Lab 2: Fingerprints CSE 402
Objective Apply image processing techniques from previous lab to fingerprint images
Fingerprint image Our fingers have a pattern of ridges and valleys We can compare the pattern of two fingers If two patterns are similar enough, we mark them as a match Separate lectures on fingerprints, this is only for lab exercise. Show what ridge ending & bifurcations are in the image.
binary Image We first binarize the image There are only two values in the image (black and white) Each pixel is marked as foreground (white) or background (black) The foreground indicates the ridge structure
Ridge skeleton image We convert a binary image to a ridge skeleton image Each ridge in the binary image is shrunk to be 1 pixel wide The ridge locations are recorded Geometric and dimensional details of the ridges are ignored
Minutiae Points We are interested in 2 types of minutiae points Ridge endings Ridge bifurcations We can detect these points with the ridge skeleton image We can compare the location and orientation of these points to match images Ridge Ending Bifurcation
Noise Images are not always “clean” They are often corrupted by different types of noise We will look at how “salt-and- pepper” noise can be reduced in an image
Image after Median Filtering Noise Removal Noisy Image Image after Median Filtering
Noise Affects Minutiae detection Minutiae from Noisy Image Minutiae from Filtered Image change color from green Many spurious minutiae points