Unraveling Multimodality with Large Language Models.pdf
Reconstruction of fingerprints
1. Juhi Khandelwal (13/PIT/031)
Laxmi Kant(13/PIT/032)
Manish Bana(10/ICS/021)
Sonam Singh Bhati(13/PIT/038)
Sudhakar Mishra(13/PIT/040)
UNDER THE GUIDANCE OF
ER. SANDHYA TARAR
2. Fingerprint Identification is the method of identification using the
impressions made by the minute ridge formations or patterns found on he
fingertips.
No two persons have exactly the same arrangement of ridge patterns
The patterns of any one individual remain unchanged throughout life.
A fingerprint in its narrow sense is an impression left by the friction ridges
of a human finger.
A friction ridge is a raised portion of the epidermis on the digits (fingers and
toes), the palm of the hand or the sole of the foot, consisting of one or more
connected ridge units of friction ridge skin.
INTRODUCTION
3. Fingerprint identification is the process of comparing two instances of friction
ridge skin impressions from human fingers or toes to determine whether these
impressions could have come from the same individual
4. LITERATURE SURVEY
Paper 1: Arun Ross, Jidnya Shah, Anil K. Jain, “From Template to Image:
Reconstructing Fingerprints from Minutiae Points” IEEE TRANSACTION
ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE VOL 29, NO. 4,
APRIL 2007
PAPER 2: N.Lalithamani and K.P. Soman, “Irrevocable Cryptographic Key
Generation from Cancelable Fingerprint Templates: An Enhanced and
Effective Scheme”, European Journal of Scientific Research (2009)
6. SOFTWARE SPECIFICATION
MATLAB (R2011a):
MATLAB (Matrix Laboratory) is a multi-paradigm numerical computing
environment and fourth-generation programming language. Developed by
Math Works.
MATLAB allows matrix manipulations, plotting of functions and data,
implementation of algorithms, creation of user interfaces, and interfacing with
programs written in other languages, including C, C++, Java, and FORTRAN.
The Image Processing Toolbox is a collection of functions that extend the
capability of the MATLAB numeric computing environment.
The graphical tool provides an automatically generated threshold.
7.
8. MINUTIAE
Minutiae are major features of a fingerprint, using which comparisons
of one print with another can be made.
Minutiae includes:
1. Ridge ending
2. Ridge bifurcation
3. Short ridge
4. Island
5. Ridge enclosure
6. Spur
7. Crossover or bridge
8. Delta
9. Core
11. 1. SEGMENTATION
The input fingerprint image is segmented from the background to actually
extract the region comprising the fingerprint, which ensures removal of
noise.
.
12. 2. ORIENTATION FIELD ESTIMATION
A fingerprint field orientation map may be described as an ensemble
of two dimensional direction fields.
Two approaches
1. filter band based approach
2. gradient based approaches
3. IMAGE ENHANCEMENT
The enhancement of fingerprint images involves the following:
Average filtering and Gabor filtering.
Initially the image is filtered with the help of average filter to
correct the frequency of the image.
The impact of noise can be decreased through simple averaging.
Subsequently, Gabor filtering is applied to the image for further
enhancement.
13.
14. 4. MINUTIAE POINTS EXTRACTION
The steps involved in the extraction of minutiae points are as
follows: Binarization, Morphological Operations, Minutiae points
extraction.
The binary images with only two levels of interest: The black
pixels that denote ridges and the white pixels that denote valleys.
Then thinning process is performed to reduce the thickness of
the lines so that the lines are only represented except the other
regions of the image.
16. 1. ESTIMATING RIDGE ORIENTATION USING
MINUTIAE POINTS
This algorithm has four main steps
1. Triplet Generation
2. Orientation Prediction
3. Triplet Pruning
4. Orientation Smoothing.
17. Fig7:(a) Minutiae distribution of a fingerprint.
(b) Examples of a good quality triplet (blue) with
(c) Estimated orientation map.
18. 1. Triplet Generation
Consider a minutiae template, M, of a fingerprint containing N minutiae points given
by, M {m1, m2 ,…mn) where mi = (xi, yi, 𝞱i). A set of 3 minutiae points, {mi}i=1;2;3,
characterized by a triangle with sides {L
i}i=1;2;3 and interior angles {φi}i=1;2;3 is said to
constitute a `valid' triplet, T.
2. Orientation Prediction
Consider a pixel P(x;y) located inside the triangular region defined by the triplet, T. Let
di = dist{mi, P}, i = 1; 2; 3, be the Euclidean distance of this pixel from all the three
vertices such that d1 < d2 < d3. The orientation of the pixel, O(x,y), is then computed
as,
19. 3. Triplet Pruning
In a fingerprint image, minutiae tend to appear in clusters. For instance, the regions
near the core and delta have dense minutiae activity. It is therefore possible for a
triplet to reside inside the triangular region of another triplet or overlap with it.
