One image segmentation method based on Otsu and fuzzy theory seeking image segment threshold

Based on the study of image segmentation algorithm, this paper presents one kind of improved image segmentation algorithm by comprehensive using the Otsu method and the method based on fuzzy theory seeking image segment threshold. The method uses Otsu method, obtained by pre-segmentation threshold value; then the image is divided into target and background. At last we use fuzzy partition method for final segmentation threshold. The experimental results show that the method can keep a greater degree of image detail. Especially for low SNR images and poor contrast image, there is a relatively good result.

[1]  Guoliang Fan,et al.  Unsupervised Bayesian image segmentation using wavelet-domain hidden Markov models , 2003, Proceedings 2003 International Conference on Image Processing (Cat. No.03CH37429).

[2]  Richard G. Baraniuk,et al.  Multiscale image segmentation using wavelet-domain hidden Markov models , 2001, IEEE Trans. Image Process..

[3]  Qian Huang,et al.  Quantitative methods of evaluating image segmentation , 1995, Proceedings., International Conference on Image Processing.

[4]  Robert D. Nowak,et al.  Wavelet-based statistical signal processing using hidden Markov models , 1998, IEEE Trans. Signal Process..

[5]  Donald Geman,et al.  Stochastic Relaxation, Gibbs Distributions, and the Bayesian Restoration of Images , 1984, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[6]  Klaus D. Tönnies,et al.  Segmentation of medical images using adaptive region growing , 2001, SPIE Medical Imaging.

[7]  Charles A. Bouman,et al.  A multiscale random field model for Bayesian image segmentation , 1994, IEEE Trans. Image Process..

[8]  Guoliang Fan,et al.  Unsupervised image segmentation using wavelet-domain hidden Markov models , 2003, SPIE Optics + Photonics.