Most of the approaches for urban characterization in the remote sensing domain, are based only on the use of spectral information, not taking into account the geometric characteristics given by the spatial distribution of urban elements. An urban area can be spatially described on the basis of the distribution of urban elements such as buildings, roads, green areas. The analysis of these local properties may enrich the spectral information, improving the differentiability among suburban areas originally belonging to homogeneous areas. On this basis the analysis of different texture of urban areas was introduced. It is based on the study of the spatial distribution of gray level variations. In this work a window and orientation independent approach is presented, based on the detection of regions (or objects) surrounded by elements of separation. This method is performed using the analysis of connected regions. Texture features are calculated within the detected regions to group regions with the same geometric structure. The data set used in this study is formed by panchromatic SPOT satellite images, that highlight urban texture. In order to validate the approach, this method has been applied on Como, Catania, Pavia and Piacenza (Italy), representative samples of different urban typologies.

Date Published: 11 December 1998
PDF: 7 pages
Proc. SPIE 3496, Earth Surface Remote Sensing II, (11 December 1998); doi: 10.1117/12.332713

Published in SPIE Proceedings Vol. 3496:
Earth Surface Remote Sensing II
Giovanna Cecchi; Eugenio Zilioli, Editor(s)