Document Server@UHasselt >
Research >
Research publications >

Please use this identifier to cite or link to this item: http://hdl.handle.net/1942/25924

Title: A Novel Hardware-Oriented Stereo Matching Algorithm and Its Architecture Design in FPGA
Authors: Li, Yanzhe
Huang, Kai
Claesen, Luc
Issue Date: 2017
Publisher: Springer-Verlag
Citation: Hollstein, Thomas; Raik, Jaan; Kostin, Sergei; Tšertov, Anton; O'Connor, Ian; Reis, Ricardo (Ed.). VLSI-SoC: System-on-Chip in the Nanoscale Era – Design, Verification and Reliability, Springer Nature, p. 213-232
Series/Report: IFIP Advances in Information and Communication Technology
Series/Report no.: 508
Abstract: Stereo matching is a crucial step to extract depth information from stereo images. However, it is still challenging to achieve good performance in both speed and accuracy for various stereo vision applications. In this contribution, a hardware-compatible stereo matching algorithm is proposed and its associated hardware implementation is also presented. The proposed algorithm can produce high-quality disparity maps with the combined use of the mini-census transform, segmentation-based adaptive support weight and effective refinement. Moreover, the proposed architecture is optimized as a fully pipelined and scalable hardware system. Implemented on an Altera Stratix-IV FPGA board, it can achieve 65 frames per second (fps) for 1024 × 768 stereo images and a 64 pixel disparity range. The proposed system is evaluated on the Middlebury benchmark and the average error rate is 6.56%. The experimental results indicate that the accuracy is competitive with some state-of-the-art software implementations.
Notes: Li, YZ (reprint author), Zhejiang Univ, Inst VLSI Design, Hangzhou, Zhejiang, Peoples R China. liyz@vlsi.zju.edu.cn; huangk@vlsi.zju.edu.cn; luc.claesen@uhasselt.be
URI: http://hdl.handle.net/1942/25924
DOI: 10.1007/978-3-319-67104-8_11
ISI #: 000432573400011
ISBN: 9783319671031
ISSN: 1868-4238
Category: C1
Type: Proceedings Paper
Appears in Collections: Research publications

Files in This Item:

Description SizeFormat
Published version2.4 MBAdobe PDF
Proof of peer review271.5 kBAdobe PDF

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.