Implementation of a Semi-global Stereo Matching Algorithm Based on Pynq
ZHAO Xin
MENG Lingjun
LIU Weihong
Abstract：As the core algorithm of binocular stereo vision technology,the stereo matching algorithm has the characteristics of high complexity,high resource consumption and poor real-time performance.The existing FPGA-based implementation methods mostly use high-performance FPGA chips,its board rich resources are used to develop pure verilog language.Although the re-al-time effect is good,but there is a development difficulty,long cycle time,high cost of the disadvantages.In this paper,an SGM stereo matching algorithm is proposed based on the relatively inexpensive Pynq-z2 development board developed collaboratively on the PL+PS side,where the computationally intensive part is hardware accelerated on the FPGA side and part of the cost-optimised algorithm is implemented in the ARM side using the Python language.Experiments show that the algorithm can effectively reduce the resource occupancy rate and basically meet the requirements of relatively inexpensive embedded devices.
Keywords：binocular stereo visionresource occupancysemi-global stereo matching algorithmPynq-z2
Publication Date：2024-02-20
Online Publishing Date：2025-08-15(First online date of this platform, not the publication date of the document)
Pages：5( 40-44 )