A Literature Survey of various Fingerprint De-noising Techniques to justify the need of a new De-noising model based upon Pixel Component Analysis

TL;DR

This paper reviews various fingerprint de-noising techniques, highlighting their limitations, and proposes the need for a new optimal model based on pixel component analysis to improve biometric authentication robustness.

Contribution

The paper surveys existing de-noising methods and emphasizes the necessity of developing a new model that enhances fingerprint image quality more effectively.

Findings

Gabor filter-based models improve MSE but not SNR significantly

Wavelet-based composite methods marginally enhance image quality

Current models are non-optimal for biometric robustness

Abstract

Image Preprocessing is a vital step in the field of image processing for biometric pattern recognition. This paper studies and reviews various classical and modern fingerprint image de-noising models. The various model used for de-noising ranges widely from transform matrix using frequency, histogram model de-noising, de-noising by introducing Gabor filter and its types to enhance fingerprint images. The output efficiency of various de-noising model proposed earlier is calculated on the basis of SNR (signal to noise ratio) and MSE (mean square error rate). Our simulated experimental results indicates that incorporating the de-noising model based on Gabor filter inside domain of wavelet ranges with composite method only betters MSE (Mean Square Error). Improved MSE without significant improvement in SNR improves the fingerprint images only by a little margin which is non-optimal in nature. Thus the objective of this research paper is to build an optimal de-noising model for fingerprint images so that its usage in biometric authentication can be more robust in nature.