/fingerprint_recognition

An implementations of fingerprint recognition algorithm

Primary LanguagePythonMIT LicenseMIT

Fingerprint recognition algorithms

comparation guide

fingerprint

Summary

This repository proposes a fingerprint recognition chain based on based on generic algorithms and tools for filtering images. The results are retrieved and validated using Python.

Prerequisites

  • python 3.7
  • python opencv

How to use it

install requirement

pip install --user --requirement requirements.txt

Run fingerprint detection. Finegerprint_pipline.py will process images from './sample_inputs/' and results will be store at './output/'

python finegerprint_pipline.py

Dataset

Dataset ussed for this project can be found in this LINK

Algorithm Pipeline

The techniques of fingerprint recognition image enhancement are based on Gabor Filters process, normalization, segmentation of ridge region, morphological thinning and estimation of the local orientation of ridges in a fingerprint.

pipeline

Results

The following image shows pipeline results at each step

Ending (Green) - Bifurcation (Blue) - Delta (Red) - Loop (Orange) - Whorl (Pink)

results

Reference