Xilinx tools, libraries, and methods integrate machine learning, computer vision, sensor fusion, and connectivity in a vision-oriented intelligence system
Beijing, March 16, 2017 — Xilinx, Inc. (NASDAQ: XLNX), a global leader in All Programmable technologies and devices, demonstrated at the upcoming 2017 Embedded World conference (2017 Embedded World) The most responsive and reconfigurable vision-oriented intelligence system, dedicated to the conference, six keynote speeches, 11 live demonstrations. Through these presentations and live demonstrations, Xilinx demonstrated its reVISION stack of tools, libraries, and methodologies to audiences from around the world, and how it integrates machine learning, computer vision, and sensor fusion in a vision-oriented intelligence system. And connectivity to create the most responsive and reconfigurable vision-oriented intelligence system.
At the same time, more than 25 Xilinx Alliance program members participated in the embedded conference and demonstrated Xilinx-based technology. These members include IP providers, EDA vendors, embedded software providers, system integrators, and hardware vendors. A demonstration by the Alliance members further demonstrates Xilinx's mature ecosystem and strong influence in embedded design.
The six major presentations of Xilinx include:
I. OpenAMP in the Industrial IoT Framework
Second, heterogeneous multi-core technology supports security: why do you need multiple different cores on SoC?
Third, the real-time physical system self-optimization virtualization method
FPGA-SoC—Manages the performance and power consumption of heterogeneous on-chip processing systems
V. Designing Industry 4.0 Silicon Carbide High Power Inverter with Zynq-7000 and Zynq UltraScale+ MPSoC Supporting Machine Learning Diagnostics
Embedded Vision - Combine OpenCV and high-level synthesis to reduce embedded vision development time
11 live demonstrations include:
◠Driver Monitoring System - demonstrated by XylonThe real-time head and eyelid tracking demonstration combined with precise line-of-sight measurements reliably detects the driver's attention and state under a variety of complex physical conditions. Based on the Xilinx All Programmable Zynq® SoC, the Xylon Driver Monitoring System is a proven automotive quality platform that flexibly optimizes system partitioning to meet visual processing performance requirements at lower cost and power.
◠Flight camera system with ADAS framework - demonstrated by XylonThe advanced panoramic vision system in this demonstration provides a three-dimensional hemisphere view around the vehicle that allows the driver to dynamically adjust the viewing angle. The system uses Zynq® UltraScaleTM+ MPSoC (includes quad-core ARM® CortexTM-A53 and dual-core ARM Cortex-R5 with MailTM-400 GPU) to support automotive model rendering. Customers can take advantage of FPGA fabric parallelism and optimize their unique features, which is critical for real-time processing.
â— Hardware Acceleration for 4K60 Dense Optical Flow Algorithm - Demonstrated by XilinxThe demonstration demonstrates the power of the Zynq Ultrascale+ MPSoC by running advanced dense optical flow algorithms in programmable logic at 4K resolution and frame rate of 60 frames per second.
â— Deep learning, machine vision and sensor integration - demonstrated by XilinxThe demonstration combines three major and complex algorithms of today's vision-oriented systems, including object-oriented detection or scene segmentation convolutional neural network (CNN) algorithms, dense optical flow algorithms for motion tracking, and depth perceptrons. Stereo vision algorithm. The demo runs on a single Zynq Ultrascale+ MPSoC device.
â— UHD HEVC real-time compression and streaming playback - demonstrated by XilinxThe video codec hardware module is integrated into the Zynq UltraScale+ MPSoC EV device, allowing you to experience low latency real-time UHD video compression/decompression and transfer. The video codec supports up to 8K video and delivers low bandwidth data with very low latency for real-time video applications such as drone operation and video conferencing.
â— Embedded vision robot system that supports accelerated image recognition - demonstrated by MLEThis industrial pick and place demonstration uses a binary neural network algorithm based on programmable logic for machine vision recognition. Compared to software implementations, programmable logic implementations can increase image recognition speed by more than 1000 times while improving recognition accuracy and reliability to meet the requirements of high-performance mission-critical vision applications.
â— 3D object recognition industrial camera for machine vision - demonstrated by iVeiaThis demo highlights the heterogeneous architecture of the Zynq UltraScale+ MPSoC and supports traditional PC algorithms on embedded camera platforms to accelerate time-to-market for machine vision applications.
