Fingerprint image thinning method based on formwork

Abstract

The invention relates to a thinning method of a binarized fingerprint image in fingerprint image preprocessing. It includes constructing neighborhood templates, deleting templates and retaining templates, and using the neighborhood template to extract the neighborhood of foreground pixels from the upper left corner of the image, and then comparing them with the deleting template and retaining template to determine whether to delete or retain the image. point, this operation scans from the upper left corner to the lower right corner of the image and ends as a cycle, and iterates repeatedly until no pixel value changes during the comparison. The feature of the present invention is to establish deletion templates and retention templates in 16 processing directions, so that the operations of deleting and retaining foreground pixels are more refined and accurate. It overcomes the shortcomings of the existing OPTA method and the improved OPTA method, such as incomplete thinning, many burrs in the lines, and twisted lines, etc., and achieves ideal thinning effects. At the same time, this effect is not achieved at the cost of increased processing time, but due to the rational design of the template and the reduction of the number of iterations.

Description

technical field [0001] The invention relates to a fingerprint image preprocessing method before fingerprint identification, in particular to a template-based fingerprint image thinning method. Background technique [0002] The preprocessing of fingerprint image is the most important part in automatic fingerprint identification system, and the result of preprocessing will directly affect the reliability and accuracy of identification. Therefore, the preprocessing of fingerprint image has become a research hotspot in the field of fingerprint identification. The preprocessing of the fingerprint image includes two major processing steps: binarization of the image and refinement of the binarized fingerprint image. Generally speaking, the binarization of the fingerprint image is to give the valley line of the fingerprint and the background of the image a white gray value. , and give the ridges of the fingerprint a black gray value to strengthen the ridges of the fingerprint; the t...

AI Technical Summary

Problems solved by technology

[0048] From its processing results, it can be seen that image 3 Compare figure 2 The refinement is complete and the ridge glitches have improved, but the effect is still not ideal

Analyzing the reason, the main reason is that the refined template is not perfect. For example, its deletion template only considers deleting pixels from eight directions: east, south, west, north, southeast, northeast, southwest, and northwest. Deleting the template is relatively rough, which will cause a lot of pixels that should be deleted to be ignored in each thinning scan, and glitches appear in the thinned image, and the number of thinning iterations is large, which is time-consuming

Images