Embedded special Embedded product update

Vision Accelerator Programme The Embedded Vision Alliance has launched a service for its members to help build products using computer vision. The Vision Accelerator Programme will give expert guidance to reduce development time and costs for companies creating products using computer vision. ‘The demand for the Vision Accelerator Programme came from product leaders who are building vision features into their products and who came to industry events like the Embedded Vision Summit. They wanted more targeted help to make specific product decisions,’ said Jeff Bier, founder of the Embedded Vision Alliance. ‘By offering guidance on critical topics

like power/cost tradeoffs, 3D sensing, deep learning and edge versus cloud processing, we can help companies reduce risk and save considerable development time and expense,’ he added.

One of the first companies to make use of the Alliance’s expertise was Camio, which makes a smart monitoring surveillance product using deep learning for real-time video search and alerts. ‘The Embedded Vision Alliance and its relationships have provided more value than any other single source of advice or guidance,’ said Carter Maslan, CEO of the firm. The Vision Accelerator Programme includes confidential expert guidance, custom online educational workshops, access to exclusive Alliance events, connections with key potential suppliers and partners, and introductions to non-competitive companies facing similar challenges.

The Embedded Vision Alliance is an industry

partnership of technology providers and end-product companies, which are enabling applications using computer vision.

DesignCore camera mezzanine board OV5640 D3 Engineering has introduced its DesignCore camera mezzanine board OV5640 for the Arrow DragonBoard 410c. The mezzanine board with MIPI CSI-2 interface gives developers direct access to camera data on the Qualcomm Snapdragon 410E vision processor, eliminating the need to decode USB or Ethernet protocols. This enables more realistic evaluation of lower power, higher performance embedded vision designs and proof- of-concept prototypes.

The D3 DesignCore camera mezzanine board

OV5640 includes an autofocus micro camera module with 5 megapixel OminiVision OV5640 image sensor. Expansion headers provide access to additional Snapdragon processor interfaces including I2C, SPI, and UART.

A complete DragonBoard 410c camera kit includes a DragonBoard 410c development board, a DesignCore camera mezzanine board OV5640 with camera module, a power supply, and Linux demonstration software.


ReVision stack Xilinx has launched its ReVision stack, a development platform designed to enable a much broader set of software and systems engineers – with little or no hardware design expertise – to develop intelligent vision-guided systems. The announcement complements the company’s recent Reconfigurable Acceleration Stack, significantly broadening the deployment of machine learning applications with Xilinx technology. ReVision offers up to six times better images per second per watt in machine learning inference, 40 times better frames per second per watt of computer vision processing, and 1/5th the latency over competing embedded GPUs and typical SoCs, according to Xilinx. Developers with limited hardware expertise can use a C/C++/OpenCL development flow with industry-standard frameworks and libraries like Caffe and OpenCV to develop embedded vision

Common Vision Blox 2016 Service Pack 1 The latest version of Stemmer Imaging’s Common Vision Blox (CVB) machine vision software library offers improved support of ARM embedded platforms. To help users create their first embedded applications in CVB, detailed documentation is supplied along with the ARM setups for CVB 2016 Service Pack 1. Service Pack 1 features upgraded Linux support with processing tools, as well as an improved Teach Bench application, which provides interactive training of CVB’s Minos and Polimago search tools. The Teach Bench application also allows an import of existing CVB Manto files so that these can easily be upgraded to take advantage of CVB Polimago’s improved performance and capabilities.

applications on a single Zynq SoC or MPSoC. The product includes a broad range of

development resources for platform, algorithm and application development. This includes support for the most popular neural networks including AlexNet, GoogLeNet, SqueezeNet, SSD, and FCN. Additionally, the stack provides library elements including pre-defined and optimised implementations for CNN network layers, required to build custom neural networks. The machine learning elements are complemented by a broad set of acceleration-ready OpenCV functions for computer vision processing. For application level development, Xilinx

supports industry-standard frameworks including Caffe for machine learning and OpenVX for computer vision. The ReVision stack also includes development platforms from Xilinx and third parties, including various types of sensors.

The modular platform uses a mix of Lattice’s

FPGA, ASSP and programmable ASSP (pASSP) products to give the right balance of flexibility and energy efficiency required for a variety of embedded vision applications in the industrial, automotive and consumer markets.

Embedded vision development kit Lattice Semiconductor, a provider of customisable smart connectivity solutions, has released its Embedded Vision Development Kit, a development kit optimised for mobile-influenced system designs that require flexible, low cost, and low power image processing architectures.

42 Imaging and Machine Vision Europe • June/July 2017

The CrossLink pASSP input board includes dual- camera HD image sensors supporting the MIPI CSI-2 interface, eliminating the need for external video sources. The ECP5 base board enables low- power pre/post processing and includes support for HD image signal processing from Helion Vision. Also included is a NanoVesta connector to support external image sensor video inputs. The HDMI output board based on the Sil1136 non-HDCP version enables connectivity to standard HDMI displays.


Page 1  |  Page 2  |  Page 3  |  Page 4  |  Page 5  |  Page 6  |  Page 7  |  Page 8  |  Page 9  |  Page 10  |  Page 11  |  Page 12  |  Page 13  |  Page 14  |  Page 15  |  Page 16  |  Page 17  |  Page 18  |  Page 19  |  Page 20  |  Page 21  |  Page 22  |  Page 23  |  Page 24  |  Page 25  |  Page 26  |  Page 27  |  Page 28  |  Page 29  |  Page 30  |  Page 31  |  Page 32  |  Page 33  |  Page 34  |  Page 35  |  Page 36  |  Page 37  |  Page 38  |  Page 39  |  Page 40  |  Page 41  |  Page 42  |  Page 43  |  Page 44  |  Page 45  |  Page 46  |  Page 47  |  Page 48  |  Page 49  |  Page 50  |  Page 51  |  Page 52  |  Page 53  |  Page 54  |  Page 55  |  Page 56