Skip to main content

Accelerate Embedded Vision design with Field Programmable Gate Array (FPGA)

According to Embedded Vision Alliance, “Embedded Vision” refers to the practical use of computer vision in machines that understand their environment through visual means. Computer vision is the use of digital processing and intelligent algorithms to interpret meaning from images or video. 

Traditionally, computer visions are implemented in computers or processor boards such as General Purpose Graphic Processor Unit (GPGPU) and Digital Signal Processor (DSP). It is always software play. However, algorithms used in computer vision becomes more complex  and  computational intensive. This leads to software bottlenecks, high power consumption and performance degradation. Designers, developers and engineers start to implement the computer vision design on reconfigurable device especially field programmable gate array (FPGA) to optimize the real-time performance and reduce the power consumption. FPGA also offers a reprogrammable solution for hardware acceleration. This can greatly reduce the design cycle. 

Xilinx, the leading FPGA manufacturer, provides embedded vision developers with a suite of technologies that support both hardware and software design. Incidentally, Xilinx’s Zynq integrates the software programmability of a processor (ARM Cortex A9) with the hardware programmability of an FPGA, providing developers with unrivaled levels of system performance, flexibility, and scalability. Many embedded vision applications are built around Xilinx Zynq

Digilent has recently released an Embedded Vision Bundle based on Xilinx Zynq 7020 This is a valuable tool for anyone aiming for developing their own embedded vision products and applications, or to learn more about embedded vision concepts. The kit includes Digilent (164-3487)  Zybo Z7-20 Zynq-7000 ARM/FPGA SoC development board, the (174-1555)  Pcam 5C 5 megapixel color camera module, and the 15-pin flat-flexible cable (FFC).

Embedded Vision Bundle

Accelerate embedded vision through Zynq-7000 ARM/FPGA SoC Platform

The (164-3487)  Digilent Zybo Z7 is the newest addition to the popular ZYBO line of ARM/FPGA SoC Platform. The Zybo Z7 surrounds Xilinx Zynq with a rich set of multimedia and connectivity peripherals to create a formidable single-board computer, even before considering the flexibility and power added by the FPGA. Hardware designers and software developers can seamlessly integrate FPGA and CPU functionality.  

The ZYBO Z7's video-capable feature set includes a MIPI CSI-2 compatible Pcam connector, HDMI input, HDMI output, and high DDR3L bandwidth. The board comes with a SDSoC voucher. The SDSoC™ development environment provides a familiar embedded C/C++ application development experience including an easy to use Eclipse IDE and a comprehensive design environment for heterogeneous Zynq® All Programmable SoC.

Zybo Z7 FPGA SoC Platform

Capture HD video signals through 5 MP Fixed Focus Color Camera Module

Digilent (174-1555)  Pcam 5C is an imaging module meant for use with FPGA development boards. The module is designed around the Omnivision OV5640 5 megapixel (MP) color image sensor. This sensor includes various internal processing functions that can improve image quality, including automatic white balance, automatic black level calibration, and controls for adjusting saturation, hue, gamma and sharpness. Data is transferred over a dual-lane MIPI CSI-2 interface, which provides enough data bandwidth to support common video streaming formats such as 1080p (at 30 frames per second) and 720p (at 60 frames per second). The module is connected to the FPGA development board via a 15-pin flat-flexible cable (FFC). The Pcam 5C comes with a set of open source Xilinx Vivado IP cores, a 10 cm flat-flexible cable and a factory-installed fixed focus lens with M12 lens mount, so it is ready to use out of the box.

PCAM 5C 5MP Camera Sensor Module

Getting Started Guide

Digilent provides a demo project to get users started with the bundle. The demo passes video data from the image sensor out through the (164-3487)  Zybo Z7 HDMI port to a display. All IP used in the demo is free and open-source, including a D-PHY receiver, CSI-2 decoder, Bayer to RGB converter and gamma correction.

awong has not written a bio yet…

Comments