Skip to main content

Lund University Publications

LUND UNIVERSITY LIBRARIES

Towards Dense Point Correspondence with PatchMatch in Low-Resolution Sonar Images

Fuchs, Louise Rixon ; Gallstrom, Andreas LU and Maki, Atsuto (2022) 2022 IEEE/OES Autonomous Underwater Vehicles Symposium, AUV 2022 In 2022 IEEE/OES Autonomous Underwater Vehicles Symposium, AUV 2022
Abstract

Robust feature correspondences between 2D sonar imagery are important for perception tasks in the underwater domain such as 3D reconstruction but involve open challenges, in particular, low-resolution as well as the fact that object appearance is view-dependent. Although sonars in the MHz range would allow for higher resolution imagery, in this paper we focus on scenarios with a lower frequency kHz sensor, in which the longer visual range is gained at the sacrifice of image resolution. To this end, we first propose to solve the correspondence task using the PatchMatch algorithm for the first time in sonar imagery, and then propose a method for feature extraction based on IC. We then compare the proposed methods against conventional... (More)

Robust feature correspondences between 2D sonar imagery are important for perception tasks in the underwater domain such as 3D reconstruction but involve open challenges, in particular, low-resolution as well as the fact that object appearance is view-dependent. Although sonars in the MHz range would allow for higher resolution imagery, in this paper we focus on scenarios with a lower frequency kHz sensor, in which the longer visual range is gained at the sacrifice of image resolution. To this end, we first propose to solve the correspondence task using the PatchMatch algorithm for the first time in sonar imagery, and then propose a method for feature extraction based on IC. We then compare the proposed methods against conventional methods from computer vision. We evaluate our method on data from a lake experiment with objects captured with an FLS sensor. Our results show that the proposed combination of IC together with PatchMatch is well-suited for point feature extraction and correspondence in sonar imagery. Further, we also evaluate the different methods for point correspondence with a 3D object reconstruction task.

(Less)
Please use this url to cite or link to this publication:
author
; and
organization
publishing date
type
Chapter in Book/Report/Conference proceeding
publication status
published
subject
keywords
3D reconstruction, feature correspondence, FLS, PatchMatch
host publication
2022 IEEE/OES Autonomous Underwater Vehicles Symposium, AUV 2022
series title
2022 IEEE/OES Autonomous Underwater Vehicles Symposium, AUV 2022
publisher
IEEE - Institute of Electrical and Electronics Engineers Inc.
conference name
2022 IEEE/OES Autonomous Underwater Vehicles Symposium, AUV 2022
conference location
Singapore, Singapore
conference dates
2022-09-19 - 2022-09-21
external identifiers
  • scopus:85143972442
ISBN
9781665416894
DOI
10.1109/AUV53081.2022.9965885
language
English
LU publication?
yes
id
902ecdb6-68d2-4556-b8e8-847cd2ad5bff
date added to LUP
2023-01-12 11:31:28
date last changed
2023-01-12 11:31:28
@inproceedings{902ecdb6-68d2-4556-b8e8-847cd2ad5bff,
  abstract     = {{<p>Robust feature correspondences between 2D sonar imagery are important for perception tasks in the underwater domain such as 3D reconstruction but involve open challenges, in particular, low-resolution as well as the fact that object appearance is view-dependent. Although sonars in the MHz range would allow for higher resolution imagery, in this paper we focus on scenarios with a lower frequency kHz sensor, in which the longer visual range is gained at the sacrifice of image resolution. To this end, we first propose to solve the correspondence task using the PatchMatch algorithm for the first time in sonar imagery, and then propose a method for feature extraction based on IC. We then compare the proposed methods against conventional methods from computer vision. We evaluate our method on data from a lake experiment with objects captured with an FLS sensor. Our results show that the proposed combination of IC together with PatchMatch is well-suited for point feature extraction and correspondence in sonar imagery. Further, we also evaluate the different methods for point correspondence with a 3D object reconstruction task.</p>}},
  author       = {{Fuchs, Louise Rixon and Gallstrom, Andreas and Maki, Atsuto}},
  booktitle    = {{2022 IEEE/OES Autonomous Underwater Vehicles Symposium, AUV 2022}},
  isbn         = {{9781665416894}},
  keywords     = {{3D reconstruction; feature correspondence; FLS; PatchMatch}},
  language     = {{eng}},
  publisher    = {{IEEE - Institute of Electrical and Electronics Engineers Inc.}},
  series       = {{2022 IEEE/OES Autonomous Underwater Vehicles Symposium, AUV 2022}},
  title        = {{Towards Dense Point Correspondence with PatchMatch in Low-Resolution Sonar Images}},
  url          = {{http://dx.doi.org/10.1109/AUV53081.2022.9965885}},
  doi          = {{10.1109/AUV53081.2022.9965885}},
  year         = {{2022}},
}