Details
| Original language | English |
|---|---|
| Pages (from-to) | 1057-1067 |
| Number of pages | 11 |
| Journal | Visual Computer |
| Volume | 32 |
| Issue number | 6-8 |
| Early online date | 9 May 2016 |
| Publication status | Published - Jun 2016 |
Abstract
Eigenvalue spectra of the Laplace-Beltrami operator have successfully been employed as fingerprints for shape and image comparison. Especially notable in this context is the work of Peinecke on Laplace spectrum fingerprinting for image data. Recently, new research on greyscale images by Berger et al. introduces the idea of attributing individual eigenfunctions to image parts and describes a mechanism for controlling their localisation. These parts are separated by sufficiently strong variations of grey value, giving the originally global fingerprint a semi-local character. This paper provides an approach to extend this idea to colour images so that not only gradients of brightness but also gradients of hue or chroma lead to localisation of eigenfunctions. This is accomplished by generalising the eigenfunctions to R2-valued functions and mapping the colours to symmetric 2 × 2 -matrices. The resulting matrix field is then used to modify the Laplacian. Finally, we present a distance function for comparing eigenvalue-based fingerprints that makes use of eigenfunction colocalization information.
Keywords
- Colour images, Eigenvalue, Fingerprint, Image comparison, Image retrieval, Laplace
ASJC Scopus subject areas
- Computer Science(all)
- Software
- Computer Science(all)
- Computer Vision and Pattern Recognition
- Computer Science(all)
- Computer Graphics and Computer-Aided Design
Cite this
- Standard
- Harvard
- Apa
- Vancouver
- BibTeX
- RIS
In: Visual Computer, Vol. 32, No. 6-8, 06.2016, p. 1057-1067.
Research output: Contribution to journal › Article › Research › peer review
}
TY - JOUR
T1 - Colocalization structures and eigenvalue spectra for colour image comparison
AU - Berger, Benjamin
AU - Wolter, Franz Erich
AU - Vais, Alexander
N1 - Publisher Copyright: © 2016, Springer-Verlag Berlin Heidelberg.
PY - 2016/6
Y1 - 2016/6
N2 - Eigenvalue spectra of the Laplace-Beltrami operator have successfully been employed as fingerprints for shape and image comparison. Especially notable in this context is the work of Peinecke on Laplace spectrum fingerprinting for image data. Recently, new research on greyscale images by Berger et al. introduces the idea of attributing individual eigenfunctions to image parts and describes a mechanism for controlling their localisation. These parts are separated by sufficiently strong variations of grey value, giving the originally global fingerprint a semi-local character. This paper provides an approach to extend this idea to colour images so that not only gradients of brightness but also gradients of hue or chroma lead to localisation of eigenfunctions. This is accomplished by generalising the eigenfunctions to R2-valued functions and mapping the colours to symmetric 2 × 2 -matrices. The resulting matrix field is then used to modify the Laplacian. Finally, we present a distance function for comparing eigenvalue-based fingerprints that makes use of eigenfunction colocalization information.
AB - Eigenvalue spectra of the Laplace-Beltrami operator have successfully been employed as fingerprints for shape and image comparison. Especially notable in this context is the work of Peinecke on Laplace spectrum fingerprinting for image data. Recently, new research on greyscale images by Berger et al. introduces the idea of attributing individual eigenfunctions to image parts and describes a mechanism for controlling their localisation. These parts are separated by sufficiently strong variations of grey value, giving the originally global fingerprint a semi-local character. This paper provides an approach to extend this idea to colour images so that not only gradients of brightness but also gradients of hue or chroma lead to localisation of eigenfunctions. This is accomplished by generalising the eigenfunctions to R2-valued functions and mapping the colours to symmetric 2 × 2 -matrices. The resulting matrix field is then used to modify the Laplacian. Finally, we present a distance function for comparing eigenvalue-based fingerprints that makes use of eigenfunction colocalization information.
KW - Colour images
KW - Eigenvalue
KW - Fingerprint
KW - Image comparison
KW - Image retrieval
KW - Laplace
UR - http://www.scopus.com/inward/record.url?scp=84966521226&partnerID=8YFLogxK
U2 - 10.1007/s00371-016-1260-x
DO - 10.1007/s00371-016-1260-x
M3 - Article
AN - SCOPUS:84966521226
VL - 32
SP - 1057
EP - 1067
JO - Visual Computer
JF - Visual Computer
SN - 0178-2789
IS - 6-8
ER -