Share Email Print

Proceedings Paper

Segmentation of textured images using local spatial-frequency representation
Author(s): Yue Tao
Format Member Price Non-Member Price
PDF $14.40 $18.00
cover GOOD NEWS! Your organization subscribes to the SPIE Digital Library. You may be able to download this paper for free. Check Access

Paper Abstract

This paper presents an algorithm for segmentation of textured images. The algorithm uses an unsupervised neural network and K-means method to classify image pixels based on image local spatial-frequency information. The short-time Fourier transform employs large size window function to extract more neighborhood information. The problems of introducing large size windows in classifying larger transient regions are also investigated. High segmentation resolution is obtained by an novel image extrapolation approach and a re-classification procedure for transient areas.

Paper Details

Date Published: 1 October 1998
PDF: 6 pages
Proc. SPIE 3460, Applications of Digital Image Processing XXI, (1 October 1998); doi: 10.1117/12.323231
Show Author Affiliations
Yue Tao, Research Institute of Marine Boiler and Turbines (China)

Published in SPIE Proceedings Vol. 3460:
Applications of Digital Image Processing XXI
Andrew G. Tescher, Editor(s)

© SPIE. Terms of Use
Back to Top