Multi-Faces Recognition in Crowd Using Support Vector Machine on Histogram of Gradient

Abstract

Face recognition plays massive role in the biometric identification. On behalf of its usability, face recognition is developed more advanced in term of the accuracy. The use of face recognition as biometric identification varies from security surveillance to business purposes. However, the existing methods for face recognition are mostly for close distance and single face purpose only, meanwhile the demand for face recognition has come into recognizing multi faces in a certain crowdy place. Thus, it is required to select the most suitable approach in order to face this challenge. This study proposes Support Vector Machine (SVM) as the classification method and Histogram of Gradient (HoG) as the feature extraction of the image. This study is expected to give the best recommendation for developing multi face recognition in crowd considering the accuracy given by the proposed method. The result of efficiency and performance is shown by the simulation and analysis.