Robust corner detection based on image structure

Bongjoe Kim, Jihoon Choi, Yongwoon Park, Kwanghoon Sohn

Research output: Contribution to journalArticle

8 Citations (Scopus)

Abstract

In this paper, we propose a robust corner detection method to improve both detection rate and localization accuracy by modifying the structure tensor-based corner detection method in two ways. First, we introduce a connected component analysis (CCA) method for constructing a CCA structure tensor in order to make the structure tensor adaptive to the structure of the image. Second, the normalized cross-correlation (NCC) method is applied for false corner rejection with the observation that the patch of a true corner has a distinctive characteristic compared with connected neighboring patches. The proposed method is compared with several corner detection methods over a number of images. Experimental results show that the proposed method shows better performance in terms of both detection rate and localization accuracy.

Original languageEnglish
Pages (from-to)1443-1457
Number of pages15
JournalCircuits, Systems, and Signal Processing
Volume31
Issue number4
DOIs
Publication statusPublished - 2012 Aug 1

Fingerprint

Corner Detection
Tensors
Correlation methods
Tensor
Connected Components
Patch
Cross-correlation
Rejection
Experimental Results

All Science Journal Classification (ASJC) codes

  • Signal Processing
  • Applied Mathematics

Cite this

Kim, Bongjoe ; Choi, Jihoon ; Park, Yongwoon ; Sohn, Kwanghoon. / Robust corner detection based on image structure. In: Circuits, Systems, and Signal Processing. 2012 ; Vol. 31, No. 4. pp. 1443-1457.
@article{b078febece7f4bb88704a4b5048764f7,
title = "Robust corner detection based on image structure",
abstract = "In this paper, we propose a robust corner detection method to improve both detection rate and localization accuracy by modifying the structure tensor-based corner detection method in two ways. First, we introduce a connected component analysis (CCA) method for constructing a CCA structure tensor in order to make the structure tensor adaptive to the structure of the image. Second, the normalized cross-correlation (NCC) method is applied for false corner rejection with the observation that the patch of a true corner has a distinctive characteristic compared with connected neighboring patches. The proposed method is compared with several corner detection methods over a number of images. Experimental results show that the proposed method shows better performance in terms of both detection rate and localization accuracy.",
author = "Bongjoe Kim and Jihoon Choi and Yongwoon Park and Kwanghoon Sohn",
year = "2012",
month = "8",
day = "1",
doi = "10.1007/s00034-012-9388-z",
language = "English",
volume = "31",
pages = "1443--1457",
journal = "Circuits, Systems, and Signal Processing",
issn = "0278-081X",
publisher = "Birkhause Boston",
number = "4",

}

Robust corner detection based on image structure. / Kim, Bongjoe; Choi, Jihoon; Park, Yongwoon; Sohn, Kwanghoon.

In: Circuits, Systems, and Signal Processing, Vol. 31, No. 4, 01.08.2012, p. 1443-1457.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Robust corner detection based on image structure

AU - Kim, Bongjoe

AU - Choi, Jihoon

AU - Park, Yongwoon

AU - Sohn, Kwanghoon

PY - 2012/8/1

Y1 - 2012/8/1

N2 - In this paper, we propose a robust corner detection method to improve both detection rate and localization accuracy by modifying the structure tensor-based corner detection method in two ways. First, we introduce a connected component analysis (CCA) method for constructing a CCA structure tensor in order to make the structure tensor adaptive to the structure of the image. Second, the normalized cross-correlation (NCC) method is applied for false corner rejection with the observation that the patch of a true corner has a distinctive characteristic compared with connected neighboring patches. The proposed method is compared with several corner detection methods over a number of images. Experimental results show that the proposed method shows better performance in terms of both detection rate and localization accuracy.

AB - In this paper, we propose a robust corner detection method to improve both detection rate and localization accuracy by modifying the structure tensor-based corner detection method in two ways. First, we introduce a connected component analysis (CCA) method for constructing a CCA structure tensor in order to make the structure tensor adaptive to the structure of the image. Second, the normalized cross-correlation (NCC) method is applied for false corner rejection with the observation that the patch of a true corner has a distinctive characteristic compared with connected neighboring patches. The proposed method is compared with several corner detection methods over a number of images. Experimental results show that the proposed method shows better performance in terms of both detection rate and localization accuracy.

UR - http://www.scopus.com/inward/record.url?scp=84864564831&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84864564831&partnerID=8YFLogxK

U2 - 10.1007/s00034-012-9388-z

DO - 10.1007/s00034-012-9388-z

M3 - Article

VL - 31

SP - 1443

EP - 1457

JO - Circuits, Systems, and Signal Processing

JF - Circuits, Systems, and Signal Processing

SN - 0278-081X

IS - 4

ER -