Fingerprint image segmentation method based on linear density calculation

A fingerprint image and line density technology, applied in image analysis, calculation, image enhancement, etc.

A fingerprint image and line density technology, applied in image analysis, calculation, image enhancement, etc.

CN106897989AInactive Publication Date: 2017-06-27MINNAN NORMAL UNIV
2 Cites 3 Cited by

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Fingerprint image segmentation method based on linear density calculation
  • Fingerprint image segmentation method based on linear density calculation
  • Fingerprint image segmentation method based on linear density calculation

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0051] see Figure 1 to Figure 1 3, a kind of fingerprint image segmentation method based on line density calculation of the present invention comprises the following steps:

[0052] A1. Separation of fingerprint image features to obtain a contour map;

[0053] A2. Find the starting point of the lines that may exist in the contour map;

[0054] A3. Starting from the starting point of each line, find a line in the contour map;

[0055] A4. Determine the fingerprint center on the line density map;

[0056] A5. Convex hull processing is performed on the area (ie mask) obtained after the fingerprint center is determined.

[0057] Preferably, said step A1 includes the following steps:

[0058] B1. Using fast cartoon texture decomposition, the image is decomposed into a cartoon part containing the background and a texture part containing details;

[0059] B2. Binarize the texture part of the image; convert a black-and-white or colored image into an image with only two colors of b...

Embodiment 2

[0082] The invention provides a fingerprint image segmentation method based on linear density calculation, comprising the following steps:

[0083] F1, use the fast cartoon texture decomposition, the figure 1 The raw fingerprint image shown is decomposed into figure 2 The textured part shown and image 3 The shown background cartoon part, wherein the value of parameter sigma in this embodiment is 3;

[0084] F2. For the previously decomposed texture image, first perform binarization processing, and the processed image is as follows Figure 4 As shown, the value of the binarization threshold V in this embodiment is 127. Then, the deburring process is performed to obtain the contour map, and the processed image is as follows Figure 5 shown;

[0085] F3. Determine the starting point of the line search, take the central pixel point of the continuous white pixel point area in the contour map as the starting point of the line, such as Image 6 The white pixels shown;

[008...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention discloses a fingerprint image segmentation method based on linear density calculation. The method comprises the following steps: the original fingerprint image is subjected to cartoon texture decomposition, a cartoon part containing a background and a texture part containing details are decomposed; the texture part is subjected to binaryzation and deburring treatment, and a contour map of the fingerprint image is acquired; lines are searched in the contour map; the image is divided to n*n pixel square grids, the number of points recorded in each grid is counted, and an image linear density is obtained; a fingerprint center is judged, the area of the fingerprint is obtained, that is, a mask area is obtained; and the mask area is subjected to convex hull treatment (or approximate convex hull treatment), and segmentation on the fingerprint foreground area and the background area of the fingerprint image is completed. On the basis of image cartoon texture decomposition, in view of a fingerprint image with each quality, a brand new method, that is, a linear density method is used for recognition and segmentation on the fingerprint area, and the segmentation efficiency and the segmentation accuracy of a low-quality fingerprint image can be effectively improved.

Description

technical field [0001] The invention designs a fingerprint image segmentation method based on linear density calculation, which belongs to the technical field of fingerprint identification. Background technique [0002] In the process of fingerprint extraction from the field to matching with the fingerprint database, fingerprint image segmentation is an extremely critical step. A good fingerprint image segmentation can remove other background noise while retaining the effective features of the fingerprint, and divide the image into two parts, the fingerprint area and the non-fingerprint area, so as to perform further optimization. At present, the segmentation algorithm for high-quality fingerprint images with simple background has been perfected day by day, but the segmentation of low-quality fingerprint images with complex background and many miscellaneous points basically needs to be done manually. Not only is this a waste of manpower, but the fingerprint area recognized ...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
27 Jun 2017
Publication
CN106897989A
IPC
G06T7/00; G06K9/00
CPC
G06T2207/20192; G06V40/12; G06V40/1347; G06T5/00
Inventors
εˆ˜δΉ¦η‚˜