Title :
Hybridization of colour information with texture by using mutual information for the classification of satellite images
Author :
El Maia, Hassan ; Hammouch, Ahmed ; Aboutajdine, Driss
Author_Institution :
Fac. of Sci., Mohamed V-Agdal Univ., Rabat, Morocco
Abstract :
In this article a construction of hybrid colour space from a set of classic colours spaces by using the algorithm MRMR is proposed. This algorithm is based on the mutual information. Our approach is evaluated on SPOT HRV (XS) image representing two forest areas in the region of Rabat. Feature extraction is done by the cooccurrence matrix. The SVM (Support Vectors Machine) classifier is used.
Keywords :
feature extraction; image classification; image colour analysis; image texture; matrix algebra; support vector machines; SPOT HRV image representation; cooccurrence matrix; feature extraction; hybrid colour space; minimal-redundancy-maximal-relevance algorithm; mutual information; satellite image classification; support vector machine classifier; texture colour information; Feature extraction; Gray-scale; Heart rate variability; Image analysis; Image color analysis; Image texture analysis; Mutual information; Satellites; Support vector machine classification; Support vector machines; MRMR; SVM; colour; cooccurrence matrix; hybrid color space; satellite image; texture;
Conference_Titel :
Multimedia Computing and Systems, 2009. ICMCS '09. International Conference on
Conference_Location :
Ouarzazate
Print_ISBN :
978-1-4244-3756-6
Electronic_ISBN :
978-1-4244-3757-3
DOI :
10.1109/MMCS.2009.5256671
