Abstract: The segmentation of fingerprint images plays a very important role in Automatic Fingerprint Identification System (AFIS). Effective segmentation can not only reduce the time of subsequent processing, but also improve the reliability of feature extraction considerably. This paper mainly contains two works; proposed a new segmentation feature for fingerprint images which is called Effective Point Cluster Degree (EPCI));proposed an effective method for the segmentation of fingerprint images based on the new feature and C1uD mentioned in reference[l]，in this new method,we first segmented fingerprint images with EPCD，and then processed the first segmentation results with iteration-based me thod, did the second segmentation with CLuD based on the processed results, and finally, morphology was applied as postprocessing to reduce misclassification. All experimental results show that: compared with other commonly used features, the EPCD holds the characteristics of good discriminability, robustness, and the segmented foreground and background are more concentrated; at the same time, the segmented method based on EPCD and CluD possesses high accuracy and adaptability.

Key words: Biometrics, AFIS, Fingerprint image segmentation, Feature extraction, Effective point cluster degree