Detection of SIFT keypoints in spherical omnidirectional view sensor

N. S. Chong, Y. H. Kho, M. L D Wong

Research output: Chapter in Book/Report/Conference proceedingConference contribution

2 Citations (Scopus)

Abstract

This paper proposed a method to detect object/scene through Scale Invariant Feature Transform (SIFT) keypoints for a hybrid sensor system comprises of a perspective view sensor and a spherical omnidirectional view sensor. A reference image is obtained from the perspective view sensor and matching is attempted with another distorted image acquired from the omnidirectional camera. The omnidirectional view image is first subjected to distortion correction (commonly termed "unwrapping") using closed form mapping functions and then SIFT keypoints of the corrected image are extracted and matched against the reference image's features. Experiment results show that the distortion correction produces acceptable performance of SIFT keypoint matching without modification to the classic SIFT algorithm.

Original languageEnglish
Title of host publicationProcedia Engineering
PublisherElsevier
Pages90-96
Number of pages7
Volume41
DOIs
Publication statusPublished - 2012
Externally publishedYes
Event2nd International Symposium on Robotics and Intelligent Sensors 2012, IRIS 2012 - Kuching, Sarawak, Malaysia
Duration: Sep 4 2012Sep 6 2012

Other

Other2nd International Symposium on Robotics and Intelligent Sensors 2012, IRIS 2012
CountryMalaysia
CityKuching, Sarawak
Period9/4/129/6/12

Fingerprint

Mathematical transformations
Sensors
Hybrid sensors
Cameras
Experiments

Keywords

  • Non-single viewpoint
  • Scale invariant feature transform
  • Scene recognition
  • Spherical omnidirectional view sensor

ASJC Scopus subject areas

  • Engineering(all)

Cite this

Chong, N. S., Kho, Y. H., & Wong, M. L. D. (2012). Detection of SIFT keypoints in spherical omnidirectional view sensor. In Procedia Engineering (Vol. 41, pp. 90-96). Elsevier. https://doi.org/10.1016/j.proeng.2012.07.147

Detection of SIFT keypoints in spherical omnidirectional view sensor. / Chong, N. S.; Kho, Y. H.; Wong, M. L D.

Procedia Engineering. Vol. 41 Elsevier, 2012. p. 90-96.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Chong, NS, Kho, YH & Wong, MLD 2012, Detection of SIFT keypoints in spherical omnidirectional view sensor. in Procedia Engineering. vol. 41, Elsevier, pp. 90-96, 2nd International Symposium on Robotics and Intelligent Sensors 2012, IRIS 2012, Kuching, Sarawak, Malaysia, 9/4/12. https://doi.org/10.1016/j.proeng.2012.07.147
Chong NS, Kho YH, Wong MLD. Detection of SIFT keypoints in spherical omnidirectional view sensor. In Procedia Engineering. Vol. 41. Elsevier. 2012. p. 90-96 https://doi.org/10.1016/j.proeng.2012.07.147
Chong, N. S. ; Kho, Y. H. ; Wong, M. L D. / Detection of SIFT keypoints in spherical omnidirectional view sensor. Procedia Engineering. Vol. 41 Elsevier, 2012. pp. 90-96
@inproceedings{6871a0ab2e7b452bb4afd28837c4f1a6,
title = "Detection of SIFT keypoints in spherical omnidirectional view sensor",
abstract = "This paper proposed a method to detect object/scene through Scale Invariant Feature Transform (SIFT) keypoints for a hybrid sensor system comprises of a perspective view sensor and a spherical omnidirectional view sensor. A reference image is obtained from the perspective view sensor and matching is attempted with another distorted image acquired from the omnidirectional camera. The omnidirectional view image is first subjected to distortion correction (commonly termed {"}unwrapping{"}) using closed form mapping functions and then SIFT keypoints of the corrected image are extracted and matched against the reference image's features. Experiment results show that the distortion correction produces acceptable performance of SIFT keypoint matching without modification to the classic SIFT algorithm.",
keywords = "Non-single viewpoint, Scale invariant feature transform, Scene recognition, Spherical omnidirectional view sensor",
author = "Chong, {N. S.} and Kho, {Y. H.} and Wong, {M. L D}",
year = "2012",
doi = "10.1016/j.proeng.2012.07.147",
language = "English",
volume = "41",
pages = "90--96",
booktitle = "Procedia Engineering",
publisher = "Elsevier",
address = "Netherlands",

}

TY - GEN

T1 - Detection of SIFT keypoints in spherical omnidirectional view sensor

AU - Chong, N. S.

AU - Kho, Y. H.

AU - Wong, M. L D

PY - 2012

Y1 - 2012

N2 - This paper proposed a method to detect object/scene through Scale Invariant Feature Transform (SIFT) keypoints for a hybrid sensor system comprises of a perspective view sensor and a spherical omnidirectional view sensor. A reference image is obtained from the perspective view sensor and matching is attempted with another distorted image acquired from the omnidirectional camera. The omnidirectional view image is first subjected to distortion correction (commonly termed "unwrapping") using closed form mapping functions and then SIFT keypoints of the corrected image are extracted and matched against the reference image's features. Experiment results show that the distortion correction produces acceptable performance of SIFT keypoint matching without modification to the classic SIFT algorithm.

AB - This paper proposed a method to detect object/scene through Scale Invariant Feature Transform (SIFT) keypoints for a hybrid sensor system comprises of a perspective view sensor and a spherical omnidirectional view sensor. A reference image is obtained from the perspective view sensor and matching is attempted with another distorted image acquired from the omnidirectional camera. The omnidirectional view image is first subjected to distortion correction (commonly termed "unwrapping") using closed form mapping functions and then SIFT keypoints of the corrected image are extracted and matched against the reference image's features. Experiment results show that the distortion correction produces acceptable performance of SIFT keypoint matching without modification to the classic SIFT algorithm.

KW - Non-single viewpoint

KW - Scale invariant feature transform

KW - Scene recognition

KW - Spherical omnidirectional view sensor

UR - http://www.scopus.com/inward/record.url?scp=84901048564&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84901048564&partnerID=8YFLogxK

U2 - 10.1016/j.proeng.2012.07.147

DO - 10.1016/j.proeng.2012.07.147

M3 - Conference contribution

VL - 41

SP - 90

EP - 96

BT - Procedia Engineering

PB - Elsevier

ER -