Title :
A novel multifractal estimation method and its application to remote image segmentation
Author :
Du, G. ; Yeo, T.S.
Author_Institution :
Electr. & Comput. Eng. Dept., Nat. Univ. of Singapore, Singapore
fDate :
4/1/2002 12:00:00 AM
Abstract :
Based on the gliding-box and relative differential box-counting algorithms, a novel method that estimates accurately the multifractal exponents, a distinct characteristics of gray-scale digital images, is proposed. Four natural texture images are used to test the performance of the novel multifractal measure. Comparisons with published methods show that the proposed method can efficiently describe texture images and can provide accurate classification results
Keywords :
fractals; geophysical signal processing; geophysical techniques; image classification; image segmentation; image texture; multidimensional signal processing; radar imaging; remote sensing by radar; synthetic aperture radar; terrain mapping; SAR; fractal method; geophysical measurement technique; gliding box algorithm; gray-scale digital images; image classification; image processing; image segmentation; image texture; land surface; multifractal estimation method; multifractal exponents; multispectral method; radar imaging; radar remote sensing; relative differential box counting algorithm; remote sensing; synthetic aperture radar; terrain mapping; Digital images; Fractals; Gray-scale; Image analysis; Image segmentation; Image texture analysis; Remote sensing; Sea ice; Statistics; Testing;
Journal_Title :
Geoscience and Remote Sensing, IEEE Transactions on
DOI :
10.1109/TGRS.2002.1006395
