Research on Statistic-Based Image Segmentation Method


Article Preview

The paper researched the image segmentation method based on statistic. For the multiple class segmentation, the K-means segmentation was employed in the first part. The segmentation method named OTSU is discussed in the second part of this paper. To solve the problem of the image noise, the method based on the Markov Random Field (MRF) is proposed in the third part of the paper. The ICM optimization algorithm is used in the procedure of MRF segmentation. In the experiments part, the methods are compared with each other, and the results showed that the method based on MRF are more efficient to remove the noise in the images.



Edited by:

B. Xu and H.Y. Li






Y. M. Hou et al., "Research on Statistic-Based Image Segmentation Method", Advanced Materials Research, Vol. 461, pp. 575-578, 2012

Online since:

February 2012




[1] Zhang Yujin: Image Engineering, Beijing (2007).

[2] Pal N.R., Pal S.K.: A Review on Image Segmentation Techniques, Pattern Recognition, 16(9), 1177-1194 (1994).

[3] Isa N.A.M., Salamah SA, Ngah UK: Adaptive fuzzy moving K-mean clustering algorithm for image segmentation, IEEE Transactions on Consumer Electronics., 55(4), 2145-2146 (2009).

DOI: 10.1109/tce.2009.5373781

[4] Mac Queen: Some Methods for Classification and Analysis of Multivariate Observations, Math. Statist, (1), 281-297 (1967).

[5] Otsu N.: A threshold Selection Method from Gray-level Histograms, IEEE Trans on Systems, 9(1), 62-66 (1979).

[6] Dubes R.C., Jain A.K.: Random Field Models in Image Analysis, Journal of Applied Statisties, 16(2), 131-164 (1989).

[7] Tolpekin V.A., Hamm N.A.S.: Fuzzy Super Resolution Mapping Based on Markov Random Fields, IEEE Geosci. Remote Sensing Symp, 875-878 (2008).

DOI: 10.1109/igarss.2008.4779134

In order to see related information, you need to Login.