An Efficient Hardware Implementation of FPGA-Based Stereo Matching Using a Single Path SGM Direction

This paper focuses on the implementation of FPGA-based stereo matching with minimal hardware utilization, particularly for medium- to low-end FPGAs with limited hardware resources. Due to its ability to rapidly reconstruct the distance of an object from two stereo images, stereo matching has been widely employed in robot navigation, advanced driver assistance systems (ADAS) in automotive, and agricultural. To achieve minimal hardware requirements for stereo matching implementation, we offer stereo matching with a tiny window buffer for cost matching computation utilizing sum of absolute difference (SAD). Furthermore, a single path semi-global matching (SGM) direction is used to boost the smoothness and discontinuity of the disparity image. According to the experimental results obtained with the Xilinx Zynq-XC7Z020 using a 4x4-pixel window size and single path SGM, the amount of required FPGA resources are 9% FF, 18% LUT, 5% BRAM, and 136 MHz working frequency. For the execution time, it takes 1.68 seconds.

Authors

Publication Details

Published
2024-10-09
DOI
https://doi.org/10.1109/ic3ina64086.2024.10732374
Primary Topic
Blind Source Separation Techniques
Type
article
Field-Weighted Citation Impact
0.00

Funders

Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

An Efficient Hardware Implementation of FPGA-Based Stereo Matching Using a Single Path SGM Direction

Ratna Indrawijaya, Arya Adhyaksa Waskita, Arief Suryadi Satyawan, Iman Firmansyah et al.
Blind Source Separation Techniques
article

An Efficient Hardware Implementation of FPGA-Based Stereo Matching Using a Single Path SGM Direction

Ratna Indrawijaya, Arya Adhyaksa Waskita, Arief Suryadi Satyawan, Iman Firmansyah, Bambang Setiadi, Dayat Kurniawan
article en

Abstract

This paper focuses on the implementation of FPGA-based stereo matching with minimal hardware utilization, particularly for medium- to low-end FPGAs with limited hardware resources. Due to its ability to rapidly reconstruct the distance of an object from two stereo images, stereo matching has been widely employed in robot navigation, advanced driver assistance systems (ADAS) in automotive, and agricultural. To achieve minimal hardware requirements for stereo matching implementation, we offer stereo matching with a tiny window buffer for cost matching computation utilizing sum of absolute difference (SAD). Furthermore, a single path semi-global matching (SGM) direction is used to boost the smoothness and discontinuity of the disparity image. According to the experimental results obtained with the Xilinx Zynq-XC7Z020 using a 4x4-pixel window size and single path SGM, the amount of required FPGA resources are 9% FF, 18% LUT, 5% BRAM, and 136 MHz working frequency. For the execution time, it takes 1.68 seconds.

University of Tsukuba
Openalex Percentile: Top 79%
Blind Source Separation Techniques
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.