Comparative Study on Different Fingerprint Classification Approaches

Automatic fingerprint recognition has been broadly useful as a personal identification device owing their reliability and uniqueness features. Fingerprints are most promising common and trusted biometric for personal identification. Fingerprint Classification is an important sub-problem for the automatic fingerprint identification system (AFIS) and automatic fingerprint recognition system (AFRS). This database may be very large in many forensic and civilian applications. In such cases, the identification takes a long response time. The authentication process can be fastened by reducing the number of comparisons that are required to be performed. Fingerprint classification was introduced to reduce the searching time of fingerprint recognition in a large database. In this database is divided into numbers of bins based on the fingerprint class. Therefore the unknown fingerprint needs to be match only with specific bin from which class the fingerprint is belonging. However, no literature to date explicitly documents all fingerprint classification approaches. On the other hand, this paper studies some of the current classification approaches and present the comparative studies of the various classification approaches and compares all the classification approaches based on the attributes like nos. of classes are used, characteristics, applied methods on which database, the accuracy of approach and problems of approach, if any.