Learning image descriptors for matching based on Haar features
This paper presents a new and fast binary descriptor for image matching learned from Haar features. The training uses AdaBoost; the weak learner is built on response function for Haar features, instead of histogram-type features. The weak classifier is selected from a large weak feature pool. The se...
Main Authors: | , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Copernicus Publications
2014-08-01
|
Series: | The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences |
Online Access: | http://www.int-arch-photogramm-remote-sens-spatial-inf-sci.net/XL-3/61/2014/isprsarchives-XL-3-61-2014.pdf |
_version_ | 1811198342112215040 |
---|---|
author | L. Chen F. Rottensteiner C. Heipke |
author_facet | L. Chen F. Rottensteiner C. Heipke |
author_sort | L. Chen |
collection | DOAJ |
description | This paper presents a new and fast binary descriptor for image matching learned from Haar features. The training uses AdaBoost; the
weak learner is built on response function for Haar features, instead of histogram-type features. The weak classifier is selected from a
large weak feature pool. The selected features have different feature type, scale and position within the patch, having correspond
threshold value for weak classifiers. Besides, to cope with the fact in real matching that dissimilar matches are encountered much
more often than similar matches, cascaded classifiers are trained to motivate training algorithms see a large number of dissimilar
patch pairs. The final trained output are binary value vectors, namely descriptors, with corresponding weight and perceptron
threshold for a strong classifier in every stage. We present preliminary results which serve as a proof-of-concept of the work. |
first_indexed | 2024-04-12T01:28:44Z |
format | Article |
id | doaj.art-914ded05a27b4f328e671e432133b76c |
institution | Directory Open Access Journal |
issn | 1682-1750 2194-9034 |
language | English |
last_indexed | 2024-04-12T01:28:44Z |
publishDate | 2014-08-01 |
publisher | Copernicus Publications |
record_format | Article |
series | The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences |
spelling | doaj.art-914ded05a27b4f328e671e432133b76c2022-12-22T03:53:33ZengCopernicus PublicationsThe International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences1682-17502194-90342014-08-01XL-3616610.5194/isprsarchives-XL-3-61-2014Learning image descriptors for matching based on Haar featuresL. Chen0F. Rottensteiner1C. Heipke2Institute of Photogrammetry and GeoInformation, Leibniz Universität Hannover, Hanover, GermanyInstitute of Photogrammetry and GeoInformation, Leibniz Universität Hannover, Hanover, GermanyInstitute of Photogrammetry and GeoInformation, Leibniz Universität Hannover, Hanover, GermanyThis paper presents a new and fast binary descriptor for image matching learned from Haar features. The training uses AdaBoost; the weak learner is built on response function for Haar features, instead of histogram-type features. The weak classifier is selected from a large weak feature pool. The selected features have different feature type, scale and position within the patch, having correspond threshold value for weak classifiers. Besides, to cope with the fact in real matching that dissimilar matches are encountered much more often than similar matches, cascaded classifiers are trained to motivate training algorithms see a large number of dissimilar patch pairs. The final trained output are binary value vectors, namely descriptors, with corresponding weight and perceptron threshold for a strong classifier in every stage. We present preliminary results which serve as a proof-of-concept of the work.http://www.int-arch-photogramm-remote-sens-spatial-inf-sci.net/XL-3/61/2014/isprsarchives-XL-3-61-2014.pdf |
spellingShingle | L. Chen F. Rottensteiner C. Heipke Learning image descriptors for matching based on Haar features The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences |
title | Learning image descriptors for matching based on Haar features |
title_full | Learning image descriptors for matching based on Haar features |
title_fullStr | Learning image descriptors for matching based on Haar features |
title_full_unstemmed | Learning image descriptors for matching based on Haar features |
title_short | Learning image descriptors for matching based on Haar features |
title_sort | learning image descriptors for matching based on haar features |
url | http://www.int-arch-photogramm-remote-sens-spatial-inf-sci.net/XL-3/61/2014/isprsarchives-XL-3-61-2014.pdf |
work_keys_str_mv | AT lchen learningimagedescriptorsformatchingbasedonhaarfeatures AT frottensteiner learningimagedescriptorsformatchingbasedonhaarfeatures AT cheipke learningimagedescriptorsformatchingbasedonhaarfeatures |