The Image Segmentation Based On Wavelet And Curvelet L1Norm

The definition of image segmentation is, for a given digital image, according to our needs, parts of the image segmentation for we are interested in and the other remaining parts.Image segmentation is one of the important subjects in the image processing problem. At present, it has become a important research direction in the field of image processing.Once upon a time, most of the image segmentation problem is processing by using the variational method and partial differential equations, which will make the image separate into one part contains the structure information and the other part contains the texture and noise information. Based on this, the morphological component analysis and sparse representation theory developed, and play an important role in the field of image segmentation. The main idea of Morphological component analysis and sparse representation is, in different dictionaries the different image compositions have its own sparse representation. This article first introduced the image segmentation model that used variational method and partial differential method, with a brief analysis; Secondly introduces the image segmentation model in curvelet domain and wavelet domain. Then introduces the image segmentation model by morphological component analysis and sparse representation; At last, we propose a new image segmentation model in this paper, based on the characteristics of wavelet transform and curvelet transform, combining the traditional LI norm to describe sparse degree, we establish the new model. Then, we use Bregman iteration method to solving the model. Finally, we can segment the image to point characteristics and linear features.