TY - JOUR
T1 - Wavelets for urban spatial feature discrimination
T2 - Comparisons with fractal, spatial autocorrelation, and spatial co-occurrence approaches
AU - Myint, Soe Win
AU - Lam, Nina S.N.
AU - Tyler, John M.
N1 - Funding Information:
This revision was accomplished during a time of support by the National Heart, lung and Blood Institute (GenTAC) and the National Human Genome Research Center (Center of Excellence in ELSI Research P50-HG-004487), both of the US National Institutes of Health, and while in residence at the Brocher Foundation, Hermance, Switzerland. I also thank all of my colleagues who shared their published and unpublished work and commented on sections of the chapter, especially Harry C. Dietz, M.D. and Irene H. Maumenee, M.D.
PY - 2004/7
Y1 - 2004/7
N2 - Traditional image processing techniques have proven inadequate for urban mapping using high spatial resolution remote-sensing images. This study examined and evaluated wavelet transforms for urban texture analysis and image classification using high spatial resolution ATLAS imagery. For the purpose of comparison and to evaluate the effectiveness of the wavelet approaches, two different fractal approaches (isarithm and triangular prism), spatial autocorrelation (Moran's I and Geary's C), and spatial co-occurrence matrix of the selected urban classes were examined using 65 X 65, 33 X 33, and 17 X 17 samples with a pixel size of 2.5 m. Results from this study suggest that a multi-band and multi-level wavelet approach can be used to drastically increase the classification accuracy. The fractal techniques did not provide satisfactory classification accuracy. Spatial autocorrelation and spatial co-occurrence techniques were found to be relatively effective when compared to the fractal approaches. It can be concluded that the wavelet transform approach is the most accurate of all four approaches.
AB - Traditional image processing techniques have proven inadequate for urban mapping using high spatial resolution remote-sensing images. This study examined and evaluated wavelet transforms for urban texture analysis and image classification using high spatial resolution ATLAS imagery. For the purpose of comparison and to evaluate the effectiveness of the wavelet approaches, two different fractal approaches (isarithm and triangular prism), spatial autocorrelation (Moran's I and Geary's C), and spatial co-occurrence matrix of the selected urban classes were examined using 65 X 65, 33 X 33, and 17 X 17 samples with a pixel size of 2.5 m. Results from this study suggest that a multi-band and multi-level wavelet approach can be used to drastically increase the classification accuracy. The fractal techniques did not provide satisfactory classification accuracy. Spatial autocorrelation and spatial co-occurrence techniques were found to be relatively effective when compared to the fractal approaches. It can be concluded that the wavelet transform approach is the most accurate of all four approaches.
UR - http://www.scopus.com/inward/record.url?scp=4143148548&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=4143148548&partnerID=8YFLogxK
U2 - 10.14358/PERS.70.7.803
DO - 10.14358/PERS.70.7.803
M3 - Review article
AN - SCOPUS:4143148548
SN - 0099-1112
VL - 70
SP - 803
EP - 812
JO - Photogrammetric Engineering and Remote Sensing
JF - Photogrammetric Engineering and Remote Sensing
IS - 7
ER -