Howe Technology and GEO Semiconductor jointly develop RGB-IR for car cabin monitoring system

Author：合众北斗长和电子科技（深圳）有限公司 Click： Time：2019-05-27 11:25:03

Howe Technology and GEO Semiconductor jointly develop RGB-IR solutions for automotive cabin monitoring systems

release time：May 22, 2019 15:05 announcer：eechina

In-cabin monitoring , OV2778 , Image Sensor

The solution features the industry's smallest and most cost-effective 2MP RGB-IR automotive image sensor with video processor for all major vehicles

OmniVision Technologies, Inc. and GEO Semiconductor today jointly launched a car cabin surveillance imaging solution that simultaneously captures and processes color (RGB) and infrared (IR) images, enabling a single camera to solve high quality problems. Day and night imaging. The solution combines OmniVision's OV2778 2MP RGB-Ir image sensor – delivering the industry's best performance and sensitivity in all lighting conditions, and GEO's dedicated hardware for processing RGB-IR sensors – the GW5200 for machine vision And visual applications provide high-quality video and provide an easy-to-integrate solution for detecting objects and unattended children, remote monitoring, and video recording of spell/rental and unmanned taxi vehicles.

Figure: Howell Technology and GEO Semiconductor jointly develop an RGB-IR solution for automotive cabin monitoring systems.

“Our partnership with OmniVision is due to OmniVision's leading image technology and successful experience working with global automotive Tier 1 suppliers," said John Casey, vice president and general manager of automotive marketing at GEO. "GEO's GW5200 has RGB-IR processing. The unique capabilities of the sensor provide superior low-light performance in challenging vehicle interiors and accelerate the adoption of advanced in-cabin cameras.”

“GEO's products have industry-leading image signal processing capabilities and a deep understanding of customer pain points," said Andy Hanvey, director of automotive marketing at OmniVision. "After OmniVision recently introduced the OV2778 image sensor, we both saw the joint development of this solution. demand. This solution enables designers to design and manufacture high-value cabin monitoring systems for mainstream vehicles more easily and quickly.”

Based on OmniVision's 2.8 micron OmniBSITM-2 Deep WellTM pixel technology, the OV2778 image sensor provides 16-bit linear output with a single exposure and the best low light sensitivity of its class. In addition, the GW5200's integrated image signal processor (ISP) features an advanced noise filter that further enhances low-light image quality. In terms of power consumption, the total power consumption of the two devices under common operating conditions is less than 750mW, reducing the heat generated by the continuous operation of the cabin surveillance camera in a confined space to ensure the best image quality.

The GW5200 processor supports the 120dB high dynamic range (HDR) output of the OV2778 image sensor, ensuring excellent image detail in bright and low light scenes. In addition, the GW5200 includes GEO's 5th generation eWARP® geometry processor for image correction of fisheye lenses with over 180 degrees of field of view. In addition, the OV2778 sensor combines RGB-IR, 4x4 pattern color filters and external frame synchronization to achieve top color imaging performance in different lighting conditions.

The joint solution occupies a small footprint, enabling designers to create a minimum camera module with a 120dB dynamic range that can be placed in an inconspicuous placement inside the vehicle. The OV2778 image sensor is available in the smallest package (6.5 x 5.7mm automotive CSP) for the automotive in-car market. The GW5200 camera video processor is available in a 7 x 7mm BGA package.

Two devices in the joint solution are available now, offering advanced ASIL functional safety and AEC-Q100 Level 2 certification for automotive applications.

Prev : 5G research and development has made many progressNext : The core of the first university's future automotive human-computer interaction design competition,