| Since B.B Mandelbrot proposed the concept of fractal, the fractal had become a new interdisciplinary subject crossing the natural science and social science. Fractal geometry theory is a new branch of mathematics since it is developed from 1980’s. The introduction of the principle of self-similar made the fractal theory become a powerful tool of studying and manipulating the anomaly image in nature and project. Fractal theory served as pioneer for image segmentation technology. Image segmentation technology is a key technology in the field of image processing. It aims to separate the object from the background, and it is also a typical problem.This paper first introduces the background of the fractal theory and its recent achievements. The main topic is to introduce the two methods which are often used on image segmentation currently. The conventional gradient method which is often used in the image segmentation is firstly discussed by using Sobel operator, Candy operator method for image segmentation, and analyzing the effect under the conditions of noise or not of the image segmentation and then to analyze the digital image segmentation techniques which is based on the fractal theory. Then paper has proposed an algorithm which is based on the fractal theory of image segmentation, and proves that the algorithm is superior to the traditional differential operator method In terms of the treatment effect, but is very slow in speed, and then an improvement on the traditional box dimension calculation method is proposed. Finally according to the analysis results, the paper put forward an improved algorithm which used two innovations can greatly reduce the running time of the program, so as to improve image segmentation speed and maintain the accuracy of image segmentation. The application of these methods is also discussed in this paper. At last, the author discusses the application of the method in aerospace image, and gives some advices. |
PDF Full Text Request