Fingerprint elastic deformation correction method and device

Abstract

The invention relates to a fingerprint elastic deformation rectification method and device, and comprises blocking of the fingerprint; selecting central points of the central template image and the to-be-measured fingerprint image; Seeking for characteristic points with the nearest distance to the central points as vertexes and base points of the triangle matched with the template; and the vertexes and base points of the triangle matched with the to-be-measured fingerprint. Determining where the two matched triangles are inter-matched; taking the orthocenter of the matched triangle as the reference point for elastic deformation rectification; determining parameters of initial elastic deformation rectification; determining parameters of the elastic deformation rectification; and carrying out elastic deformation rectification for characteristic data of the to-be-measured fingerprint using the rectifying parameters after two-dimensional spatial domain filtering. The fingerprint elastic deformation rectification parameters are derived through analyzing differences of digitalized information between the to-be-measured sample and the fingerprint template. The elastic deformation of the to-be-measured sample is rectified in accordance with the rectifying parameters, so the fingerprint matching accuracy is increased, and error rate and rejection rate of finger matching is decreased.

Description

technical field [0001] The invention relates to an automatic fingerprint identification system, in particular to a method and device for correcting elastic deformation of fingerprints. Background technique [0002] Automated fingerprint identification system (AFIS for short) has a wide application background. At present, the research on automatic fingerprint identification system mainly includes four aspects, namely image enhancement, fingerprint classification, fingerprint feature extraction and detail matching. The block diagram of the fingerprint identification system is shown in Figure 1. Among them, the processing of fingerprint deformation is concentrated in the detail matching, and the process is mainly the deformation correction of the feature information of the sample to be tested and the matching degree calculation of the feature points within the matching threshold. The prior art mainly solves the problem of rigid deformation of fingerprints, and does not propose...

AI Technical Summary

Problems solved by technology

Although the above processing methods can solve the problem of surface deformation very well, the application of fingerprint recognition is mostly in the case of limited computing resources. Obviously, the three-dimensional surface method with complex matrix changes is not suitable for fingerprint recognition in the environment of limited computing resources. technology application

[0004] In the field of graphics processing applications such as computer-aided design, the consumption of computing resources is not the main consideration, so its surface deformation processing method cannot be applied in the field of fingerprint recognition with limited resources.

[0005] In the resource-constrained fingerprint recognition application, only the rigid deformation of the fingerprint is corrected, and no special correction method is proposed for the elastic deformation

Images