High-Speed Image Processing and Data Transmission Based on Vivado HLS and AXI4-Stream Interface

In this paper, the embedded image processing system based on ZYNQ-7000 SOC is designed to achieve feature extraction and high-speed data transmission. This system overcomes the shortcomings of high power consumption, long delay time and limited bandwidth in traditional embedded image processing systems. To maximize the advantages of ZYNQ Processor System (PS) and Programmable Logic (PL), the computationally intensive feature extraction algorithm was implemented not only in PL with its speciality of parallel computing to enhance the processing speed but also in PS to increase the flexibility of the system. Moreover, High-Level Synthesis (HLS) tool in Vivado was used to generate the Intellectual Property (IP) core in PL for accelerating the processing of feature extraction and coordinates calculation. Furthermore, in the data transmission interface, the fast Advanced eXtensible Interface (AXI) 4-Stream bus was selected for image Stream. Meanwhile, the coordinate information of the feature points was stored as the pixels of image via AXI4-Stream. This method simplified the circuit and increased the transmission rate. The comparative experimental results reveal the proposed image processing and transmission scheme has a higher speed with 15.6 times faster than implementing in ARM.

[1]  Yan Han,et al.  Hardware/Software Co-Design of a Traffic Sign Recognition System Using Zynq FPGAs , 2015 .

[2]  Haitham Akkary,et al.  On the effectiveness of accelerating MapReduce functions using the Xilinx Vivado HLS tool , 2016, Int. J. High Perform. Syst. Archit..

[3]  Luis Angel Barragan,et al.  High-Level Synthesis for Accelerating the FPGA Implementation of Computationally Demanding Control Algorithms for Power Converters , 2013, IEEE Transactions on Industrial Informatics.

[4]  Ulrich Rückert,et al.  FPGA-based Generic Architecture for Rapid Prototyping of Video Hardware Accelerators using NoC AXI4-Stream Interconnect and GigE Vision Camera Interfaces , 2014 .

[5]  Holger Blume,et al.  HLS-based FPGA implementation of a predictive block-based motion estimation algorithm — A field report , 2014, Proceedings of the 2014 Conference on Design and Architectures for Signal and Image Processing.

[6]  Sangil Lee,et al.  Distributed and parallel real-time control system equipped FPGA-Zynq and EPICS middleware , 2016, 2016 IEEE-NPSS Real Time Conference (RT).

[7]  Mohammad Hosseinabady,et al.  Energy Optimization in Commercial FPGAs with Voltage, Frequency and Logic Scaling , 2016, IEEE Transactions on Computers.

[8]  Du-ming Tsai,et al.  Machine Vision-Based Positioning and Inspection Using Expectation–Maximization Technique , 2017, IEEE Transactions on Instrumentation and Measurement.

[9]  Quang-Cherng Hsu,et al.  Development of a simple three-dimensional machine-vision measurement system for in-process mechanical parts , 2017 .

[10]  José M. P. Nascimento,et al.  System-on-chip field-programmable gate array design for onboard real-time hyperspectral unmixing , 2016 .

[11]  Yinan Kong,et al.  Efficient Hardware Implementation For Fingerprint Image Enhancement Using Anisotropic Gaussian Filter , 2017, IEEE Transactions on Image Processing.

[12]  A. Cortes,et al.  High level synthesis using Vivado HLS for Zynq SoC: Image processing case studies , 2016, 2016 Conference on Design of Circuits and Integrated Systems (DCIS).

[13]  Christopher Blochwitz,et al.  Hardware-accelerated pose estimation for embedded systems using Vivado HLS , 2016, 2016 International Conference on ReConFigurable Computing and FPGAs (ReConFig).