Search in Medwell
 
 
Journal of Engineering and Applied Sciences
Year: 2019 | Volume: 14 | Issue: 7 | Page No.: 2171-2176
DOI: 10.36478/jeasci.2019.2171.2176  
Local Quadrant Pattern with Co-occurrence Matrix (LQP-CM): Hybrid Method for Image Classification and Feature Extraction
Hassan Mohammed Mahdi Al-Jawahry and Hind Rustum Mohammed
 
Abstract: Image classification is important in several fields which depend on the methods of extracting the features. This study proposes a new method for features extraction called Local Quadrant Pattern with Co-occurrence Matrix (LQP-CM) that related with Local Ternary Pattern (LTP) and Gray-Level Co-occurrence Matrix (GLCM). LQP-CM will map the values into four types instead of two like Local Binary Pattern (LBP) or three like LTP. For classification, this study will use the Euclidean Distance (ED) to classifying the features that extracting. The data set that used in this study is Brodatz dataset. The MATLAB environment was adopted in the programming and the criteria was used to evaluate the performance of the proposed method is percentage of correct classification which proved successful in classification the database used in high efficiency.
 
How to cite this article:
Hassan Mohammed Mahdi Al-Jawahry and Hind Rustum Mohammed, 2019. Local Quadrant Pattern with Co-occurrence Matrix (LQP-CM): Hybrid Method for Image Classification and Feature Extraction. Journal of Engineering and Applied Sciences, 14: 2171-2176.
DOI: 10.36478/jeasci.2019.2171.2176
URL: http://medwelljournals.com/abstract/?doi=jeasci.2019.2171.2176