DocumentCode :
2869807
Title :
Application of high-order fractal feature to edges detection
Author :
Gan, Du ; Zhanghouhong
Author_Institution :
Key Lab. for Radar Signal Process., Xidian Univ., Xi´´an, China
Volume :
2
fYear :
1998
fDate :
1998
Firstpage :
1006
Abstract :
A gray image of natural texture satisfies the model of fractional Brownian random field. The limitation of available methods for edge detection based on fractal characteristics is analyzed and a novel method is proposed. The lacunarity, which is a high-order fractal feature and another fractal parameter (II value) are extracted for detection. With the new method, the experiments show that the edges of gray images can be obtained clearly and completely
Keywords :
Brownian motion; edge detection; feature extraction; fractals; image texture; II value; edge detection; feature extraction; fractal characteristics; fractional Brownian random field; gray image; high-order fractal feature; lacunarity; natural texture; Computer vision; Fractals; Image edge detection; Image segmentation; Laboratories; Motion detection; Motion measurement; Power measurement; Radar detection; Radar signal processing;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Signal Processing Proceedings, 1998. ICSP '98. 1998 Fourth International Conference on
Conference_Location :
Beijing
Print_ISBN :
0-7803-4325-5
Type :
conf
DOI :
10.1109/ICOSP.1998.770784
Filename :
770784
Link To Document :
بازگشت