Face Recognition Using Binary SIFT and it’s Robustness Against Face Variations

SIFT algorithm is one of the most notable algorithm being used for feature extraction. In order to detect the object, these extracted features should be matched with the features extracted from the target image. For the matching purpose there are number of algorithms (Euclidian distance, Cityblock, correlation etc) to compute the distance between extracted features. But the space and time complexity of this algorithm is high enough to meet the real time requirements because of having the large feature vector space (256¹²⁸). To overcome this drawback, Binary-SIFT is introduced by Kadir A. Peker having a very small feature vector space (comparatively) and meets the real time requirements by using the XOR function for matching purpose which needs very less time in comparison with the technique mentioned above. In this paper we have done the performance evaluation of SIFT and Binary-SIFT on test images of Indian face database and The ORL Database of Faces. The performance of both the algorithms is same if the variation (change in scale, illumination and rotation) in the image is very low. But as the variation increases the Binary-SIFT algorithm began to lack the performance in comparison with SIFT.