A study of the efficacy of using Wavelet Transforms for Palm Print Recognition

Palm Print Recognition is a vital technique in the burgeoning Biometrics industry as an answer to security and identity threats. In this paper we analyze the various details associated with Palm Print Recognition and devise an algorithm to do so that works with Discrete Wavelet Transforms. Palm Prints, being easily discernable and permanent features of the human body, can be subjected to many Fingerprint analysis methods [2]. These methods work by identifying and demarcating the unique lines, ridges etc. on the human palm. The Wavelets used are those of the D.C.T., Eigen, Haar, Hartley, Walsh, Slant Transforms. Further experimenting, we decided to incorporate wavelets of the lesser-used Helmert and Kekre Transforms. It is seen that all these Wavelets give us accuracies close to 93% with our database of over 8000 images. Also, they do so with providing a great decrease in computational requirements due to the usage of Fractional Coefficients. Our study also deals with the importance of segregating these high information zones from each Transformed Image as seen in the results.