Arm offers smaller chip, better images for computer vision systems
UK chip-designer Arm has announced the launch of a new image signal processor which the company says is its smallest and most configurable ISP yet. Arm’s Mali-C55 image signal processor is primarily designed to provide improved image quality and better performance for complex IoT vision applications. These capabilities are packaged on a chip that is half the silicon area of the previous generations of Arm ISPs.
Arm Mali-C55 is the updated version of the existing Mali-C52 ISP, delivering upgraded image quality features while enabling the device to operate efficiently under different lighting and weather conditions. The design allows it to be used in power and space-constrained AI vision applications, such as surveillance and security cameras. The support for multi-camera capability provides developers with the flexibility to connect the device with up to 8 separate inputs with an image resolution of up to 8K. With a maximum image size of 48 megapixels, the Mali-C55 image signal processor is said to be the “most efficient” combination of image quality, throughput, power consumption and silicon area.
“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 Mohamed Awad, vice president of IoT and Embedded, Arm.
The significant upgrades from the Arm Mali-C52 ISP show improvements in the sensor interface, throughput, and maximum resolution. The new Arm Mali-C55 ISP provides up to 8 sensor interfaces compared to 1 in Mali-C52, the throughput of 1200Mpix/sec compared to 600Mpix/sec, and resolution support for 8192×6144 pixels than 4608×3456 pixels in the previous generation.
On the software side of the Arm Mali-C55 ISP, the manufacturer has provided a complete software package for tuning and calibration functionalities to help customers achieve the most efficient image quality for their custom applications. For pre-tuned evaluation purposes of the Mali-C55 image quality, Arm offers a bit-exact simulation model with a reference platform.
“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 compromising on inferencing,” Awad further added.
As part of the Arm’s latest roadmap of Total Solutions for IoT, the company spills the news of the upcoming Total Solution for vision systems that will be capable of integrating the new Arm Mali-C55 image signal processor. The Mali-C55 ISP will be shipped with a pre-validated solution to support a specific sensor and dual output mode for connecting the device to a machine learning accelerator.
Article Topics
ARM | chip | computer vision | edge AI | IoT
Comments