Unit 7

Question 1

Fingerprint matching

Answer 1

Is a process of evaluating the degree of similarity or differences between two given fingerprints

Question 2

Matching problems occur whenever there are:

Answer 2

~ Small inter-class differences
~ Large intra-class differences
The intra-class variations are particularly problematic, as they are much more likely to happen

Question 3

Reasons for intra-class variations:

Answer 3

1) Displacement
2) Rotation
3) Pressure of the impression
4) Skin condition
5) Condition of the sensor surface
6) Feature extraction accuracy
(Figure slide 5)

Question 4

Displacement

Answer 4

Different parts of the fingertip are presented to the sensor

Question 5

Rotation

Answer 5

The fingertip is presented to the sensor at a different angle

Question 6

Pressure of the impression

Answer 6

The finger is pressed onto the sensor with a different force and effort

Question 7

Skin condition

Answer 7

On different occasions, the fingertip maybe dry, wet, scratched or dirty

Question 8

Condition of the sensor’s surface

Answer 8

On different occasions, the surface may be clean, dirty, dusty, or greasy

Question 9

Matching methods for fingerprints can be broadly classified into three categories:

Answer 9

1) Correlation-based matching
2) Minutiae-based matching
3) Ridge-feature-based matching

Question 10

Correlation-based matching:

Answer 10

Works by superimposing one image over another image and changing their alignments (by moving and rotating one of the images) until the correlation between the corresponding pixels of the two images is maximized. A downside however is that the time and resources required to match the images pixel by pixel are huge

Question 11

Minutiae-based matching:

Answer 11

Addresses the correlation-based shortcomings
In a minutiae-based matching, the matching is conducted by working out the similarity between minutiae instead of pixels.
It is the most commonly used matching method adopted by automatic fingerprint recognition systems

Question 12

Ridge-feature-based matching:

Answer 12

Is used when the qualities of the fingerprint images are not good (minutiae extraction is difficult)
In a Ridge-feature-based matching, other ridge features may be used to help in matching fingerprints such as: ridge orientation, ridge frequency, shape and texture,and ridge counts between minutiae

Question 13

There are three main steps in the matching process

Answer 13

1) Alignment
2) Evaluating the distance between each pair of minutiae
3) Determining the outcome