Increasing the Precision of Junction Shaped Features

Research output: Chapter in book/report/conference proceedingConference contributionResearchpeer review

Authors

Research Organisations

View graph of relations

Details

Original languageEnglish
Title of host publicationProceedings of the 14th IAPR International Conference on Machine Vision Applications
Subtitle of host publication MVA 2015
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages295-298
Number of pages4
ISBN (electronic)9784901122153
Publication statusPublished - Jul 2015
Event14th IAPR International Conference on Machine Vision Applications, MVA 2015 - Tokyo, Japan
Duration: 18 May 201522 May 2015

Abstract

The scale invariant feature operator (SFOP) detects circular features from an image using a spiral shape model. Special cases of the spiral model are junctions and circular symmetric shapes. The spatial localization is determined with subpixel accuracy which is obtained by an interpolation of the structure tensor in the scale space. For the interpolation, SFOP uses a 3D quadratic function. This leads to suboptimal solutions since the structure tensor surrounding a feature does not show the shape of a 3D quadratic. The aim of this paper is to improve the localization of the features detected by SFOP. A Difference of Gaussians function is proposed for the signal approximation which leads to improved precision values and to more accurate features. The proposed method improves the localization such that 72.5% of the features increase their precision. Hence, more features are extracted while increasing their repeatability by up to 9% on standard benchmarks.

ASJC Scopus subject areas

Cite this

Increasing the Precision of Junction Shaped Features. / Cordes, Kai; Ostermann, Jorn.
Proceedings of the 14th IAPR International Conference on Machine Vision Applications: MVA 2015. Institute of Electrical and Electronics Engineers Inc., 2015. p. 295-298 7153189.

Research output: Chapter in book/report/conference proceedingConference contributionResearchpeer review

Cordes, K & Ostermann, J 2015, Increasing the Precision of Junction Shaped Features. in Proceedings of the 14th IAPR International Conference on Machine Vision Applications: MVA 2015., 7153189, Institute of Electrical and Electronics Engineers Inc., pp. 295-298, 14th IAPR International Conference on Machine Vision Applications, MVA 2015, Tokyo, Japan, 18 May 2015. https://doi.org/10.1109/mva.2015.7153189
Cordes, K., & Ostermann, J. (2015). Increasing the Precision of Junction Shaped Features. In Proceedings of the 14th IAPR International Conference on Machine Vision Applications: MVA 2015 (pp. 295-298). Article 7153189 Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/mva.2015.7153189
Cordes K, Ostermann J. Increasing the Precision of Junction Shaped Features. In Proceedings of the 14th IAPR International Conference on Machine Vision Applications: MVA 2015. Institute of Electrical and Electronics Engineers Inc. 2015. p. 295-298. 7153189 doi: 10.1109/mva.2015.7153189
Cordes, Kai ; Ostermann, Jorn. / Increasing the Precision of Junction Shaped Features. Proceedings of the 14th IAPR International Conference on Machine Vision Applications: MVA 2015. Institute of Electrical and Electronics Engineers Inc., 2015. pp. 295-298
Download
@inproceedings{39c3a92d660f4ffbafc0ee96de7c40b1,
title = "Increasing the Precision of Junction Shaped Features",
abstract = "The scale invariant feature operator (SFOP) detects circular features from an image using a spiral shape model. Special cases of the spiral model are junctions and circular symmetric shapes. The spatial localization is determined with subpixel accuracy which is obtained by an interpolation of the structure tensor in the scale space. For the interpolation, SFOP uses a 3D quadratic function. This leads to suboptimal solutions since the structure tensor surrounding a feature does not show the shape of a 3D quadratic. The aim of this paper is to improve the localization of the features detected by SFOP. A Difference of Gaussians function is proposed for the signal approximation which leads to improved precision values and to more accurate features. The proposed method improves the localization such that 72.5% of the features increase their precision. Hence, more features are extracted while increasing their repeatability by up to 9% on standard benchmarks.",
author = "Kai Cordes and Jorn Ostermann",
year = "2015",
month = jul,
doi = "10.1109/mva.2015.7153189",
language = "English",
pages = "295--298",
booktitle = "Proceedings of the 14th IAPR International Conference on Machine Vision Applications",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
address = "United States",
note = "14th IAPR International Conference on Machine Vision Applications, MVA 2015 ; Conference date: 18-05-2015 Through 22-05-2015",

}

Download

TY - GEN

T1 - Increasing the Precision of Junction Shaped Features

AU - Cordes, Kai

AU - Ostermann, Jorn

PY - 2015/7

Y1 - 2015/7

N2 - The scale invariant feature operator (SFOP) detects circular features from an image using a spiral shape model. Special cases of the spiral model are junctions and circular symmetric shapes. The spatial localization is determined with subpixel accuracy which is obtained by an interpolation of the structure tensor in the scale space. For the interpolation, SFOP uses a 3D quadratic function. This leads to suboptimal solutions since the structure tensor surrounding a feature does not show the shape of a 3D quadratic. The aim of this paper is to improve the localization of the features detected by SFOP. A Difference of Gaussians function is proposed for the signal approximation which leads to improved precision values and to more accurate features. The proposed method improves the localization such that 72.5% of the features increase their precision. Hence, more features are extracted while increasing their repeatability by up to 9% on standard benchmarks.

AB - The scale invariant feature operator (SFOP) detects circular features from an image using a spiral shape model. Special cases of the spiral model are junctions and circular symmetric shapes. The spatial localization is determined with subpixel accuracy which is obtained by an interpolation of the structure tensor in the scale space. For the interpolation, SFOP uses a 3D quadratic function. This leads to suboptimal solutions since the structure tensor surrounding a feature does not show the shape of a 3D quadratic. The aim of this paper is to improve the localization of the features detected by SFOP. A Difference of Gaussians function is proposed for the signal approximation which leads to improved precision values and to more accurate features. The proposed method improves the localization such that 72.5% of the features increase their precision. Hence, more features are extracted while increasing their repeatability by up to 9% on standard benchmarks.

UR - http://www.scopus.com/inward/record.url?scp=84941253812&partnerID=8YFLogxK

U2 - 10.1109/mva.2015.7153189

DO - 10.1109/mva.2015.7153189

M3 - Conference contribution

AN - SCOPUS:84941253812

SP - 295

EP - 298

BT - Proceedings of the 14th IAPR International Conference on Machine Vision Applications

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 14th IAPR International Conference on Machine Vision Applications, MVA 2015

Y2 - 18 May 2015 through 22 May 2015

ER -

By the same author(s)