New low-power FPGA video and image processing solutions speed smart embedded vision designs

Microchip, through Microsemi, has released its Smart Embedded Vision initiative that offers solutions for creating intelligent machine vision systems with Microchip’s low-power PolarFire FPGAs. With today’s announcement, the company extends its high-resolution smart embedded vision FPGA solutions with new enhanced high-speed imaging interfaces, an IP bundle for image processing and an extended partner ecosystem.

The solution initiative presents a suite of FPGA offerings that incorporates IP, hardware and tools for low-power, small form factor machine vision designs over the industrial, medical, automotive, broadcast, aerospace and defence markets.

PolarFire FPGAs offer 30% to 50% lower total power over competing SRAM-based mid-range FPGAs. With family members ranging from 100K to 500K LEs, they give five to ten times lower static power, making them excellent for a new series of compute-intensive edge devices, incorporating those deployed in thermally- and power-constrained environments.

As well as the new high-speed imaging IP cores and the PolarFire Imaging IP bundle, a new MIPI-CSI2-based machine learning camera reference design is offered for smart embedded system implementations. Based on the PolarFire FPGA imaging and video kit that employs inference algorithms from the company's partner ASIC Design Services, the reference design is free for customers to evaluate. All Smart Embedded Vision solutions are supported by the Libero SoC Design Suite, the company's comprehensive development tool.