Latent Fingerprint Image Segmentation using …vislab.ucr.edu/Biometrics16/CVPR_presentation_June_2016.pdfFigure: Sample latent fingerprints from NIST SD27:- Quality levels (a) good,

Latent Fingerprint Image Segmentation using

Fractal Dimension Features and

Weighted Extreme Learning Machine Ensemble

Jude Ezeobiejesi and Bir Bhanu

Center for Research in Intelligent Systems

University of California, Riverside, CA 92521, USA

e-mail: [email protected], [email protected]

June 26,2016

Outline Outline

Problem definitionProblem Definition

Related work Related Work

Technical approachTechnical Approach

Extreme Learning Machine(ELM)

and Weighted ELM

Extreme Learning Machine (ELM)

and Weighted ELM

Performance evaluation metricsPerformance Evaluation Metrics

Experiments – ELM hyper parameter selection

Experiments – ELM Hyper Parameter Selection

Fractal dimension (FD) features

Fractal Dimension

FD Lacunarity

Fractal Dimension (FD) Features

Experiments - setupExperiments - Setup

Results & comparison with other segmentation approaches

Results & Comparison with Other

Segmentation Approaches

Conclusions and future workConclusions and Future Work