Abstract

An Empirical Analysis of Filtering Techniques on Fingerprint Images

K. Kanagalakshmi and Archana C S*

Head & Associate Professor, PG & Research Department Computer Science, Nehru Arts & Science College, Coimbatore, Tamil Nadu-641105, INDIA. Research Scholar, PG & Research Department Computer Science, Nehru Arts & Science College, Coimbatore, Tamil Nadu-641105, INDIA.

DOI : http://dx.doi.org/10.29055/jcms/939

ABSTRACT

Most recently popular biometric systems are based on recognition and classification of unique finger print patterns. Finger print images have several peaks and valleys on human fingertip. These peaks and valleys form ridge direction and ridge frequency. In open literature of finger print and researches done by novel researcher, it was studied that researches have been done considering ridge direction mostly. Complete image enhancement both features of fingerprint images are required to be addressed. In proposed work 23 samples of fingerprint images are used to find comparison, dataset image taken real time college students these datasets are considered as two types without Mehndi as original image and with mehandi as noise image. Proposed system Keeping the view in mind five filters, such as weiner, median, gradient, knv, and laplacian in the parameters MSE, PSNR, CORELATION and CPU time vales are compared and analyses. Results of this paper have been simulated on MATLAB 2013 version a. From the result This Paper proves that wiener filter is best filter for removing noise from the fingerprint images.

Keywords :Mean Square Error, Peak Signal to Noise Ratio, cpu time, laplacian, gradient, weiner, median, KNV neighbourhood.

map1 map2 map3