Color Image Segmentation Using Gaussian Mixture Model and EM Algorithm - Multimedia and Signal Processing - page 66

image, and use EM algorithm to estimate the various parameters of Gaussian model to
achieve image segmentation. This method's merit lies in using the randomness of the
image, might give a stable result. But here uses only the image's color information, the
segmentation effect is not very ideal. This indicates that if we make full use of the
image's edge, texture and other features, will obtain better segmentation results.
Acknowledgment. This work was supported in part by the National Natural Science
Foundation of China (NSFC) under Grant No. 61071193.
References
1. Permuter, H., Francos, J., Jermyn, I.H.: Gaussian mixture models of texture and colour for
image database retrieval. In: Proc. of the IEEE International Conference on Acoustics,
Speech and Signal Processing (2003)
2. Maclean, J., Jepson, A.: Recovery of ego-motion and segmentation of independent object
motion using the EM algorithm. In: Proceedings BMVC, pp. 175-184 (1994)
3. Yuille, A.L., Stolorz, P., Utans, J.: Statistical physics, mixtures of distributions, and the EM
algorithm. Neural Computation 6, 334-340 (1994)
4. Moss, S., Hancock, E.R.: Registering incomplete radar images using the EM algorithm.
Image and Vision Computing 15, 637-648 (1997)
5. Cho, W.H., Kim, S.H., Park, S.Y., Park, J.H.: Mean field annealing EM for image
segmentation. In: International Conference on Image Processing, vol. 3, pp. 568-571 (2000)
6. Park, J.-H., Cho, W., Park, S.: Deterministic Annealing EM and Its Application in Natural
Image Segmentation. In: Zhang, J., He, J.-H., Fu, Y. (eds.) CIS 2004. LNCS, vol. 3314, pp.
639-644. Springer, Heidelberg (2004)