Abstract: In order to effectively detect the singularities in fingerprint images,a method combining multi-scale complex filter and Poincare Index is proposed.Firstly,the directional fields of the fingerprint image at multiple scales are calculated and the direction field of each scale is filtered by using the complex filter.Subsequently the PI values of the core and delta points whose filtering response are greater than a certain threshold are calculated.The singularities whose PI values satisfy the PI judgment condition are identified as the candidate singularities.Then a coarse scale is gradually shifted to the fine scale to refine and filter the candidate singular points.Finally,the accurate singular point detection results are reached.The proposed method is subjectively and objectively compared with approaches in literature by using the open fingerprint image database FVC2004DB1.The experimental results verify that the correct singularity detection ratio of the proposed algorithm is 93%,which is higher than those of the compared methods,indicating the good performance of the proposed method in detecting singularities in poor quality fingerprint images and the accuracy in conserving the genuine singularity location.