Title :
Image Segmentation Based on Gabor Filter and MRF Model
Author :
Wei, Xiaoli ; Shen, Weiming
Author_Institution :
LIESMARS, Wuhan Univ., Wuhan, China
Abstract :
This paper proposed a Gabor filter and MRF model for image segmentation. First, extract color features through transformation of color space; second, get image texture through Gabor filtering and Gaussian smoothing of the original color texture image; and then set up MRF image segmentation model, combining the color and texture features, to calculate the maximum posterior probability(MAP), and using ICM algorithm to optimize the computing complexity. We also proposed parameters estimation method using EM algorithm. Experiment shows that this mixture feature model is efficient than using only color or texture features.
Keywords :
Gabor filters; computational complexity; expectation-maximisation algorithm; feature extraction; image colour analysis; image segmentation; image texture; maximum likelihood estimation; EM algorithm; Gabor filter; Gaussian smoothing; MRF model; computing complexity; feature extraction; image segmentation; maximum posterior probability; original color texture image; Color; Colored noise; Feature extraction; Filtering; Gabor filters; Gray-scale; Image segmentation; Image texture; Image texture analysis; Smoothing methods;
Conference_Titel :
Wireless Communications, Networking and Mobile Computing, 2009. WiCom '09. 5th International Conference on
Conference_Location :
Beijing
Print_ISBN :
978-1-4244-3692-7
Electronic_ISBN :
978-1-4244-3693-4
DOI :
10.1109/WICOM.2009.5301963