DocumentCode :
2911641
Title :
A Supervised Method for Determining Displacement of Gray Level Co-Occurrence Matrix
Author :
Nikoo, Hassan ; Talebi, Hossein ; Mirzaei, Abdolreza
Author_Institution :
Dept. of Electr. & Comput. Eng., Isfahan Univ. of Technol., Isfahan, Iran
fYear :
2011
fDate :
16-17 Nov. 2011
Firstpage :
1
Lastpage :
5
Abstract :
Gray Level Co-occurrence Matrix (GLCM) is one of the most powerful methods for extracting texture information. GLCM extracts occurrence probability of intensities with a specific displacement between pair pixels. Most applications of GLCM use a single displacement or a distinct number of displacements. Maximum displacement between pair pixels is dependent on Texel size. So approaches that use distinct displacements for all textures taken from different distances from texture surface do not classify textures properly. This paper introduces a supervised method for determining Texel size of a texture. Texel size can be used as the maximum displacement for constructing gray-level co-occurrence matrix. This method uses variances of intensifies in scale space levels to extract features. The proposed method calculates Texel size of 91.9% of test textures correctly.
Keywords :
feature extraction; image texture; matrix algebra; probability; Texel size; feature extraction; gray level co-occurrence matrix; intensity occurrence probability; supervised method; texture displacement determination; texture information extraction; Aluminum; Databases; Feature extraction; Kernel; Probability density function; Surface texture; Training;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Machine Vision and Image Processing (MVIP), 2011 7th Iranian
Conference_Location :
Tehran
Print_ISBN :
978-1-4577-1533-4
Type :
conf
DOI :
10.1109/IranianMVIP.2011.6121563
Filename :
6121563
Link To Document :
بازگشت