Content Based Image Retrieval and Support Vector Machine Methods for Face Recognition
Content Based Image Retrieval and Support Vector Machine Methods for Face Recognition
Author(s): Anton Satria Prabuwono, Wendi Usino, Arif Bramantoro, Khalid Hamed S. Allehaibi, A. Hasniaty, Tomi DefisaSubject(s): ICT Information and Communications Technologies
Published by: UIKTEN - Association for Information Communication Technology Education and Science
Keywords: Face recognition; Content-based image retrieval; Euclidean distance;Support Vector Machine;
Summary/Abstract: The development of biometrics is growing rapidly. The recognition as non-trivial element in biometrics is not only using fingerprints, but also human face. The purpose of this research is to implement both Content Based Image Retrieval (CBIR) and Support Vector Machine (SVM) methods in the face recognition system with a combination of features extraction. CBIR method interprets images by exploiting several features. The feature usually consists of texture, color, and shape. This research utilizes color, texture, shape and shape coordinate features of the image. The proposed algorithms are HSV Color Histogram, Color Level Co-Occurrence Matrix (CLCM), Eccentricity, Metric, and Hierarchical Centroid. SVM method is used to train and classify the extracted vectors. The result shows that the proposed system is 95% accurate in recognizing faces with different resolutions.
Journal: TEM Journal
- Issue Year: 8/2019
- Issue No: 2
- Page Range: 389-395
- Page Count: 7
- Language: English