4. Averaging The Orientation Map
To obtain a smooth transition in orientations, the predicted orientation map
is convolved with a 3 *3 local averaging filter.
21. 2. CONSTRUCTING STREAMLINES
It uses a linear interpolation scheme for constructing streamlines.
Once a streamline is initiated from a seed point, the next position
is obtained by updating its current position based on the
orientations of the immediate neighbors of the seed point.
The streamline is terminated if it encounters a boundary point in
the grid,
If it arrives in the vicinity of a minutia point.
Streamlines generated and original fingerprint
22. 3. GENERATING RIDGE STRUCTURE
In order to impart texture-like appearances to ridges, we use Linear Integral
Convolution(LIC)
The application of the streamlines results in the generation of binarized thin
ridge lines for the estimated orientation map.
The LIC technique involves calculating the intensity of all the pixels
constituting the streamline.
it locally blurs an uncorrelated input texture image such as white nose,
along the path of the streamlines to impart a dense visualization of the flow
field.
23. 4. ENHANCING THE RIDGE MAP
To enhance the ridge width
Use a low pass filter to smooth the texture image
Then perform histogram equalization of the ridge structure for contrast
enhancement.
Reconstructing fingerprints. (a) Minutiae distribution of a fingerprint image.
(b) Predicted orientation map. (c) Reconstructed fingerprint.
25. IMAGE THINNING
Ridge Thinning is to eliminate the redundant pixels of ridges till the
ridges are just one pixel wide.
Thinned Image
26. MINUTIAE EXTRACTION
Finally, the minutiae points are extracted from the enhanced fingerprint image.
The steps involved in the extraction of minutiae points are as follows:
1. Binarization.
2. Morphological Operations.
3. Minutiae points Extractions.
28. BINARIZATION
The Binarization step is basically stating the obvious, which is that the true
information that could be extracted from a print is simply binary; ridges vs.
valleys.
Binarization
32. SCOPE OF PROJECT
An approach may be used here to enhance the accuracy of generated
friction ridge structure. More sophisticated numerical integration methods
can be used to improve the quality of streamlines.
The major scope of this project is to answer the question: How much
information does the minutiae template reveal about the original fingerprint
image? The benefits of such a study are twofold: 1) It helps in
understanding the vulnerability of decrypted fingerprint templates to a
masquerade attack and 2) It provides insight into the individuality of
fingerprints as assessed using the minutiae distribution.
This project can also be further extended so that it can help the
organizations to generate the fingerprints from minutiae if the original
database has been manipulated or any other mishap has happened.
33. CONCLUSION
In our project first minutiae points are extracted and then minutiae points are
used to generate fingerprints. Thus it is concluded that minutiae points can
reveal substantial information about the parent fingerprint.
There is a scope of further improvement in terms of efficiency and accuracy
which can be achieved by improving the hardware to capture the image or by
improving the image enhancement techniques.
34. REFERENCES
C.Hill, “Risk of Masquerade Arising from the Storage of Biometrics”,
Master’s thesis, Australian Nat’l University, 2001.
A. moenssen, Fingerprint technique. Chilton Book Company, 1971.
B.Sherlock and D. Monro,” A Model for Interpreting Fingerprint Topology,”
Pattern Recognition. Vol. 26, no 7, pp. 1047-1055, 1993.
K.Karu and A.Jain “Fingerprint Classification,” Pattern Recognition, vol. 29.
389-404 2000.
D.Maltoni, D.Maio, A.K Jain, and S.Prabhakar, Handbook of Fingerprint
recognition, first ed. Springer-Verlag, 2003.
Jain A.K, Ross, A. and Prabhakar, S,”An introduction to biometric
recognition,” IEEE Transactions on Circuit and Systems for Video
Technology vol.14, no. 1 pp.4-20, 2004.
35. T.C Clancy, N.Kiyavash and D.J. Lin,”Secure smart card –based fingerprint
authentication,”proceedings of the 2003 ACM SIGMM Workshop on Biometric
Methods and Application,WBMA 2003 .
John Chirillo and Scott Blaul, “Implementing Biometric Security,” Wiley Red
Books, ISBN 978-0764525025, April 2003.
In.Wikipedia.org/wiki/minutiae
F.Galton, Finger Prints. Mc Millan 1996.
N.Lalithamani and K.P. Soman, “Irrevocable Cryptographic Key Generation
from Cancelable Fingerprint Templates: An Enhanced and Effective Scheme”,
European Journal of Scientific Research (2009).
Bryan Morse, ”Low-Pass Filtering”