Title :
Contributions in belief functions and plausible and paradoxical reasoning theories: Application to supervised classification of satellite images
Author :
Ben Dhiaf, Zouhour ; Haouas, Fatm ; Hamouda, Atef
Author_Institution :
Fac. of Sci. of Tunis, Tunis, Tunisia
Abstract :
We present in this paper a new approach to determine the relevant focal elements within the framework of theories of belief functions and plausible and paradoxical reasoning. This approach is based on spectral information represented in the form of gray levels histograms. The particularity of this approach lies in its ability to reduce the sets of composites classes by keeping only the focal elements. Thus, the spatial and temporal complexity of the fusion and classification system based on the two theories is reduced.
Keywords :
geophysical image processing; geophysical techniques; image classification; image fusion; belief functions; classification system; composites classes; focal elements; fusion system; gray level histograms; paradoxical reasoning theories; plausible reasoning; satellite image classification; spatial complexity; spectral information; temporal complexity; Cognition; Complexity theory; Conferences; Histograms; Image resolution; Uncertainty; Belief functions theory; Classification; Focal elements; Plausible and paradoxical reasoning theory;
Conference_Titel :
Image Processing, Applications and Systems Conference (IPAS), 2014 First International
Print_ISBN :
978-1-4799-7068-1
DOI :
10.1109/IPAS.2014.7043308
