FACE RECOGNITION BY GREY-LEVEL CO-OCCURRENCE MATRICES IN HEXAGONAL DIGITAL IMAGE PROCESSING


Face Recognition (FR) has been an attractive research area for decades, since the face is one of the handiest, and deterministic biometrics. Image processing has been square-pixel-based, that is called Square-based Image Processing (SIP), since the date of its presence. Inherently, studies on facial recognition have also been built on the square pixel processing logic. However, hexagonal image processing (HIP), which is based on the idea of designing the pixels hexagonally, can yield significant benefits in terms of saving time and memory. In this study, a HIP based face recognition method is proposed. The method is inspired by one of the fundamental methods, Gray Level Co-occurrence Matrices (GLCM). Since it is based on the ordinary GLCM, the name Hex_ Direct_GLCM is given to the method. As presented by the simulation results, Hex_ Direct_GLCM poses competitive results regarding recognition accuracy in addition to saving time and space.


Keywords


face recognition, gray-level co-occurrence matrices, hexagonal image processing.

Author : Nazife ÇEVİK
Number of pages: 149-165
DOI: http://dx.doi.org/10.29228/TurkishStudies.22825
Full text:
Share:
Alıntı Yap:
Turkish Studies-Information Technologies and Applied Sciences
E-Mail Subscription

By subscribing to E-Newsletter, you can get the latest news to your e-mail.