Please use this identifier to cite or link to this item:
http://hdl.handle.net/2080/357
Full metadata record
DC Field | Value | Language |
---|---|---|
dc.contributor.author | Mohapatra, P J | - |
dc.contributor.author | Nanda, P K | - |
dc.contributor.author | Panda, Sucheta | - |
dc.date.accessioned | 2006-11-20T04:43:08Z | - |
dc.date.available | 2006-11-20T04:43:08Z | - |
dc.date.issued | 2006 | - |
dc.identifier.citation | Proceedings of Soft Computing Technique for Engineering Applications SCT-2006, March 24-26, Rourkela, India | en |
dc.identifier.uri | http://hdl.handle.net/2080/357 | - |
dc.description | Copyright for this article belongs to National Institute of Technology, Rourkela, India | en |
dc.description.abstract | In this paper color image segmentation is accomplished using MRF model. The problem is formulated as a pixel labeling and the true labels are modeled as the MRF model. The observed color image is assumed to be the degraded version of the true labels. We assume the degradation to be Gaussian distribution. The label estimates are obtained by using Bayesian framework and MAP criterion. The (I1, I2, I3) color model is used to represent the color and the MAP estimate is obtained by Simulated Annealing. Simulation results of both indoor and outdoor scenes are considered to validate our approach. | en |
dc.format.extent | 899606 bytes | - |
dc.format.mimetype | application/pdf | - |
dc.language.iso | en | - |
dc.publisher | NIT, Rourkela, India | en |
dc.subject | (I1, I2, I3) color Space | en |
dc.subject | MAP Estimate | en |
dc.subject | Segmentation | en |
dc.subject | Simulated Annealing (SA) | en |
dc.title | Color Image Segmentation Using MRF Model and Simulated Annealing | en |
dc.type | Article | en |
Appears in Collections: | Conference Papers |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
Pmohapatra_pknanda_spanda_SCT20062.pdf | 878.52 kB | Adobe PDF | View/Open |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.