1 / 22

Iris Recognition Under Various Degradation Models

Iris Recognition Under Various Degradation Models. Hans Christian Sagbakken. Outline. Introduction Scope and research questions Experimental setup Results Conclusions. Introduction. Biometrics technology.

elani
Download Presentation

Iris Recognition Under Various Degradation Models

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Iris Recognition Under Various Degradation Models Hans Christian Sagbakken

  2. Outline • Introduction • Scope and research questions • Experimental setup • Results • Conclusions Hans Christian Sagbakken

  3. Introduction Hans Christian Sagbakken

  4. Biometrics technology • Biometrics refers to technologies that measure and analyze human physical and behavioural characteristics • Examples of characteristics include fingerprints, eye retinas and irises, facial patterns and handmeasurements • Two main application: • Verification (mobile banking) • Identification (security control) Hans Christian Sagbakken

  5. Example of an iris pattern Hans Christian Sagbakken

  6. Iris recognition prosess 1. Segmentation prosess 2. Normalisation prosess 4. Comparison/decision 3. Iris code generation Comparison Hans Christian Sagbakken

  7. Scope and research questions Hans Christian Sagbakken

  8. Research questions • 1. Under which conditions is iris-based recognition feasible? • 2. Which filter to perform under certain degradation conditions? Hans Christian Sagbakken

  9. Scope of the thesis • The thesis was restricted to experiments in MATLAB only • Adapt Libor Masek’s open source code for the experiments (different filters, inter-class and intra-class comparisions) • The iris images degradations are simulated in MATLAB with different parameters (to find the best filter under different conditions) Hans Christian Sagbakken

  10. Experimental setup Hans Christian Sagbakken

  11. Implementation • Expanded Libor Masek’s open source code for iris recognition with four filters • Log-Gabor filter (9600 bit, original filter) • 702-bit Haar wavelet filter • 87-bit Haar wavelet filter • Log of Gaussian filter (9600 bit) • Expanded the search function with inter-class and intra-class comparisons Hans Christian Sagbakken

  12. Iris database • The filters are tested on 500 images from the UBiris database. Five images per person for 100 persons. • The images are simulated with different paramenters • Add noise in the image database (Gaussian noise) • Add blur in the image database • Change the light intensity in the image database • Rotate the images in the database Hans Christian Sagbakken

  13. Evaluation • For each filter under different conditions, the False Acceptance Rate (FAR) and False Rejection Rate (FRR) are computed • Inter-class comparisons (to experiment with FAR). For each test 123,750 comparisons are done • Intra-class comparisons (to experiment with FRR). For each test 1000 comparisons are done • Totally 6,930,000 inter-class and 56,000 intra-class comparisons are performed. Hans Christian Sagbakken

  14. Example of hamming distribution Inter-class comparisons Intra-class comparisons Hans Christian Sagbakken

  15. Example of FAR and FRR Optimal threshold value = 0.32 Hans Christian Sagbakken

  16. Results Hans Christian Sagbakken

  17. Results under noisy conditions Hans Christian Sagbakken

  18. Results under blur conditions Hans Christian Sagbakken

  19. Results under light changes Hans Christian Sagbakken

  20. Results under rotation Hans Christian Sagbakken

  21. Conclusions • Under noisy conditions the best results where achieved with 702-bit Haar wavelet filter • Under blur conditions the best results where achieved with 702-bit Haar wavelet filter • Under light changes the best results where achieved with 702-bit Haar wavelet filter and Log-Gabor filter • Under rotation the best results where achieved with Log-Gabor filter • Totally the best filter is 702-bit Haar wavelet filter Hans Christian Sagbakken

  22. Questions??? Hans Christian Sagbakken

More Related