Hi-resolution image recognition for driver assistance systems
The R-Car V2H SoC implements advanced image recognition technology to support high-resolution surround viewing in advanced driver assistance systems (ADAS). This is the first R-Car ADAS device from Renesas and includes additional functionality for higher performance, lower power, and improved safety in traditional and emerging, self-driving, vehicles.
Renesas applied technology from its R-Car SoCs for automotive infotainment equipment and added high-performance image recognition and low-power consumption technologies in the development of the R-Car V2H. The R-Car V2H enables embedded system manufacturers to deliver high-resolution surround-view monitoring systems, with multiple cameras, for advanced point-of-view switching. This includes rear-view and surround-view capabilities that expand what is seen by the system to every direction around the vehicle, aiding in driver oversight prevention. These types of sophisticated systems demand substantially higher processing performance in order to handle the expanded range of image recognition targets and increased processing load imposed by multiple cameras with higher pixel counts. At the same time, there is demand for both functional safety, to maintain high quality and performance, and reduced power consumption. The rapidly increasing cost of software development creates a need for better development efficiency. The Renesas R-Car V2H addresses all of these requirements, Renesas claims, in a single-chip solution.
Features of the R-Car V2H SoC include;
– Integration of essential ADAS functions on a single chip and low power consumption – The R-Car V2H SoC incorporates the IMR, a core dedicated to viewpoint conversion, the Renesas IMP-X4 image recognition core, and a 3-D graphics core, all of which provide the imaging-related technologies needed to support ADAS requirements on vehicles targeting new government regulations and insurance safety ratings. The overall architecture of the R-Car V2H is optimised in order to reduce memory bandwidth usage and overall latency. Six IMR channels can each support a high-resolution camera where the combined output can produce high-definition surround view images. The IMP-X4, in combination with the CPU, offers high performance for real-time processing of image data as it is captured from the camera. The 3-D graphics core enables a human-machine interface (HMI) that can deliver high visibility and easy identification of hazards and is optimised to support texture mapping from the camera feeds. While integrating the above functionality on a single chip, the R-Car V2H device also incorporates functions to reduce power consumption to the levels required by automotive systems. This includes a system design that balances performance and power consumption, a power cutoff function for the integrated cores, and a DDR3-SDRAM power backup function.
– Expansion of image recognition library support to include an open source library – OpenCV for improved software development efficiency – The IMP-X4 image recognition core delivers image recognition processing performance that is approximately eight times that of the earlier Renesas SH7766 and approximately four times that of the R-Car H1. The IMP-X4 incorporates six programmable image processing accelerators making possible image recognition functionality involving, for example, processing of input from multiple cameras, while altering their configuration as needed.
The Renesas exclusive image recognition library support has been expanded to include an open source library (OpenCV), substantially increasing the efficiency of software development for the developer. The image recognition library does not require fine tuning to extract full performance from the image recognition core, delivering consistently stable performance and allowing for smooth software upgrades. In addition, the library is backward compatible with the IMP series employed in the earlier R-Car H1 and SH7766 Renesas products, making it easy to reuse existing software resources.
– Ethernet AVB in combination with optimised video codecs for next-generation automotive networks and camera systems – Ethernet AVB offers features such as guaranteed bandwidth and synchronisation of multiple cameras. It is expected to become the mainstream technology for automotive high-speed networking. The R-CAR V2H includes a Gigabit Ethernet AVB controller to work with next-generation automotive networks. In combination with the introduction of the Ethernet AVB technology with optimised features for multiple-camera reception, Renesas has integrated multiple-channel video codecs supporting up to five high resolution cameras in parallel. In order to support technical transition processing of video codecs for surround view monitoring systems, R-Car V2H supports MJPEG & H.264.
R-Car V2H has been designed, Renesas says, to achieve high level functional safety. For example, ECC has been implemented on external memory and CPU caches in order to support system manufacturers’ safety concepts. Hardware CRC functionality makes it easy to achieve end-to-end protection of data transfers. In addition, the early adoption of the highly reliable INTEGRITY real-time OS from Green Hills Software is an important step toward achieving functional safety in both hardware and software.
Renesas; www.renesas.eu
Green Hills; www.ghs.com
If you enjoyed this article, you will like the following ones: don't miss them by subscribing to :
