Multiscale texture segmentation of dip‐cube slices using wavelet‐domain hidden Markov trees
| Autor: | Richard G. Baraniuk, Ivan Magrin-Chagnolleau, Rutger L. C. van Spaendonck, Hyeokho Choi, Philippe Steeghs |
|---|---|
| Rok vydání: | 1999 |
| Předmět: |
business.industry
Computer science Maximum-entropy Markov model ComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISION Scale-space segmentation Wavelet transform Pattern recognition Image segmentation Wavelet Tree structure Computer Science::Computer Vision and Pattern Recognition Segmentation Artificial intelligence Hidden Markov model business |
| Zdroj: | SEG Technical Program Expanded Abstracts 1999. |
| DOI: | 10.1190/1.1820808 |
| Popis: | Wavelet-domain hidden Markov models (HMMs) are powerful tools for modeling the statistical properties of wavelet transforms. By characterizing the joint statistics of wavelet coe cients, HMMs e ciently capture the characteristics of many real-world signals. When applied to images, the model can characterize the joint statistics between pixels, providing a very good classi er for textures. Utilizing the inherent tree structure of waveletdomain HMMs, classi cation of textures at various scales is possible, furnishing a natural tool for multiscale texture segmentation. In this paper, we introduce a new multiscale texture segmentation algorithm based on wavelet-domain hidden Markov trees (HMTs). We apply this new technique to the segmentation of dip-cube time slices. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|