Computer vision is almost always the first application that comes to mind when anyone starts talking about industrial IoT applications, and its increasing appeal means that image sensing and processing capabilities need to be embedded in a variety of device form factors and footprints.
Mohamed Awad, VP of IoT and Embedded at Arm said in a blog post this week that the company kept that in mind while developing the Mali-C55 image signal processor, what he described as the smallest and most configurable image signal processor from Arm. The successor to the Mali-C-52, the new processor covers just half the silicon area size of previous product generations, he said.
Already licensed by Japan’s Renesas Electronics, an existing Arm customer for IoT-focused processors, the Mali-C55 supports multi-camera capability for up to eight separate inputs, image resolutions up to 8K and a maximum image size up to 48 megapixels, with low power consumption, translating to lower costs. Combining multiple units allows for applications like video conferencing. The Mali-C55 also works under a variety of lighting and weather conditions, and can process images of fast-moving objects to such a degree that it can read a license plate on a car going 75 mph.
“Image signal processors (ISPs) 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,” Awad stated in the blog post. “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.”
Awad also said that Arm’s Total Solutions for IoT portfolio, announced last year, includes a Total Solution for vision offering, which will integrate the Mali-C55 and be available in the near future.
He also noted that the new ISP integrates with machine learning accelerators in a way that reduces processing time and cost. “As ML moves closer to the edge, advanced image processing can be leveraged by integrating more ISPs into the SoCs. By enabling easy integration between Mali-C55 and machine learning accelerators, we’re 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.”
This week’s announcement comes after the April unveiling of Arm’s Cortex-M85 processors focused on machine learning and security capabilities for IoT applications.