CN101303728A

CN101303728A - Method for identifying fingerprint facing image quality

Info

Publication number: CN101303728A
Application number: CNA2008101381170A
Authority: CN
Inventors: 尹义龙; 杨公平; 骆功庆; 张宇; 詹小四; 任春晓
Original assignee: Shandong University
Current assignee: Shandong University
Priority date: 2008-07-01
Filing date: 2008-07-01
Publication date: 2008-11-12
Anticipated expiration: 2028-07-01
Also published as: CN100592323C

Abstract

The invention discloses a fingerprint identifying method facing to the quality of an image, which solves the defect that the compatibility of the existing fingerprint identifying method to an ideal fingerprint image and a non ideal fingerprint image is not good. The method of the invention includes: (1) reading the collected fingerprint image g(x, y), wherein, g (x, y) is the gray value of the pixel points (x, y); (2) carrying out quality characteristic extracting on the fingerprint images and respectively extracting three characteristics of grads consistency QT, the frequency spectrum characteristic QF and the gray standard difference Qs; (3) adopting an SVM sorter to carry out learning and sorting on the qualities of the fingerprint images and confirming the qualities of the fingerprint images as two defined quality types; (4) respectively adopting a matching arithmetic based on detail points and a matching arithmetic based on textures to the two fingerprints with better and poorer qualities to accomplish the identification.

Description

Fingerprint identification method facing image quality

Technical Field

The invention relates to a fingerprint identification method, in particular to a fingerprint identification method facing to image quality.

Background

In the automatic fingerprint identification technology, the quality of an image is an important factor influencing the identification performance. In the existing fingerprint identification processing method, the minutiae of the fingerprint image are usually acquired for matching, but the performance of the method is obviously reduced for the image with poor quality. While other recognition methods, such as the recognition method based on the lines and the recognition method based on the textures, have a certain effect on the image with poor quality, but for the image with good quality, not only the performance is not greatly improved, but also the occupied resources and the time complexity are relatively large. Therefore, a fingerprint identification method for distinguishing image quality is urgently needed, which not only ensures that a certain identification accuracy is achieved, but also can save resources to a greater extent.

Disclosure of Invention

The invention aims to provide a fingerprint identification method facing image quality, aiming at solving the defect that the prior fingerprint identification method has poor compatibility with ideal and non-ideal fingerprint images. The method is a processing method based on image quality judgment, and divides fingerprint images into two types of better quality and poorer quality, and further adopts different identification algorithms to carry out fingerprint identification.

In order to achieve the purpose, the invention adopts the following technical scheme:

(1) reading a collected fingerprint image g (x, y), wherein g (x, y) is the gray value of a pixel point (x, y);

(2) extracting the quality characteristics of the fingerprint image, and respectively extracting the gradient consistency Q_TSpectral characteristic Q_FStandard deviation of gray scale Q_sThree features in total;

(3) learning and classifying the quality of the fingerprint image by adopting an SVM (support vector machine) classifier, and determining the quality as two defined quality types;

(4) and for the two fingerprints with better quality and poorer quality, respectively adopting a minutiae-based matching algorithm and a texture-based matching algorithm to finish the identification.

In the step (2), the three characteristics respectively reflect the quality from different aspects, and the specific calculation is as follows,

r is the total number of foreground blocks,

the gradient consistency of one block in the block image is calculated by the following formula:

\tilde{k} = \frac{{(j_{11} - j_{22})}^{2} + 4 j_{12}^{2}}{{(j_{11} + j_{22})}^{2}},

wherein j is₁₁，j₁₂，j₂₁，j₂₂Are elements in the gradient vector covariance matrix J. If the image block size is b x b, all b in the block²Gradient vector covariance matrix of points

