An Image Segmentation Calculation Based on Differential Box-Counting of Fractal Geometry

Abstract:

Differential box-counting of fractal geometry has been widely used in image processing.A method which uses the differential box-counting to segment the gathered images is discussed in this paper . It is to construct a three-dimensional gray space and use the same size boxes to contain the three dimensional space.The number of boxes needed to cover the entire image are calculated .Different sizes of boxes can receive different number of boxes, so least squares method is used to calculate the fractal dimension. According to the fractal dimension parameters, appropriate threshold is chose to segment the image by using binarization .From the handle case of bearing pictures can be seen that image segmentation based on differential box-counting method can get clear image segmentation .This method is easy to understand, to operate, and has important significance on computer image segmentation .

• Periodical: Applied Mechanics and Materials (Volumes 719-720)
• Pages: 964-968
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.719-720.964
• Online since: January 2015
• Authors: Tao He*, Long Fei Cheng, Qing Hua Wu, Zheng Jia Wang, Lian Gen Yang, Lang Yu Xie
• Keywords: Binarization, Fractal Dimension Differential Box-Counting, Image Segmentation

References

[1] W. KennethFalconer, Fratal geometry-Mathematical foundations and applications, third ed., Chichester: John Wiley & Sons Ltd, (1990).

[2] J. Pentland A P, Fractal-based description of natural scenes, IEEE Transactions on Pattern Analysis and Machine Intelligence, 6(1984)661-674.

DOI: 10.1109/tpami.1984.4767591

[3] J. Sarkar N, Chaudhuri B B. An efficient differencial box-counting approach to compute fractal dimension of image, IEEE Transactions on System, Man and Cybernetics, 24(1994)115-120.

DOI: 10.1109/21.259692

[4] J. Liu Ming-qing, Zhang Xiao-guang. Application of differencial box-counting in detecting edge of weld, Hebei Journal of Industrial Science and Technology, 26(2009)230-232.

[5] J. Yang Shu-shen, Shao Long-yi, Estimation of fractal dimensions of image based on MATLAB, Journal of China University of Mining & Technology, 35(2006)478-482.

[6] J. Zhang Tao, Sun Lin, Huang Ai-min. Research on improvement of DBC algorithm for image's fractal dimension calculation. Electronics Optics &Control, 14(2007)54-57.

[7] J. Yu Hai-jing, Li Gui-ju. Smoke segmention method based on improved differential box=counting fractal dimension, Chinese Journal of Liquid Crystals and Displays, 28(2013)115-119.

DOI: 10.3788/yjyxs20132801.0115

[8] J. Yai Yuan, Liang Zhi-yi, Image segmention for space target based on differential box counting, Computer Science, 39(2012)359-361.