Abstract
We present a robust facial recognition method in this paper, which utilizing improved multi-histogram based features from spatial domain, as well as frequency domain, respectively. In spatial domain, we utilize Local Binary Pattern (LBP) histogram due to its excellent robustness and high discriminative capability. In frequency domain, we utilize two types of histogram named binary vector quantization (BVQ) histogram and energy histogram extracted from low-frequency (LF) DCT domain. The former histogram feature is made by applying BVQ on DCT coefficients per frequency band. The latter is energy histogram which can be considered to add amplitude information of DCT coefficients. In addition, in order to add spatial information of a facial image so as to increase the accuracy, the region-division (RD) algorithm is applied. Publicly available databases are utilized to evaluate our proposed method, and experiments demonstrate that the proposed method utilizing multi-histogram features can realize superior recognition performance.
Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
