Chen, Guo-Huei and Wilson, Roland (1999) Image segmentation based on the multiresolution Fourier transform and Markov random fields. University of Warwick. Department of Computer Science. (Department of Computer Science research report). (Unpublished)

Preview

PDF (Department of Computer Science Research Report) WRAP_cs-rr-351.pdf - Other - Requires a PDF viewer. Download (614Kb) | Preview

Request Changes to record.

Abstract

In this work, the Multiresolution Fourier Transform (MFT) and Markov Random Fields (MRFs) are combined to produce as a tool for image segmentation. Firstly, a Laplacian Pyramid is used as a high-pass filter. Then, the MFT is applied in order to segment images based on the analysis of local properties in the spatial frequency domain. A methodology for edge detection in image segmentation in the Bayesian framework using Markov random field models is then developed. Stochastic Relaxation is also adopted to maximise the likelihood and find the globally minimum energy states using simulated annealing.

Item Type:	Report
Subjects:	T Technology > TA Engineering (General). Civil engineering (General)
Divisions:	Faculty of Science > Computer Science
Library of Congress Subject Headings (LCSH):	Image segmentation, Markov random fields, Fourier transformations
Series Name:	Department of Computer Science research report
Publisher:	University of Warwick. Department of Computer Science
Official Date:	21 September 1999
Dates:	Date Event 21 September 1999 Completion
Number:	Number 351
Number of Pages:	23
Identifier:	CS-RR-351
Institution:	University of Warwick
Theses Department:	Department of Computer Science
Status:	Not Peer Reviewed
Publication Status:	Unpublished
Related URLs:	Organisation

Request changes or add full text files to a record

Repository staff actions (login required)

View Item

Downloads