Finger Knuckle Based Biometric Identification of person

1 Ms. Arti B Waghode, M.E. (Digital System), JSPM’s JSCOE Pune, Maharashtra, India 2 Assistant Professor. C A Manjare, Dept. of E&TC Engineering, JSPM’s, JSCOE, Pune, Maharashtra, India ---------------------------------------------------------------------***--------------------------------------------------------------------Abstract Finger knuckle print can be incredibly beneficial for person identity. FKP as a new biometric modality supplying huge scope for researchers in few years. In this system we present a finger knuckle identification method that uses subspace techniques. In proposed system we use the three subspace techniques, first we use Gabor filter in preprocessing for removing the noise from acquired image and we get the noise free image. Secondly we use PCA for feature extraction and then last we use the LDA as well as Knn classifier for matching purpose. Result obtained from knn classifier and LDA algorithm gives the approximately 98% recognition rate. Also gives the high efficiency as compare to other methods.