<math> <mrow> <mi>J</mi> <mo>=</mo> <mfrac> <mn>1</mn> <msup> <mi>b</mi> <mn>2</mn> </msup> </mfrac> <munder> <mi>Σ</mi> <mrow> <mi>s</mi> <mo>&Element;</mo> <mi>B</mi> </mrow> </munder> <msub> <mi>g</mi> <mi>s</mi> </msub> <msubsup> <mi>g</mi> <mi>s</mi> <mi>T</mi> </msubsup> <mo>&equiv;</mo> <mfenced open='[' close=']'> <mtable> <mtr> <mtd> <msub> <mi>j</mi> <mn>11</mn> </msub> </mtd> <mtd> <msub> <mi>j</mi> <mn>12</mn> </msub> </mtd> </mtr> <mtr> <mtd> <msub> <mi>j</mi> <mn>21</mn> </msub> </mtd> <mtd> <msub> <mi>j</mi> <mn>22</mn> </msub> </mtd> </mtr> </mtable> </mfenced> <mo>,</mo> </mrow> </math>

b²Is the size of the block image, s is the point in the block, B is the set of all the pixel points in the block, g_sIs the gradient vector of point s, g_s ^TIs a transpose of the gradient vector; the final quality characteristic of the whole image is the mean of the gradient consistency of all blocks.

Q_FThe calculation formula of (2) is as follows:

wherein,

<math> <mrow> <mi>Q</mi> <mrow> <mo>(</mo> <mi>r</mi> <mo>)</mo> </mrow> <mo>=</mo> <mfrac> <mn>1</mn> <mrow> <mo>#</mo> <msub> <mi>C</mi> <mi>r</mi> </msub> </mrow> </mfrac> <munder> <mi>Σ</mi> <mrow> <mrow> <mo>(</mo> <mi>u</mi> <mo>,</mo> <mi>v</mi> <mo>)</mo> </mrow> <mo>&Element;</mo> <msub> <mi>C</mi> <mi>r</mi> </msub> </mrow> </munder> <mo>|</mo> <msub> <mi>G</mi> <mrow> <mo>(</mo> <mi>u</mi> <mo>,</mo> <mi>v</mi> <mo>)</mo> </mrow> </msub> <mo>|</mo> </mrow> </math>

as a function of the intensity of the energy,

is a ring (r)₀-4＜＝r＜＝r₀+4) number of pixels within, and

|G_(u，v)i reflects the energy intensity at the point (u, v) in the frequency domain image after Fourier transform, | G_(u，v)L constitutes the intensity spectrum of the frequency domain. Let g (x, y) denote the gray value of a pixel point with coordinates (x, y) in a digital image with size NxN, the distance of g (x, y)The scattered Fourier transform (DFT) G (u, v) is defined as,

Q_Fthe quality is represented by calculating the energy of the bright ring zone in the spectrum image and using the size of the energy.

Q_sThe calculation formula of (2) is as follows:

wherein,

<math> <mrow> <msub> <mi>S</mi> <mi>k</mi> </msub> <mo>=</mo> <msqrt> <mfrac> <mn>1</mn> <msup> <mover> <mi>w</mi> <mo>&OverBar;</mo> </mover> <mn>2</mn> </msup> </mfrac> <munderover> <mi>Σ</mi> <mrow> <mi>x</mi> <mo>=</mo> <mn>1</mn> </mrow> <mover> <mi>w</mi> <mo>&OverBar;</mo> </mover> </munderover> <munderover> <mi>Σ</mi> <mrow> <mi>y</mi> <mo>=</mo> <mn>1</mn> </mrow> <mover> <mi>w</mi> <mo>&OverBar;</mo> </mover> </munderover> <msup> <mrow> <mo>(</mo> <mi>g</mi> <mrow> <mo>(</mo> <mi>x</mi> <mo>,</mo> <mi>y</mi> <mo>)</mo> </mrow> <mo>-</mo> <mi>g</mi> <mrow> <mo>(</mo> <mi>k</mi> <mo>)</mo> </mrow> <mo>)</mo> </mrow> <mn>2</mn> </msup> </msqrt> </mrow> </math>

the standard deviation of the k block in the block image is shown, g (x, y) is the gray value of the pixel point (x, y), g (k) is the gray average value of the k block, and w is the block side length of the block image. Q_sThe quality of the whole image is represented by averaging the standard deviations of all the block images.

In the step (4), both the minutiae-based matching algorithm and the texture-based matching algorithm are classic algorithms.

The block size of the block image in the processing method of the invention is 8 multiplied by 8.

The invention has the beneficial effects that: the processing method integrates the characteristics of a plurality of aspects such as a fingerprint image space domain, a frequency domain and the like, so that the quality of the fingerprint image can be well distinguished, and the matching algorithm based on the minutiae and the matching algorithm based on the texture have better adaptability.

Drawings

FIG. 1 is a flow chart of the identification method of the present invention.

