OmniVision Technologies, Inc., announced the availability of an evaluation kit developed in collaboration with Jungo Connectivity. The kit combines Jungo’s award-winning CoDriver software development kit (SDK)—featuring deep-learning, machine-learning and computer-vision algorithms—with OmniVision’s OV2311, a high-resolution CMOS image sensor built on the OmniPixel3-GS™ global shutter technology. This evaluation kit makes it easy for OEMs and tier-1 automotive designers to develop the next generation of driver and occupant monitoring systems.
According to the National Highway Traffic Safety Administration, 80% of car accidents are caused by distracted driving. Carmakers are constantly looking for new ways to mitigate this issue, such as enhancing their advanced driver assistance systems (ADAS) to allow for real-time monitoring, alerts and other measures. In semi-autonomous vehicles, features are being added that allow the onboard computer to seize or relinquish control of the vehicle based on the driver’s condition. In fully autonomous scenarios, the car systems will even acquire information about the passengers’ well-being, mood and overall condition.
“Increasingly advanced capabilities are necessary to enable safer and more intelligent automotive systems that will power not only next-generation ADAS, but also the first fully autonomous vehicles,” said Cliff Cheng, senior director of automotive marketing at OmniVision. “For all of these next-generation occupant monitoring and identification applications, the ability to perform optimally in low- or no-light conditions is a must. OmniVision and Jungo’s combined solution provides customers such high performing, time-to-market solution.”
The combination of Jungo’s SDK and OmniVision’s image sensor provides automotive designers with the tools to obtain the most complete, real-time picture of the driver’s condition—regardless of the lighting conditions. Vehicles designed using this evaluation kit will be able to accurately determine whether the driver is ready to take control in a semi-autonomous emergency scenario. If not, the vehicle can take an alternative action, such as pulling off the road and parking. In a fully autonomous experience, the sensors and software provide reliable information about the passengers’ characteristics, possessions, and emotional and medical states. OmniVision’s high-performance sensing capabilities are crucial for Jungo’s CoDriver software to perform accurate facial recognition—not only of the driver, but also of passengers as far back as the third row.
“We partnered with OmniVision because their OV2311 image sensor offers optimal performance and compact form factor for in-cabin monitoring applications,” said Ophir Herbst, CEO of Jungo Connectivity Ltd. “In combination with our CoDriver SDK, which provides a user-friendly environment for designing with our leading-edge deep-learning, machine-learning and computer-vision algorithms, automotive developers have everything they need to quickly take their occupant monitoring and identification systems to the next level.”
The kit’s OV2311 high-resolution, 2-megapixel, 3-micron pixel-size sensor features a high frame rate of 60 frames per second and a small optical form factor of 1/2.9 inches—all in an ultra-compact automotive chip-scale package (a-CSP™) that is Automotive Safety Integrity Level (ASIL) B and AEC-Q100 qualified. Additionally, the sensor can control an infrared LED to sync with the exposure interval to capture extremely clear images in low- or no-light conditions. The OV2311 also achieves high near-infrared quantum efficiency, which minimizes the amount of LED illumination power required in total darkness.
Version 1.5 of Jungo’s full CoDriver SDK is included in the kit, with all the APIs, computer-vision algorithms and drivers needed by automotive designers, featuring both whole-cabin and driver-only monitoring configurations. The SDK has been configured and rigorously tested to perform optimally with OmniVision’s OV2311 image sensor, in real time and using a live video stream.