J4 ›› 2011, Vol. 29 ›› Issue (02): 152-.

Previous Articles     Next Articles

Image,s Fractal Dimension Calculation and Its Edge Extraction

LI Peng-fei1,XING Li-xin1,PAN Jun1,GU Xiao-feng2   

  1. 1.College of Earth Exploration Science and Technology, Jilin University, Changchun 130026, China|2.School of Resource and Environment Science, Wuhan University, Wuhan 430079, China
  • Online:2011-03-25 Published:2011-04-25

Abstract:

 Edge information in remote sensing image provides a rich content for image understanding.An algorithm to calculate the fractal dimension of a single pixel by using the model of DFBR(Discrete Fractal Brown Random field)is designed.This algorithm is programmed by Matlab. Map the image gray level space to the fractal dimension of space, and then transform and edge detection in the latter one. Spatial distribution of surface features and the differences of image texture structure and edge features, make the selected window size used to calculate fractal dimension as the key. Select part of the high spatial resolution remote sensing image in the study area as the experimental data. By calculating the fractal dimension of pixels in different windows, the calculation window of the best edge detection results is required. The results show that this algorithm is more suitable for remote sensing data. And the accuracy of his remote sensing image extraction is improved.

Key words: fractal dimension, discrete fractal brown random field, edge extraction, remote sensing image

CLC Number: 

  • TP301