TY - GEN
T1 - On the segmentation of fingerprint images
T2 - 2012 International Conference on Image Processing, Computer Vision, and Pattern Recognition, IPCV 2012
AU - Msiza, Ishmael S.
AU - Nelwamondo, Fulufhelo V.
AU - Marwala, Tshilidzi
PY - 2012
Y1 - 2012
N2 - Fingerprint image segmentation is the process of separating the fingerprint ridge area - that is, the foreground - from the background of the captured image. It is an important process because it prevents a fingerprint recognition system from analysing the unnecessary fragments of the captured image. This is the reason why segmentation should be executed during the initial stages of the fingerprint analysis process. This manuscript presents a method based on the gray-level variance of the captured image, where this variance is calculated per block of image pixels, instead of doing it for each image pixel. This then necessitates careful and informed choices for the values of - both - the gray-level variance threshold, and the pixel-block size. This manuscript solely focuses on studying the effects of these two variable parameters, and the results indicate that the choice of their respective values should be well-informed.
AB - Fingerprint image segmentation is the process of separating the fingerprint ridge area - that is, the foreground - from the background of the captured image. It is an important process because it prevents a fingerprint recognition system from analysing the unnecessary fragments of the captured image. This is the reason why segmentation should be executed during the initial stages of the fingerprint analysis process. This manuscript presents a method based on the gray-level variance of the captured image, where this variance is calculated per block of image pixels, instead of doing it for each image pixel. This then necessitates careful and informed choices for the values of - both - the gray-level variance threshold, and the pixel-block size. This manuscript solely focuses on studying the effects of these two variable parameters, and the results indicate that the choice of their respective values should be well-informed.
KW - Fingerprint segmentation
KW - Gray-level variance
KW - Pixel-block size
KW - Variance threshold
UR - http://www.scopus.com/inward/record.url?scp=84873315084&partnerID=8YFLogxK
M3 - Conference contribution
AN - SCOPUS:84873315084
SN - 9781601322258
T3 - Proceedings of the 2012 International Conference on Image Processing, Computer Vision, and Pattern Recognition, IPCV 2012
SP - 1107
EP - 1113
BT - Proceedings of the 2012 International Conference on Image Processing, Computer Vision, and Pattern Recognition, IPCV 2012
Y2 - 16 July 2012 through 19 July 2012
ER -