图像分割介绍（Introduction to image segmentation）.docVIP

图像分割介绍（Introduction to image segmentation） Digital image processing technology is an interdisciplinary field. With the continuous development of computer science, image processing and analysis has gradually formed a scientific system of their own, the new treatment methods emerge in an endless stream, although its history is not long, but it caused widespread concern in all sectors. First of all, the vision is the most important means of human perception, image is the basis of vision, therefore, the digital image becomes an effective tool for researchers of visual perception physiology, psychology, computer science and other fields in the. Secondly, image processing has a growing demand in military, remote sensing, meteorological and other large applications. Image segmentation based on graph theory is a new research focus in the field of image segmentation in recent years. The method maps the image to weighted undirected graphs and treats pixels as nodes. The minimum cut criterion is used to obtain the best segmentation of the image. This method essentially turns the image segmentation problem into an optimization problem. It is a kind of point to point clustering method. It also has a good application prospect for data clustering. But because of the great theoretical knowledge involved, the application is still in the initial stage. Therefore, domestic studies in this area is rare, this paper will be used for the basic theory of image segmentation are introduced briefly on the graph theory and the method of graph theory for the latest research progress of image segmentation are summarized, and emphatically introduces the method of image segmentation based on graph cuts week. [summary] image target segmentation and extraction technology overview Image segmentation is an important image technology, and has received extensive attention in theoretical research and practical applications. There are many kinds of image segmentation methods. Some segmentation operations