Title :
A fuzzy shell clustering approach to recognize hyperbolic signatures in subsurface radar images
Author :
Delbò, S. ; Gamba, P. ; Roccato, D.
Author_Institution :
Dipt. di Elettronica, Pavia Univ., Italy
fDate :
5/1/2000 12:00:00 AM
Abstract :
The authors propose a pattern recognition approach to analyze subsurface radar images and recognize the hyperbolic signatures produced by targets. After enhancing these signatures using a wavelet denoising procedure, pixels are grouped into significant hyperbolic shapes by fuzzy clustering. The approach also provides a validation measure for each recognized shape using so-called “shell thickness”. Excellent experimental results justify the use of this algorithm for automatic interpretation of subsurface radar images
Keywords :
buried object detection; geophysical signal processing; geophysical techniques; pattern recognition; radar imaging; remote sensing by radar; terrain mapping; algorithm; automatic interpretation; buried object detection; fuzzy clustering; fuzzy shell clustering; geophysical measurement technique; ground penetrating radar; hyperbolic shape; hyperbolic signature; land surface; pattern recognition; radar remote sensing; shape; shell thickness; subsurface radar image; terrain mapping; wavelet denoising; Clustering algorithms; Ground penetrating radar; Image analysis; Image recognition; Noise reduction; Pattern analysis; Pattern recognition; Shape measurement; Target recognition; Thickness measurement;
Journal_Title :
Geoscience and Remote Sensing, IEEE Transactions on
