Monday, June 12, 2017

FPGA enables energy-efficient embedded vision processing at the edge

By Nick Flaherty www.flaherty.co.uk

An FPGA from Lattice Semiconductor is being used for high performance image recognition systems at the edge of the Internet of Things (IoT), highlighting the importance of edge processing to reduce power consumption. 

The ECP5 FPGA is being used for CPU acceleration for license plate detection and image enhancement in intelligent traffic cameras, as well as image stitching and 3D merging for Advanced Driver Assistance Systems (ADAS) 360 surround view systems.

Intelligent traffic systems (ITS), including traffic flow monitoring, traffic violations identification, smart parking, and toll collection are a key part of the vision of tomorrow’s Smart Cities. Such systems typically require intelligent traffic cameras that can accurately detect many aspects of a vehicle, such as license plates, even in harsh environments, to perform video analytics at the Edge, rather than sending raw video streams back to the Cloud. Shanghai Microsharp Intelligent Technology (Microsharp) is using the ECP5 for intelligent traffic cameras with up to 95 percent license plate recognition rate.

“When seeking a partner to collaborate on our intelligent traffic camera, we were looking for a scalable, low power solution to ensure flawless real-time license plate image capture,” said Ting Zhou, CEO of Microsharp. “Lattice’s ECP5 FPGA was a natural fit, allowing us to take advantage of the product’s energy efficient image enhancement and processing capabilities to accelerate the development of our intelligent camera.”

360 degree surround view system is a popular automotive ADAS technology for parking assistance. These systems typically capture their surroundings using at least four cameras facing different directions around the car to generate a composite view for the driver to assist with parking and low-speed maneuvering. The NEX-ADAS 360ยบ 3-D Surround View Monitoring Technology from Shenzhen Moorechip Technologies (Moorechip) synthesizes images from four cameras to create a 3D view of a vehicle's surroundings.

“Lattice’s ECP5 FPGA family offers the low power, small form factor and flexible connectivity requirements needed to support our ADAS 360 surround view system, providing advanced, high-quality image-stitching technologies to further differentiate our product from others on the market,” said Andrew Liu, General Manager at Moorechip. 

The demonstrations come after Lattice was bought by the Canyon Bridge fund with backing from a Chinese state fund.

Related stories:

No comments:

PLATINUM SPONSOR

South West Innovation News - news from across the region for oneof the world's hottest tech clusters