Frequency domain subpixel registration using HOG phase correlation

We present a novel frequency-domain image registration technique, which employs histograms of oriented gradients providing subpixel estimates. Our method involves image filtering using dense Histogram of Oriented Gradients (HOG), which provides an advanced representation of the images coping with re...

Full description

Bibliographic Details
Main Authors: Argyriou, Vasileios, Tzimiropoulos, Georgios
Format: Article
Published: Elsevier 2016
Subjects:
Online Access:https://eprints.nottingham.ac.uk/39855/
_version_ 1848795929440681984
author Argyriou, Vasileios
Tzimiropoulos, Georgios
author_facet Argyriou, Vasileios
Tzimiropoulos, Georgios
author_sort Argyriou, Vasileios
building Nottingham Research Data Repository
collection Online Access
description We present a novel frequency-domain image registration technique, which employs histograms of oriented gradients providing subpixel estimates. Our method involves image filtering using dense Histogram of Oriented Gradients (HOG), which provides an advanced representation of the images coping with real-world registration problems such as non-overlapping regions and small deformations. The proposed representation retains the orientation information and the corresponding weights in a multi-dimensional representation. Furthermore, due to the overlapping local contrast normalization characteristic of HOG, the proposed Histogram of Oriented Gradients - Phase Correlation (HOG-PC) method improves significantly the estimated motion parameters in small size blocks. Experiments using sequences with and without ground truth including both global and local/multiple motions demonstrate that the proposed method out- performs the state-of-the-art in frequency-domain motion estimation, in the shape of phase correlation, in terms of subpixel accuracy and motion compensation prediction for a range of test material, block sizes and motion scenarios.
first_indexed 2025-11-14T19:39:54Z
format Article
id nottingham-39855
institution University of Nottingham Malaysia Campus
institution_category Local University
last_indexed 2025-11-14T19:39:54Z
publishDate 2016
publisher Elsevier
recordtype eprints
repository_type Digital Repository
spelling nottingham-398552020-05-04T18:22:31Z https://eprints.nottingham.ac.uk/39855/ Frequency domain subpixel registration using HOG phase correlation Argyriou, Vasileios Tzimiropoulos, Georgios We present a novel frequency-domain image registration technique, which employs histograms of oriented gradients providing subpixel estimates. Our method involves image filtering using dense Histogram of Oriented Gradients (HOG), which provides an advanced representation of the images coping with real-world registration problems such as non-overlapping regions and small deformations. The proposed representation retains the orientation information and the corresponding weights in a multi-dimensional representation. Furthermore, due to the overlapping local contrast normalization characteristic of HOG, the proposed Histogram of Oriented Gradients - Phase Correlation (HOG-PC) method improves significantly the estimated motion parameters in small size blocks. Experiments using sequences with and without ground truth including both global and local/multiple motions demonstrate that the proposed method out- performs the state-of-the-art in frequency-domain motion estimation, in the shape of phase correlation, in terms of subpixel accuracy and motion compensation prediction for a range of test material, block sizes and motion scenarios. Elsevier 2016-11-03 Article PeerReviewed Argyriou, Vasileios and Tzimiropoulos, Georgios (2016) Frequency domain subpixel registration using HOG phase correlation. Computer Vision and Image Understanding . ISSN 1077-3142 Phase correlation; Registration in frequency domain; Subpixel; Fourier; Histogram of oriented gradients http://www.sciencedirect.com/science/article/pii/S1077314216301722 doi:10.1016/j.cviu.2016.10.019 doi:10.1016/j.cviu.2016.10.019
spellingShingle Phase correlation; Registration in frequency domain; Subpixel; Fourier; Histogram of oriented gradients
Argyriou, Vasileios
Tzimiropoulos, Georgios
Frequency domain subpixel registration using HOG phase correlation
title Frequency domain subpixel registration using HOG phase correlation
title_full Frequency domain subpixel registration using HOG phase correlation
title_fullStr Frequency domain subpixel registration using HOG phase correlation
title_full_unstemmed Frequency domain subpixel registration using HOG phase correlation
title_short Frequency domain subpixel registration using HOG phase correlation
title_sort frequency domain subpixel registration using hog phase correlation
topic Phase correlation; Registration in frequency domain; Subpixel; Fourier; Histogram of oriented gradients
url https://eprints.nottingham.ac.uk/39855/
https://eprints.nottingham.ac.uk/39855/
https://eprints.nottingham.ac.uk/39855/