[DGJ18] Different approaches for the texture classication of a remote sensing image bank
Conférence Internationale avec comité de lecture :
Ninth International Conference on Graphic and Image Processing,
April 2018,
Vol. 10615(Proc. of SPIE Vol. 10615 1061530-1),
pp.1-9,
Series Series proceedings for ISI2017,
Qingdao,
China,
(
DOI:
doi: 10.1117/12.2302497)
Mots clés: Expert system, Prolog, Neural network, natural textures
Résumé:
In this paper, we summarize and compare two dierent approaches used by the authors, to classify dierent
natural textures. The rst approach, which is simple and inexpensive in computing time, uses a data bank
image and an expert system able to classify dierent textures from a number of rules established by discipline specialists. The second method uses the same database and a neural networks approach.