DocumentCode
658826
Title
Classification of Dayas Formations on Landsat TM Imagery Using Morphological Profiles
Author
Kemmouche, Akila ; Atmani, Nassim
Author_Institution
Telecommun. Dept., Univ. of Sci. & Technol. Houari Boumediene (USTHB), Algiers, Algeria
fYear
2013
fDate
28-30 Oct. 2013
Firstpage
471
Lastpage
475
Abstract
The semi-arid southern piedmont plain in the Algerian Saharan region is characterized by dayas, i.e. small semi-circular depressions colonized by vegetation. These dayas show specific morphological variations and vegetation characteristics. A correlation was noted between their morphology, vegetation cover and stage of development. Thus, it is of interest to develop an approach for combining the vegetation and shape criteria within a feature vector of each pixel to map the dayas. Vegetation and geometrical properties concerning these day a structures, as determined from remote sensing data, were used in this work to perform chronological classification of all dayas in this region. The differential morphological profile and pattern recognition parameters are used for extraction and classification of the dayas formation on Land sat Thematic Mapper imagery. Each class corresponds to a specific morphological and vegetation evolution stage.
Keywords
geophysical image processing; remote sensing; vegetation; Algerian Saharan region; Dayas formation; Land sat thematic mapper imagery; Landsat TM imagery; chronological classification; differential morphological profile; feature vector; geometrical property; morphological variation; pattern recognition parameter; remote sensing data; semiarid southern piedmont plain; semicircular depression; shape criteria; vegetation characteristic; Differential morphological profile; image analysis; mathematical morphology; remote sensing; semi-arid area;
fLanguage
English
Publisher
ieee
Conference_Titel
Broadband and Wireless Computing, Communication and Applications (BWCCA), 2013 Eighth International Conference on
Conference_Location
Compiegne
Type
conf
DOI
10.1109/BWCCA.2013.82
Filename
6690931
Link To Document