Presentation is loading. Please wait.

Presentation is loading. Please wait.

Small-Area Fingerprint Verification

Similar presentations


Presentation on theme: "Small-Area Fingerprint Verification"— Presentation transcript:

1 Small-Area Fingerprint Verification
曾子家

2

3

4

5 True match

6

7 False match

8 Outline 7 Moments MSER: Section 1. Minutiae matching Normalization
Enhancement Binarization Thinning Minutiae extraction Minutiae descriptor Matching Section 2. Local features matching 7 Moments MSER: Maximally Stable Extremal Regions

9 Minutiae matching Normalization Enhancement Binarization Thinning
Minutiae extraction Minutiae descriptor Matching Minutiae matching Fingerprint Verification System using Minutiae Extraction Technique (Kaur, 2008)

10 Normalization Histogram equalization add in fingerprint

11 Normalization Histogram equalization add in fingerprint

12 Enhancement

13 Binarization

14 Thinning Fingerprint Reference Point Detection Based on High Curvature Points

15 Minutiae matching

16 Local features matching
7 Moments MSER: Maximally Stable Extremal Regions Local features matching

17 7 Moments

18 Introduce “Moments” Definition: Weighted average (moment) of the image pixels' intensities Usage: Describe objects after segmentation Raw moment: Centroid: Centroid mement: Digital Image Processing (3rd Edition) 3 by Rafael C. Gonzalez

19 Invariants Translation The nature of centroid moment Scale Rotation

20 http://limitless-thoughts. blogspot

21 http://limitless-thoughts. blogspot

22 http://limitless-thoughts. blogspot

23 Similar pattern, similar value?

24 Exp. Compare images someone_0_ someone_0_10

25 8 Splits someone_0_ someone_0_10

26

27

28

29 Same image -> maskSize, characteristics

30 maskSize = 3

31 maskSize = 10

32 maskSize = 20

33 maskSize = 30

34 maskSize = 40

35 maskSize = 50

36 Summary Insignificant results in every mask size.

37 MSER Maximally Stable Extremal Regions

38

39

40

41

42

43 Introduce “MSER” Affine transformation-invariant Scale-invariant

44

45

46 Normalization

47 Affine transforamtion
Ellipses to circular Intensity normalization

48

49

50 Summary Shortcomings Can only detect few features Not robust to blur

51 References M. Dubey and S. Sahu, “Fingerprint Minutiae Extraction and Orientation Detection Using ROI (Region Of Interest) for Fingerprint Matching,” Orientation-Detection.html, 2017. M. Kaur, et al., “Fingerprint Verification System using Minutiae Extraction Technique,” system-using-minutiae-extraction-technique, 2017. Wikipedia, “Histogram Equalization,”

52 References N. H. Barnouti, “Fingerprint Recognition Improvement Using Histogram Equalization and Compression Methods”, J. –F. Mainguet, “Fingerprint Algorithms: Algorithmes de Reconnaissance d‘Empreintes Digitales,” R. C. Gonzalez, Digital Image Processing, 3rd Edition, Pearson, London, England, 2007.

53 References Limitless Thoughts, “Hu's Seven Moments Invariant (Matlab Code for invmoments.m),” seven-moments-invariant-matlab-code.html, 2017. Yung-Yu Chuang, “Digital Visual Effect,”


Download ppt "Small-Area Fingerprint Verification"

Similar presentations


Ads by Google