Fingerprint Classification Using Double k-Means Clustering

Fingerprint clusters recognize fingerprint images faster by minimizing the database search space. To this end, the Gaussian filter algorithm is first used to improve the image contrast and calculate the directional fields of the fingerprint, and then the K-means algorithm is used twice, the first time, it is used to classify the fingerprint in terms of the length and width of the fingerprint, and the age group it belongs to, the second time, it is used to classify the fingerprint in terms of to which pattern the direction of ridgelines belong, plain arch, tented arch, ulnar loop, central pocket loop, plain whorl, and double whorl. Due to the smaller search space in the database, good quality results and faster fingerprint recognition are produced.