Open AccessAnalysis of Detection of Drow and Entire Co-Occurrence Matrix GLCM Method on The Classification of Image
Author(s) -
Taufik Ismail Simanjuntak,
Saib Suwilo,
Rahmat Widia Sembiring
Publication year - 2019
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1361/1/012028
Subject(s) - grey level , randomness , artificial intelligence , pattern recognition (psychology) , co occurrence matrix , support vector machine , computer science , feature extraction , classifier (uml) , image (mathematics) , texture (cosmology) , feature (linguistics) , matrix (chemical analysis) , computer vision , image texture , mathematics , image processing , statistics , materials science , linguistics , philosophy , composite material
Grey level co-occurrence matrix or GLCM is a method for obtaining features of a textural image. On the previous study of GLCM, using only 4-degree degrees 0, 45, 90 and 135 in calculating the result of features. Periodicity, directionalityand randomness are the three most important factors in characterizing textures. Therefore, textural directionality is a basic feature of the image and plays an important role in image descriptions, which can be used to describe image textures. So this research will analyse the effect of modification of directionality or direction of adegree of texture distribution on GLCM by using 8 degree direction, which is degree 0, 45, 90, 135, 180, 225, 270, and 315 to know theeffect that impact on theimage by using support vector machine as aclassifier. The results showed that the SVM classifier was not able to work well on GLCM feature extraction, but the performance of the modification resulted in a consistent increase in accuracy by performing 10 tests.