Xilinx is assisting machine vision application designers to speed up the design using its Zynq All Programmable SoC by leveraging HALCON and VisualApplets development platforms. HALCON software from MVTec support for video analytics and runs on multicore chips including Xilinx's Zynq. HALCON's library supports blob analysis, morphology, matching, measuring, identification, and 3D vision. VisualApplets from Silicon Software provides advanced image processing libraries and high-level design entry and simulation tools that enable users to quickly and efficiently program FPGA hardware for industrial image processing.
"In the past, machine vision worked via central processing on a PC with cameras transmitting video and images; however, with the development of embedded technology, smart vision cameras and compact vision systems have developed to support the distributed intelligence at nodes," said Christoph Fritsch, director of industrial, science and medical segment markets for Xilinx. "The growth of smart vision applications requires All Programmable embedded processing technology that can deliver real-time processing and system flexibility for customization."
"The ARM Cortex-A9 dual-core processor inherent in the Zynq-7000 All Programmable SoC is the best match for advanced image processing applications," said Klaus-Henning Noffz, CEO of Silicon Software. "VisualApplets is the development platform to make maximum use of this processing power by offering short design cycles, simple verification, automatically generated interface to the ARM processor, and a huge image processing library. Software and application engineers are able to create complex designs efficiently – with the benefit of short time-to-market and dramatically reduced risks."
"Adding support for the Zynq-7000 All Programmable SoC is a natural extension of the flexible architecture inherent to HALCON software for machine vision," said Dr. Gerhard Blahusch, director of solutions and services for MVTec. "While the FPGA fabric provides the real-time performance required for pixel processing at high resolution and high frame rates, HALCON's automatic operator parallelization capability takes full advantage of the ARM Cortex-A9 dual-core processor, without the need for additional programming."