Fingerprint image segmentation method irrelevant to collecting device

Abstract

The invention discloses a fingerprint image segmentation method irrelevant to a collecting device, which not only has better segmentation effect, but also avoids the defect that when the traditional algorithm segments fingerprint images from different devices, different threshold values need to be set or different classifiers are trained. The fingerprint image segmentation method basically has collecting device independency and is more suitable for Internet application environment which uses multiple types of fingerprint collecting devices. The method comprises the following steps: 1) an input picture is divided into W*W no-overlapping image blocks; 2) CMV indexes of the image blocks are obtained and are normalized; 3) K-means clustering is carried out for the image blocks; 4) the classification of foreground blocks and background blocks is ensured to obtain a primary segmentation result; and 5) a final segmentation result is obtained after the morphologic processing.

Description

technical field [0001] The invention relates to the field of automatic fingerprint identification, in particular to a fingerprint image segmentation method which has nothing to do with collection equipment. Background technique [0002] Fingerprint image segmentation is a very important preprocessing step in the automatic fingerprint identification system. By accurately segmenting the foreground and background, the subsequent processing can be concentrated in the effective area, thereby reducing the amount of calculation and improving the accuracy of fingerprint detail feature acquisition. Fingerprint image segmentation first selects and calculates one or more description features of fingerprints, and then designs and implements a suitable segmentation algorithm. [0003] It should be pointed out that when the existing fingerprint image segmentation algorithms are performing effect analysis, they usually only do experimental comparison and analysis on a single fingerprint da...

AI Technical Summary

Problems solved by technology

At present, most automatic fingerprint identification systems are designed for a specific fingerprint collection device. When the system processes image data from different types of collection devices, the fingerprint images collected by these devices usually have different image quality, resolution and The gray level causes the performance of the system to decline to varying degrees when performing segmentation, enhancement, matching, etc.

[0005] The current fingerprint image segmentation technology also has device interoperability problems, which are specifically reflected in: (1) As far as the features selected during segmentation are concerned, for fingerprint images from multiple different types of acquisition devices, the eigenvalues ​​of the background and foreground appear significantly Intersection, the same feature value, which belongs to the background for one fingerprint image, may belong to the foreground for another image

(2) As far as the segmentation method is concerned, taking the supervised algorithm as an example, since most current algorithms design classifiers, the training samples and test samples come from images collected by the same device, resulting in the application of this classifier to fingerprints from other devices. A large segmentation error occurs when the image is segmented

As an important preprocessing step of automatic fingerprint identification system, the issue of device interoperability of fingerprint image segmentation method deserves enough attention, but as far as the inventors currently have knowledge, there is no open literature discussing such issues

Images