â— Dobby AI Drone - Demonstrated by Zero Technology and Deep TechnologyZero Technology's Dobby AI is a pocket-sized drone with Xilinx Zynq SoC devices that uses deep learning to detect human gestures. This demo will showcase the deep learning inference techniques provided by Shenjian Technology.
â— Functional Safety Reference System - demonstrated by IkerlanThis demonstration demonstrates a comprehensive design template for a custom solution with hardware fault tolerance (HFT=1) implemented on a single Xilinx Zynq SoC single chip that conforms to the IEC 61508:2010 standard. Advanced support for the PC-controlled error injecTIon funcTIon will be demonstrated using Xilinx's latest Zynq SoC Safety Reference System.
â— Intelligent Network Security Gateway - demonstrated by SoCeThis demonstration combines sensor fusion, network security, real-time communication and control, and machine learning to deliver end-to-cloud high availability on the Xilinx Zynq SoC platform. The mass production intelligent gateway is time-aware and provides secure network interoperability and control functions, highlighting the advantages of combining operational technology with information technology.
â— All Programmable Industrial Control System - Demonstrated by BE Services, Matrikon and RTIThis demo introduces a user-ready platform available from Industrial Control Systems (ICS) that not only enables network interoperability, but also extends secure edge analysis to the cloud and is provided by Zynq SoC and Zynq UltraScale+ MPSoC devices. Implemented All Programmable functionality.
Demonstration of Xilinx Alliance Program membersThis year, more than 25 Xilinx Alliance program members, including IP providers, EDA vendors, embedded software providers, system integrators and hardware vendors, will showcase Xilinx-based technologies, including:
Senior member of the affiliate program:
· Avnet Memec-Silica – Hall 1/1-370, Hall 1/1-205
· Topic Embedded Products – Hall 3/3-111
· Xylon – Hall 1/1-205
Alliance Program Certification Members:
· DAVE Embedded Systems – Hall 4/4-481
· Enclustra GmbH – Hall 4A/4A-121, Hall 1/1-101
· Missing Link Electronics – Hall 4/4-478, Hall 1/1-205
· SoC-e – Hall 1/1-205
Alliance Program Members:
· Aldec Inc. – Hall 4/4-560
· Digilent Inc. – Hall 1/1-208
· eSOL Company – Hall 4/4-634
· HMS Industrial Networks – Hall 1/1-550
· iVeia – Hall 1/1-205
· MathWorks – Hall 4/4-110
· Mentor Graphics (Deutschland) – Hall 4/4-510
· NI – Hall 1/1-652, Hall 4/4-108
· NetModule – Hall 3/3-320
· Port GmbH – Hall 1/1-648
· Solectrix – Hall4A/4A-126
· X2E – Hall 4/4-370
Associate Members:
· 3S Smart Software SoluTIons (Codesys) – Hall 1/1-205
· Antmicro – Hall 4A/4A-121
· dSPACE – Hall 4/4-316
· Ikerlan – Hall 1/1-205
· Koenig-pa – Hall 4/4-380
· Lynx Software Technologies – Hall 4/4-423
· Matrikon – Hall 1/1-205
· RBZ Robot Design SL RBZ Embedded Logics – Hall 3/3-500
For more complete information on Xilinx's participation in this embedded conference, please visit Xilinx's official website to learn more:
Https://china.xilinx.com/about/events/2017_xilinx_embedded_world.html#participation
About XilinxXilinx, Inc. (NASDAQ: XLNX) is the world's leading provider of All Programmable FPGAs, SoCs, MPSoCs, RFSoCs and 3D ICs, uniquely enabling applications that are both software-defined and hardware-optimized. The development of cloud computing, 5G wireless, embedded vision and industrial Internet of Things. For more information, please visit the Xilinx Chinese website at http://china.xilinx.com/.
The advantages of the solution of motorcycle engine starter lithium battery, compared to the traditional lead-acid battery, motorcycle special starting lithium battery in addition to high energy density, there are the following advantages:
1. More environmental protection
Lead-acid batteries are polluted in the production process and may cause pollution if they are not recycled properly. Lithium batteries are green and environmental friendly.
2. Longer service life.
3. Low self-discharge rate, less than 10% per month, lead acid battery can reach 30%.
Starting Lithium Battery,Rechargeable Motorcycle Battery,Motorbike Starter Battery,Motorcycle Starting Battery Deep Cycle,E-motor Starter Battery,Motor Scooter Lithium-ion Battery
Shenzhen Enershare Technology Co.,Ltd , https://www.enersharepower.com