A Segmentation Model Based On Gray Information And Variance Of The Cv

Image segmentation technology is one of the important tasks in image-processing. The purpose of image segmentation is making the part that people take interest in separated from the rest of the image. As traditional methods, region segmentation technology, edge segmentation technology and texture segmentation technology are widely used in the field of image segmentation. Recent years, image segmentation technology, which is based on partial differential equations, is becoming very popular and having a widespread concern in image segmentation.We study the image segmentation method based on partial differential equations. Firstly, the existing methods of image segmentation are introduced, at the same time; the ideas of image segmentation, the meaning and purpose of the image segmentation are summarized. Secondly, the Snake model, the Mumford-Shah model and the Chan-Vese model that based on partial differential equations are analysed and introduced. Then, the curve evolution theory, the level set solution and numerical calculation methods of partial differential equation model are introduced in detail.Thirdly, the advantages and disadvantages df C-V model are discussed. When it comes to non-binary image and multi-object image, the ideal segmentation result can not be achieved through C-V model easily.In this paper, an improved C-V model is proposed, which takes advantage of the regional gray information and regional variance information in an image. Lastly, a new method is proposed to advance the operating efficiency of improved C-V model. The experimental results show that a more accurate result can be achieved through the improved method when processing the non-binary image and multi-object image. In addition to that, the number of iterations is reduced, the segmenting time is shortened and the operating efficiency is improved.