Enhanced and fast face recognition by hashing algorithm

Abstract

This paper presents a face hashing technique for fast face recognition. The proposed technique employs the two existing algorithms, i.e., 2-D discrete cosine transformation and K-means clustering. The image has to go through different pre-processing phases and the two above-mentioned algorithms must be used in order to obtain the hash value of the face image. The searching process is increased by introducing a modified form of binary search. A new database architecture called Facebases has also been introduced to further speedup the searching process.