OPMAOP: Opposite Pair Matching Approach in Offline Palmprint

Authentication by biometric verification is becoming increasingly common in corporate, public security and other such systems. There is scads of work done in the area of offline palmprints like palmprint segmentation, crease extraction, special areas, feature matching etc. But to the best of our knowledge no work has been done yet to extract and identify the right hand of a person, given his/her left hand or vice versa from a given database. This kind of identification assumes special significance in cases like bomb blasts, air crash etc., where body parts of various persons get mutilated and mixed up. A framework has been designed where palmprint feature vectors are extracted using 2-D wavelet transform and then a OPMAOP Clustering algorithm (proposed in this paper) is applied to cluster the palmprints to get the opposite hand. Using this approach one can easily achieve the said target with a very high accuracy rate. The FAR of the result has been discussed graphically in subsequent sections.