This work intends to test the use of remotely sensed data, as a mean to identify degraded lands with a high environmental hazard. The approach uses data from the sensor Thematic Mapper on Landsat 5 in synergy with digital ortho-photos (1:10000) and land cover map Corine 1990 to create a methodology useful to identify areas with dumps. The analyzed scene is relative to an area located in the Apulia Region in Southern Italy, where it is known the presence of a dump near the Margherita di Savoia "saline" (salt evaporation pool). As this dump is in its early phase, it is impossible to use thermal anomaly as a characteristic sign of its presence. So its identification proceeds through the extraction of the spectral signatures of the dump area and of the neighborhood zones. The analysis is developed in three steps: (1) Monitoring the change in the zone nearby the pools, especially if abandoned; (2) Pointing out the dump presence by the spectral signature specificity; (3) Individuating areas characterized by the same spectral properties. A pre-processing analysis is carried out by the Principal Component Transformation in order to minimize spectral noise and redundancy. Subsequently, the images are classified by the unsupervised algorithm ISODATA aiming at automatically individuating radiometric classes. The regions of interest are identified by help of the land cover map and then characterized by their spectral signatures. The identification of the dump is a feasible objective because of the temporal stability of its spectral signature, with respect to those of the other areas.
[1]
Vincent V. Salomonson,et al.
The Landsat Program: Its Origins, Evolution, and Impacts
,
1997
.
[2]
William A. Brenneman.
Statistics for Research
,
2005,
Technometrics.
[3]
P. Duncombe,et al.
Multivariate Descriptive Statistical Analysis: Correspondence Analysis and Related Techniques for Large Matrices
,
1985
.
[4]
L. Bartolucci,et al.
Atmospheric effects on Landsat TM thermal IR data
,
1988
.
[5]
J. Campbell.
Introduction to remote sensing
,
1987
.
[6]
R. Rockwell,et al.
The detection of vegetational change by multitemporal analysis of LANDSAT data: the effects of goose foraging
,
1998
.
[7]
A. Huete.
A soil-adjusted vegetation index (SAVI)
,
1988
.
[8]
P. Switzer,et al.
A transformation for ordering multispectral data in terms of image quality with implications for noise removal
,
1988
.
[9]
Robert A. Schowengerdt,et al.
Remote sensing, models, and methods for image processing
,
1997
.