Supported by the Natural Science Research Project of Anhui Province Education Office of China under Grant No.2002KJ234 (安徽省教育厅自然科学研究项目)
It is very important to detect singularities (core and delta) accurately and reliably for classification and matching of fingerprints. In this paper, a method for singularity detection in fingerprint images is presented to improve accuracy of the position and reliability of the singularity. Firstly, the singularities are detected based on block images through shifting position of the whole image time after time at the same block size and the concentrative region of singularities detected under different positions is got and the centroid of the region is computed to gain the accurate position of singularities. Then, the reliability of singularities detected above is determined with multilevel block sizes. In this method, the characteristics of the relative concentration of the position of singularities detected through image shift and of the corresponding relationship of the singularities detected with multilevel block sizes are used and the singularities are detected accurately and reliably. Experimental results show that the method performs well and it is robust to poor quality images.