Title :
Unsupervised textured image segmentation
Author :
Gregoriou, George K. ; Tretiak, Oleh J.
Author_Institution :
Imaging & Comput. Vision Center, Drexel Univ., Philadelphia, PA, USA
Abstract :
A new algorithm for unsupervised textured image segmentation is presented. The image comprises M textured regions, each of which is modeled by a stationary Gaussian Markov random field. A feature vector is computed for each pixel in the original image where these vectors are normally distributed and cluster about some vector means. Thus, the problem is reduced to one of restoring a vector valued underlying field embedded in additive Gaussian noise. The vector means corresponding to the different regions are estimated by using the expectation-maximization (EM) algorithm. An iterative algorithm is used with the underlying field modeled as a multilevel logistic Markov random field. The results obtained on two-region and four-region textured images are impressive, and the classification error is less than 3%. The algorithm is not limited to textured images but can also be applied to any vector-valued signals
Keywords :
Markov processes; image segmentation; image texture; iterative methods; maximum likelihood estimation; random noise; additive Gaussian noise; classification error; expectation-maximisation algorithm; feature vector; four-region textured images; iterative algorithm; multilevel logistic Markov random field; stationary Gaussian Markov random field; two-region textured images; unsupervised textured image segmentation; Additive noise; Clustering algorithms; Distributed computing; Gaussian noise; Image restoration; Image segmentation; Iterative algorithms; Logistics; Markov random fields; Pixel;
Conference_Titel :
Acoustics, Speech, and Signal Processing, 1992. ICASSP-92., 1992 IEEE International Conference on
Conference_Location :
San Francisco, CA
Print_ISBN :
0-7803-0532-9
DOI :
10.1109/ICASSP.1992.226273
