Slap fingerprint segmentation

This paper proposes an efficient algorithm to extract all four fingerprints from a slap-image and to identify the indices of these fingers such as index, middle, ring or little finger of left or right hand. Entropy of the image is used to determine different components of fingers, while geometrical and spatial properties of these components are considered to segment fingerprints from the slap image. The algorithm gives accurate results even when there exist some dull prints, large rotational angles or non-elliptical shape of components in the image. It makes use of a new distance measure to index these fingerprint components. It has been tested on a database of 6,096 images of 1,016 subjects, which has accurately segment each slap image into four fingerprints.