Memory Saving Feature Descriptor Using Scale and Rotation Invariant Patches around the Feature Ppoints

Masamichi KITAGAWA; Ikuko SHIMIZU

doi:10.1587/transinf.2018EDL8176

Open Access
Memory Saving Feature Descriptor Using Scale and Rotation Invariant Patches around the Feature Ppoints

Masamichi KITAGAWA, Ikuko SHIMIZU

Full Text Views

68

Cite this

Free PDF (707KB)

Summary :

To expand the use of systems using a camera on portable devices such as tablets and smartphones, we have developed and propose a memory saving feature descriptor, the use of which is one of the essential techniques in computer vision. The proposed descriptor compares pixel values of pre-fixed positions in the small patch around the feature point and stores binary values. Like the conventional descriptors, it extracts the patch on the basis of the scale and orientation of the feature point. For memories of the same size, it achieves higher accuracy than ORB and BRISK in all cases and AKAZE for the images with textured regions.

Publication: IEICE TRANSACTIONS on Information Vol.E102-D No.5 pp.1106-1110

Publication Date: 2019/05/01

Publicized: 2019/02/05

Online ISSN: 1745-1361

DOI: 10.1587/transinf.2018EDL8176

Type of Manuscript: LETTER

Category: Image Recognition, Computer Vision

Authors

Masamichi KITAGAWA
Tokyo University of Agriculture and Technology
Ikuko SHIMIZU
Tokyo University of Agriculture and Technology

Keyword

image matching, feature descriptor, keypoints

Cite this

Copy

Masamichi KITAGAWA, Ikuko SHIMIZU, "Memory Saving Feature Descriptor Using Scale and Rotation Invariant Patches around the Feature Ppoints" in IEICE TRANSACTIONS on Information, vol. E102-D, no. 5, pp. 1106-1110, May 2019, doi: 10.1587/transinf.2018EDL8176.
Abstract: To expand the use of systems using a camera on portable devices such as tablets and smartphones, we have developed and propose a memory saving feature descriptor, the use of which is one of the essential techniques in computer vision. The proposed descriptor compares pixel values of pre-fixed positions in the small patch around the feature point and stores binary values. Like the conventional descriptors, it extracts the patch on the basis of the scale and orientation of the feature point. For memories of the same size, it achieves higher accuracy than ORB and BRISK in all cases and AKAZE for the images with textured regions.
URL: https://global.ieice.org/en_transactions/information/10.1587/transinf.2018EDL8176/_p

Copy

@ARTICLE{e102-d_5_1106,
author={Masamichi KITAGAWA, Ikuko SHIMIZU, },
journal={IEICE TRANSACTIONS on Information},
title={Memory Saving Feature Descriptor Using Scale and Rotation Invariant Patches around the Feature Ppoints},
year={2019},
volume={E102-D},
number={5},
pages={1106-1110},
abstract={To expand the use of systems using a camera on portable devices such as tablets and smartphones, we have developed and propose a memory saving feature descriptor, the use of which is one of the essential techniques in computer vision. The proposed descriptor compares pixel values of pre-fixed positions in the small patch around the feature point and stores binary values. Like the conventional descriptors, it extracts the patch on the basis of the scale and orientation of the feature point. For memories of the same size, it achieves higher accuracy than ORB and BRISK in all cases and AKAZE for the images with textured regions.},
keywords={},
doi={10.1587/transinf.2018EDL8176},
ISSN={1745-1361},
month={May},}

Copy

TY - JOUR
TI - Memory Saving Feature Descriptor Using Scale and Rotation Invariant Patches around the Feature Ppoints
T2 - IEICE TRANSACTIONS on Information
SP - 1106
EP - 1110
AU - Masamichi KITAGAWA
AU - Ikuko SHIMIZU
PY - 2019
DO - 10.1587/transinf.2018EDL8176
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E102-D
IS - 5
JA - IEICE TRANSACTIONS on Information
Y1 - May 2019
AB - To expand the use of systems using a camera on portable devices such as tablets and smartphones, we have developed and propose a memory saving feature descriptor, the use of which is one of the essential techniques in computer vision. The proposed descriptor compares pixel values of pre-fixed positions in the small patch around the feature point and stores binary values. Like the conventional descriptors, it extracts the patch on the basis of the scale and orientation of the feature point. For memories of the same size, it achieves higher accuracy than ORB and BRISK in all cases and AKAZE for the images with textured regions.
ER -