Improvement on SGM-based stereo matching algorithm and FPGA implementation

Mei Jin; Wenzhe Wu; Liguo Zhang

doi:10.1117/12.3033805

13 June 2024 Improvement on SGM-based stereo matching algorithm and FPGA implementation

Mei Jin, Wenzhe Wu, Liguo Zhang

Proceedings Volume 13180, International Conference on Image, Signal Processing, and Pattern Recognition (ISPP 2024); 131804A (2024) https://doi.org/10.1117/12.3033805
Event: International Conference on Image, Signal Processing, and Pattern Recognition (ISPP 2024), 2024, Guangzhou, China

Abstract

This paper presents a low-resource binocular stereo matching algorithm. This algorithm begins with an optimized design of semi-global stereo matching (SGM) by employing a sparser version of the census transform and introducing filtering operations in the sparse transformation factors. This optimization enhances the reliability of the sparse template and incorporates color information for fusion. Furthermore, it effectively preserves aggregation direction while curtailing other directions to alleviate the hardware cache resource pressure. The complete hardware system is implemented on an Artix7 AC7A035 platform. The overall matching accuracy is approximately 2.7% higher than that of the classic SGM algorithm. Under the condition that the input image is 1024*768 resolution, it consumes 42,456 lookup tables (LUTs), 40,175 flipflops (FF), and 4.5MB block RAM (BRAM). Resource utilization for both lookup tables and block RAM is reduced. Experimental validation confirms the effectiveness of the proposed algorithm in this paper.

(2024) Published by SPIE. Downloading of the abstract is permitted for personal use only.

Citation Download Citation

Mei Jin, Wenzhe Wu, and Liguo Zhang "Improvement on SGM-based stereo matching algorithm and FPGA implementation", Proc. SPIE 13180, International Conference on Image, Signal Processing, and Pattern Recognition (ISPP 2024), 131804A (13 June 2024); https://doi.org/10.1117/12.3033805

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available