tech Tuesday takes the opportunity to highlight three facial-recognition products—a development kit, sensors, and a USB accelerator—available to designers for solution-based development.

NXP Semiconductors SLN-VIZNAS-IOT Solution for Face Recognition development kit uses the i.MX RT106F crossover microprocessor to help developers add facial recognition with liveness detection to their design. Liveness detection is designed to identify instances of image spoofing when images or objects are used to fake out facial recognition devices. The kit includes drivers for infrared and RGB cameras that can deploy the liveness detection feature to head this off. Facial recognition is done entirely at the network edge to ease privacy issues. For those in a hurry, who like to crack open the box and get going, the kit comes with fully integrated turnkey software that minimizes time to market, risk, and the development effort.

Omron Electronics B5T HVC-P2 Image Sensor Modules can enable embedded developers to add facial recognition functionality to a system without the need to understand the algorithms or optical design. Algorithms are handled entirely within the module without making any demand on the host system. The sensors feature OKAO^® Vision technology, which allows developers to implement 10 key image-sensing functions: body, face, and hand detection; face direction, gaze, blink, age, gender, and expression estimation; and facial recognition. Interestingly, it can even give a mood estimation—so be sure to smile.

Intel^® Neural Compute Stick 2 is a USB 3.0-based plug-in that's a fully functional deep-learning inference kit and self-contained artificial intelligence accelerator. Its superpower enables developers to bring computer vision and artificial intelligence to IoT and edge-device prototypes. The NCS 2 allows developers to allow smart security devices, drones, and industrial robots to see, interact, and respond. Because the NCS 2 plugs in, it has no reliance on cloud commute dependence. The NCS 2 includes the Myriad X chip found in some commercial drones moving toward obstacle avoidance and objection recognition—so, no more drone crashes.

Over the years, facial recognition technology has evolved into some aspects of artificial intelligence technology with debates over who owns the data once collected. As a result, the conversation about facial recognition's benefits and perils will continue on a global scale, just like the technology's development.