This paper reports investigations aimed at developing a variographic tool which visually describes or discriminates perceivable qualities of textures. In particular, it will be shown that the variogram measure commonly used in geostatistic can be used to identify different structures. In texture analysis the first and most important task is to extract texture features which most completely embody information about the spatial distribution of intensity variations in an image. We present, here, an approach to texture segmentation using variography. Semi-variograms are essentially graphs measuring the difference between grade values relative to the distance separating the grade sample locations in a particular orientation. This variance therefore changes as the separation distance increases where repetitive structures are described as hole-effects. Since the nesting structure on the semi-variogram defines the separation distance up to which the primitives can be considered repetitive, it can also be used as a tool for identifying the principal directions and size of texture structures. In all likelihood, this will be the distance that exhibits a hole-effect. Using the textural informations provided by the variogram, we performed segmentation schemes based on the comparison of variogram features in the four main directions of land cover and built-up environment to constitute four classes. Urban regions of Alger (aerial photographs) have been used for evaluating the performance of the segmentation scheme proposed in the characterisation and discrimination of the texture aspect of urban images. The results obtained demonstrate the potential usefulness of the variogram measure. Its success is tied directly to the fidelity with which it captures the underlying structure of urban images.
[1]
S. Rambal,et al.
Identifying spatial patterns of Mediterranean landscapes from geostatistical analysis of remotely-sensed data
,
1994
.
[2]
Noel A Cressie,et al.
Statistics for Spatial Data, Revised Edition.
,
1994
.
[3]
David G. Goodenough,et al.
Comparison of fixed-size and variable-sized windows for the estimation of tree crown position
,
1998,
IGARSS '98. Sensing and Managing the Environment. 1998 IEEE International Geoscience and Remote Sensing. Symposium Proceedings. (Cat. No.98CH36174).
[4]
Phil Brodatz,et al.
Textures: A Photographic Album for Artists and Designers
,
1966
.
[5]
R. Mohan Srivastava,et al.
Robust Measures of Spatial Continuity
,
1989
.
[6]
Geostatistical Analysis of Stereoscopic Pairs of Satellite Images
,
2001
.
[7]
J. M. Hans du Buf,et al.
A review of recent texture segmentation and feature extraction techniques
,
1993
.
[8]
Noel Cressie,et al.
Deterministic/Stochastic Wavelet Decomposition for Recovery of Signal From Noisy Data
,
2000,
Technometrics.
[9]
J. Dubois,et al.
Evaluation Of The Grey-level Co-occurrence Matrix Method For Land-cover Classification Using Spot Imagery
,
1990
.