SVM Based Approach for Multiface Detection and Recognition in Static Images

Abstract

Recognizing and identifying a face from the real world, capture data that senses images is the demanding process in this advanced world. Because of varied face appearances, lighting effects and illumination of the background of the images, perceiving and recognizing multiple faces in a single image is a challenging process. This paper proposes a method that recognizes multiple faces in a single image using a different face recognition algorithm. Here, different approaches of face recognition using OpenCV and SVM algorithm have been compared and implemented for recognizing the multiple faces in a single image. In this method, the Haar Cascade Classifier, which is given by Viola Jones is used to detect the multiple faces in a single image. Local binary pattern histogram, eigenfaces and fisherfaces and Support Vector Machine learning algorithms are used to recognize multiple faces in a single image. These multiple face recognition algorithms are compared and tested over a different set of images.