Detailed Description

The invention is further described with reference to the following figures and examples.

In fig. 1, a fingerprint identification method for image quality includes the steps of:

(3) forming a three-dimensional characteristic vector by the three characteristic indexes extracted in the step (2) and using the three characteristic indexes as input vectors of a support vector machine; learning and classifying the quality of the fingerprint image by adopting an SVM (support vector machine) classifier, and determining the quality as two defined quality types;

r is the total number of foreground blocks (where r is 40),

\tilde{k} = \frac{{(j_{11} - j_{22})}^{2} + 4 j_{12}^{2}}{{(j_{11} + j_{22})}^{2}},

wherein j is₁₁，j₁₂，j₂₁，j₂₂Are elements in the gradient vector covariance matrix J. Gradient vector covariance matrix of all points in a block

Where b is²64, i.e. the size of the block image, s is the point in the block, B is the set of all pixel points in the block, g_sIs the gradient vector of point s, g_s ^TIs a transpose of the gradient vector; the final quality characteristic of the whole image is the mean of the gradient consistency of all blocks.

Q_FThe calculation formula of (2) is as follows:

wherein,

as a function of the intensity of the energy,is the number of pixel points within the ring (r0-4 r0+4), and

|G_(u，v)i reflects the energy intensity at the point (u, v) in the frequency domain image after Fourier transform, | G_(u，v)L constitutes the intensity spectrum of the frequency domain. Let G (x, y) denote the gray value of a pixel point with coordinates (x, y) in a digital image with size NxN, then the Discrete Fourier Transform (DFT) G (u, v) of G (x, y) is defined as,

Q_sThe calculation formula of (2) is as follows:

wherein,

the standard deviation of the k-th block in the block image is shown, g (x, y) is the gray value of the pixel point (x, y), g (k) is the gray average value of the k-th block, and w is the block edge length of the block image (where w is 8). Q_sRepresenting the quality of the entire image by averaging the standard deviations of all the block imagesAmount of the compound (A).

Claims

1. A fingerprint identification method facing image quality is characterized in that the method comprises the following steps:

(1) reading an acquired fingerprint image g (x, y), wherein g (x, y) is the gray value of a pixel point (x, y);

(3) learning and classifying the quality of the fingerprint image by adopting an SVM (support vector machine) classifier, and determining the quality of the fingerprint image into two quality types of better quality or poorer quality;

2. The image quality-oriented fingerprint recognition method according to claim 1, wherein in said step (2), three features respectively reflect the quality from different aspects, and are calculated as follows,

r is the total number of foreground blocks,

\tilde{k} = \frac{{(j_{11} - j_{22})}^{2} + 4 j_{12}^{2}}{{(j_{11} + j_{22})}^{2}},

wherein j is₁₁，j₁₂，j₂₁，j₂₂Is an element in the gradient vector covariance matrix J;

if the image block size is b x b, all b in the block²Gradient vector covariance matrix of points

Wherein b is²Is the size of the block image, s is the point in the block, B is the set of all the pixel points in the block, g_sIs the gradient vector of point s, g_s ^TIs a transpose of the gradient vector; the quality of the final whole image is characterized by the consistent gradient of all blocksThe mean value of sex;

Q_Fthe calculation formula of (2) is as follows:

wherein,

as a function of the intensity of the energy,

is a ring (r)₀-4＜＝r＜＝r₀+4) number of pixels within, and

|G_(u，v)i reflects the energy intensity at the point (u, v) in the frequency domain image after Fourier transform, | G_(u，v)L constitutes the intensity spectrum of the frequency domain; let G (x, y) denote the gray value of a pixel point with coordinates (x, y) in a digital image with size NxN, the discrete Fourier transform G (u, v) of G (x, y) is defined as,

Q_Fcalculating the energy of a bright ring band in the frequency spectrum image, and using the size of the energy to represent the quality;

Q_sthe calculation formula of (2) is as follows:

wherein,

the standard deviation of the k block in the block image is shown, g (x, y) is the gray value of a pixel point (x, y), g (k) is the gray average value of the k block, and w is the block edge length of the block image; q_sThe quality of the whole image is represented by averaging the standard deviations of all the block images.

3. The image quality-oriented fingerprint recognition method according to claim 1, wherein in the step (4), the minutiae-based matching algorithm and the texture-based matching algorithm are both classical algorithms.