Wavelet-based circular hough transform and its application in embryo development analysis

Marcelo Cicconet, Davi Geiger, Kris Gunsalus

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Detecting object shapes from images remains a challenging problem in computer vision, especially in cases where some a priori knowledge of the shape of the objects of interest exists (such as circle-like shapes) and/or multiple object shapes overlap. This problem is important in the field of biology, particularly in the area of early-embryo development, where the dynamics is given by a set of cells (nearly-circular shapes) that overlap and eventually divide. We propose an approach to this problem that relies mainly on a variation of the circular Hough Transform where votes are weighted by wavelet kernels, and a fine-tuning stage based on dynamic programming. The wavelet-based circular Hough transform can be seen as a geometric-driven pulling mechanism in a set of convolved images, thus having important connections with well-stablished machine learning methods such as convolution networks.

Original languageEnglish (US)
Title of host publicationVISAPP 2013 - Proceedings of the International Conference on Computer Vision Theory and Applications
Pages669-674
Number of pages6
Volume1
StatePublished - 2013
Event8th International Conference on Computer Vision Theory and Applications, VISAPP 2013 - Barcelona, Spain
Duration: Feb 21 2013Feb 24 2013

Other

Other8th International Conference on Computer Vision Theory and Applications, VISAPP 2013
CountrySpain
CityBarcelona
Period2/21/132/24/13

Fingerprint

Hough transforms
Convolution
Dynamic programming
Computer vision
Learning systems
Tuning

Keywords

  • Biological cells
  • Circular hough transform
  • Machine vision

ASJC Scopus subject areas

  • Computer Vision and Pattern Recognition

Cite this

Cicconet, M., Geiger, D., & Gunsalus, K. (2013). Wavelet-based circular hough transform and its application in embryo development analysis. In VISAPP 2013 - Proceedings of the International Conference on Computer Vision Theory and Applications (Vol. 1, pp. 669-674)

Wavelet-based circular hough transform and its application in embryo development analysis. / Cicconet, Marcelo; Geiger, Davi; Gunsalus, Kris.

VISAPP 2013 - Proceedings of the International Conference on Computer Vision Theory and Applications. Vol. 1 2013. p. 669-674.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Cicconet, M, Geiger, D & Gunsalus, K 2013, Wavelet-based circular hough transform and its application in embryo development analysis. in VISAPP 2013 - Proceedings of the International Conference on Computer Vision Theory and Applications. vol. 1, pp. 669-674, 8th International Conference on Computer Vision Theory and Applications, VISAPP 2013, Barcelona, Spain, 2/21/13.
Cicconet M, Geiger D, Gunsalus K. Wavelet-based circular hough transform and its application in embryo development analysis. In VISAPP 2013 - Proceedings of the International Conference on Computer Vision Theory and Applications. Vol. 1. 2013. p. 669-674
Cicconet, Marcelo ; Geiger, Davi ; Gunsalus, Kris. / Wavelet-based circular hough transform and its application in embryo development analysis. VISAPP 2013 - Proceedings of the International Conference on Computer Vision Theory and Applications. Vol. 1 2013. pp. 669-674
@inproceedings{c9cdf0f77f184eac81817168d96f9bc6,
title = "Wavelet-based circular hough transform and its application in embryo development analysis",
abstract = "Detecting object shapes from images remains a challenging problem in computer vision, especially in cases where some a priori knowledge of the shape of the objects of interest exists (such as circle-like shapes) and/or multiple object shapes overlap. This problem is important in the field of biology, particularly in the area of early-embryo development, where the dynamics is given by a set of cells (nearly-circular shapes) that overlap and eventually divide. We propose an approach to this problem that relies mainly on a variation of the circular Hough Transform where votes are weighted by wavelet kernels, and a fine-tuning stage based on dynamic programming. The wavelet-based circular Hough transform can be seen as a geometric-driven pulling mechanism in a set of convolved images, thus having important connections with well-stablished machine learning methods such as convolution networks.",
keywords = "Biological cells, Circular hough transform, Machine vision",
author = "Marcelo Cicconet and Davi Geiger and Kris Gunsalus",
year = "2013",
language = "English (US)",
isbn = "9789898565471",
volume = "1",
pages = "669--674",
booktitle = "VISAPP 2013 - Proceedings of the International Conference on Computer Vision Theory and Applications",

}

TY - GEN

T1 - Wavelet-based circular hough transform and its application in embryo development analysis

AU - Cicconet, Marcelo

AU - Geiger, Davi

AU - Gunsalus, Kris

PY - 2013

Y1 - 2013

N2 - Detecting object shapes from images remains a challenging problem in computer vision, especially in cases where some a priori knowledge of the shape of the objects of interest exists (such as circle-like shapes) and/or multiple object shapes overlap. This problem is important in the field of biology, particularly in the area of early-embryo development, where the dynamics is given by a set of cells (nearly-circular shapes) that overlap and eventually divide. We propose an approach to this problem that relies mainly on a variation of the circular Hough Transform where votes are weighted by wavelet kernels, and a fine-tuning stage based on dynamic programming. The wavelet-based circular Hough transform can be seen as a geometric-driven pulling mechanism in a set of convolved images, thus having important connections with well-stablished machine learning methods such as convolution networks.

AB - Detecting object shapes from images remains a challenging problem in computer vision, especially in cases where some a priori knowledge of the shape of the objects of interest exists (such as circle-like shapes) and/or multiple object shapes overlap. This problem is important in the field of biology, particularly in the area of early-embryo development, where the dynamics is given by a set of cells (nearly-circular shapes) that overlap and eventually divide. We propose an approach to this problem that relies mainly on a variation of the circular Hough Transform where votes are weighted by wavelet kernels, and a fine-tuning stage based on dynamic programming. The wavelet-based circular Hough transform can be seen as a geometric-driven pulling mechanism in a set of convolved images, thus having important connections with well-stablished machine learning methods such as convolution networks.

KW - Biological cells

KW - Circular hough transform

KW - Machine vision

UR - http://www.scopus.com/inward/record.url?scp=84878233243&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84878233243&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:84878233243

SN - 9789898565471

VL - 1

SP - 669

EP - 674

BT - VISAPP 2013 - Proceedings of the International Conference on Computer Vision Theory and Applications

ER -