Chip designer Arm has introduced an image signal processor (ISP) to advance vision systems for IoT and embedded markets.

The Arm Mali-C55 is said to be the company’s smallest and most configurable image signal processor, and is already seeing success with licensees such as Japanese chip maker Renesas.

It delivers improved image quality and higher performance for embedded and IoT vision systems with half the silicon area of previous generations. There are multi-camera high resolution support and seamless integration for more on-device machine learning.

“Image signal processors continue to be one of the most important information-generating devices, supporting a broad range of IoT vision system applications including commercial, industrial or home smart cameras and drones,” said Mohamed Awad, vice president of IoT at Arm. “With increased demand for both more and higher quality image processing in future devices, Arm’s ISP technology roadmap is an area of continuing investment. Today, we are pleased to share the latest results of that work with the launch of the new Arm Mali-C55 ISP, the smallest and most configurable ISP from Arm to date.”

The Mali-C55 delivers upgraded image quality features, works under a wide range of lighting and weather conditions, and is designed to enable increased performance and capability in area and power constrained applications, making it suitable for smart camera and edge AI vision use cases.

This technology should deliver improved capabilities in several markets: surveillance and security cameras will be able to detect more critical detail, such as recognising the exact information on licence plates traveling at up to 120kmph; home cameras and security systems will capture higher resolution images both inside and out; and smart home hubs can include features such as secure visual unlock.

“For developers, we’re providing a complete software package for controlling the ISP, as well as a full set of tuning and calibration tools to help our partners to achieve the desired image quality for their application,” said Awad.

With multi-camera capability for up to eight separate inputs, support for image resolutions up to 8K and a maximum image size up to 48Mpixel, the Mali-C55 provides a combination of image quality, throughput, power consumption and silicon area.

Building on the Mali-C52 ISP, it enables image quality through features including improved tone mapping and spatial noise reduction, enhanced support for high dynamic range (HDR) sensors and seamless integration with machine-learning accelerators to take advantage of neural networks for various de-noising techniques. By combining multiple Mali-C55s, larger image sizes can be achieved for applications that require greater than 48Mpixel capabilities, such as video conferencing.

“For embedded and IoT vision applications, silicon footprint and cost are important factors and, with the Mali-C55, we’ve provided these enhanced features in almost half of the silicon area size of previous generations, significantly lowering power consumption for extended battery life, and in the process also lowering the cost of these devices,” said Awad.

As machine learning moves closer to the edge, advanced image processing can be leveraged by integrating more ISPs into the SoCs. By enabling integration between Mali-C55 and machine-learning accelerators, Arm says it is delivering new levels of on-device processing in devices that require high-quality vision systems because the output from the ISP can be sent directly to the ML accelerator. This reduces cost and processing time by having less data sent from device to cloud, without having to compromise on inferencing.

“The applications that require ever-advancing vision systems are wide-ranging, and so imaging technology has to be adaptable to specific market requirements,” said Awad. “For example, a low-cost home camera system may require a simple, limited set of features whereas a commercial camera might demand more sophisticated capabilities such as high resolution, noise reduction and enhanced security. Silicon partners and OEMs need the freedom to easily add or remove features in line with their application requirements, and that’s why we’ve designed the Mali-C55 to be our most configurable ISP.”

In addition to the multi-camera support and integration with ML accelerators, the Mali-C55 also includes standard AXI and AHB interfaces for integration with either Cortex-A or Cortex-M-based SoCs.

“The Mali-C55 ISP illustrates Arm’s continued pursuit to provide industry-leading imaging technology,” said Awad. “This new technology will enable SoC architects, imaging experts and embedded software developers to create market-leading products in the growing smart camera and edge AI vision markets.”