Japanese electronics company Renesas has launched a microprocessor to process data from multiple cameras in applications such as AI-equipped gateways, video servers, security gates, PoS terminals and robotic arms.

The device enables AI processing of image data from multiple cameras, offering accurate image recognition for vision AI applications. Equipped with two 64bit Arm Cortex-A53 cores, the device can deliver high computing performance with a maximum operating frequency of 1GHz.

The RZ/V2MA uses a proprietary low power DRP (dynamically reconfigurable processor) AI accelerator that can process vision AI at one tera operations per second, per watt (Tops/W).

High-speed interfaces such as Ethernet, USB and PCI Express allow image input from multiple external cameras. In addition to the DRP AI accelerator, it includes an Open CV accelerator that allows rule-based image processing simultaneously. These features bring accurate image recognition capabilities for machine-vision products such as AI-equipped gateways, video servers, security gates, PoS terminals and robotic arms.

“Renesas’ RZ/V series is ideal for embedded devices since it does not need fans or heat sinks, due to its extremely low power consumption and low heat capability when running AI,” said Chiharu Nakabayashi, president of Yokogawa subsidiary Amnimo, a provider of IoT and AI-based services. “With these devices, we are confident that we can develop powerful image AI gateways that can be installed anywhere.”

There is a suite of development tools to aid vision AI system design. In addition to the existing DRP-AI translator, the device adds DRP-AI TVM, which is based on the open-source deep learning compiler Apache TVM technology. While DRP AI translator is designed to convert AI models into DRP-AI executables, the TVM compiler lets the DRP-AI accelerator work with the CPU, allowing DRP-AI to convert and generate more AI models. As a first phase, Renesas supports Onnx and PyTorch AI models and plans to support Tensorflow in the future.

“One of the challenges for embedded systems developers who want to implement machine learning is to keep up with the latest AI models that are constantly evolving,” said Shigeki Kato, vice president at Renesas. “With the new DRP-AI TVM tool, we are offering designers the option to expand AI frameworks and AI models that can be converted to executable formats, allowing them to bring the latest image recognition capabilities to embedded devices using new AI models.”

Renesas has developed a vision AI gateway, which is an AI-based object detection and recognition platform that uses multiple cameras to collect and transmit data wirelessly. This high-speed processing option combines the RZ/V2MA MPU with complimentary Renesas products such as power ICs, VersaClock clock generator and communication modules for wifi, Bluetooth and LTE. This not only provides flexible connectivity but has an optimised power supply and has been tested to accelerate development of robust AI gateway devices.

• Energy harvesting firm EnOcean has acquired the assets of Renesas’ edge computing business. The deal includes the transition of personnel as well as hardware and software products and was closed at the start of this month.

“This acquisition brings us, our partners and our customers one step closer to a carbon-neutral future,” said EnOcean CEO Raoul Wijgergangs.