Fingerprint Recognition under Missing Image Pixels Scenario

TL;DR

This paper investigates fingerprint recognition when images have missing pixels, using compressive sensing for reconstruction, and demonstrates successful identification with up to 90% missing pixels.

Contribution

It introduces a method applying compressive sensing to recover fingerprint images with missing pixels for recognition tasks, which is a novel application in this context.

Findings

Successful fingerprint recognition with less than 90% missing pixels.

Effective image reconstruction using compressive sensing.

Validation through experiments confirming the approach's viability.

Abstract

This work observed the problem of fingerprint image recognition in the case of missing pixels from the original image. The possibility of missing pixels recovery is tested by applying the Compressive Sensing approach. Namely, different percentage of missing pixels is observed and the image reconstruction is done by applying commonly used approach for sparse image reconstruction. The theory is verified by experiments, showing successful image reconstruction and later person identification even if less then 90% of the image pixels is missing.