TWI416432B - Method of using singular value decompostion for enhancing and segmenting fingerprint images and a system thereof - Google Patents

Abstract

A method of using singular value decomposition for fingerprint images includes: decomposing an original image in a singular value decomposition manner to obtain an enhanced image; transforming energy of the enhance image to obtain an energy distribution; searching a fingerprint boundary by the energy distribution to obtain a plurality of landmarks to surround a segment boundary polygon. The enhancing and segmenting fingerprint image system includes an input unit, a calculating unit and an output unit. The input unit is used to input the original image, the calculating unit is used to generate the landmarks and segment boundary polygon, and the output unit is used to output a segmented fingerprint image according to the segment boundary polygon.

Description

Method and system for enhancing and cutting fingerprint image by using singular value decomposition

The invention relates to a method and a system for enhancing and cutting a fingerprint image by using a singular value decomposition; in particular, the method for eliminating background noise by using a Singular Value Decomposition (SVD) method; Enhancement and cutting methods and systems thereof.

In general, according to the fingerprint classification method of British fingerprint expert ER Henry, the fingerprint is classified into right-handed pattern [right loop], left-hand hoove-like pattern [left loop], tented arch-like pattern [tented arch], arched pattern [plain arch or arch] and vortex pattern [whorl, spiral pattern]. In addition, eddy patterns and s-shaped lines or double-ring patterns [S-type or twin loop] are also classified as spiral patterns.

A conventional fingerprint recognition device, for example, a total reflection type fingerprint identification device of the Republic of China Patent Publication No. 200641700, which discloses a total reflection type fingerprint identification device, which includes a body, a light source, a joint body, and A sensor. The body has a touch zone, a first reflection zone and a light transmission zone. The light generated by the light source is illuminated by the light-transmitting area of the body on the finger placed on the touch area. The joint body has another second reflection area, and the sensor is disposed on the joint body, and the aperture of the body and the joint body is provided with an aperture, and the aperture is a black or black opaque hole-like film layer, allowing the passage The set light passes through the aperture and filters other stray light, and is imaged on the sensor, so that the fingerprint identification device can achieve high contrast and low noise when capturing fingerprint images.

Another conventional fingerprint image classification method and device, for example, the method and system device for automatic classification of fingerprint images in the Republic of China Patent Publication No. 354397, and a method for discovering a fingerprint image, and a system device according to the fingerprint The number of core points and the fingerprint line around the core point flow to [ridge flow direction], and the input fingerprints are divided into eight different categories. The pre-processing of the input original fingerprint image is first performed. The background portion and the fingerprint portion of the fingerprint image are separated according to the grayscale mean value [mean] and the variation value [variance] in the fingerprint image. All subsequent processing steps are only for the fingerprint portion, which can improve the speed of this classification method. After the fingerprint portion is separated, the average direction of the fingerprint line of the block is calculated for each fingerprint block, and a block directional image is generated. The location of the core points in the fingerprint is then derived for the distribution of the entire fingerprint image block pattern. Finally, the number of core points and the flow of fingerprint lines around the core points are used. According to the classification rules, the input fingerprint images are divided into eight categories.

Although the aforementioned patents have disclosed related fingerprint image recognition techniques, they do not provide related techniques for how to eliminate background noise. In fact, in terms of fingerprint image recognition technology, it is necessary to eliminate considerable background noise, otherwise it will affect the reliability of fingerprint image recognition. Therefore, the conventional fingerprint image recognition technology must have the need to further eliminate background noise.

The above-mentioned patents of the Japanese Patent Publication No. 200641700 and the publication No. 354397 are only for reference to the technical background of the present invention and the state of the art is not limited to the scope of the present invention.

In view of the above, the present invention provides a method and system for enhancing and cutting a fingerprint image by using a singular value decomposition method, which uses a singular value decomposition method to decompose an original fingerprint image to obtain a gain fingerprint image. To achieve the purpose of accurately cutting fingerprint images.

The main object of the present invention is to provide a method and system for enhancing and cutting a fingerprint image by using a singular value decomposition method, which uses a singular value decomposition method to decompose an original fingerprint image to obtain a gain fingerprint image to achieve accurate cutting of the fingerprint image. The purpose.

In order to achieve the above object, the method for enhancing and cutting a fingerprint image by using the singular value decomposition method comprises: decomposing an original fingerprint image by using a singular value decomposition method to obtain a gain fingerprint image; and performing energy conversion on the gain fingerprint image. Obtaining an energy profile; and using the energy profile to find a fingerprint profile to obtain a plurality of demarcations around which a cut polygon is formed.

The enhancement and cutting system using the singular value decomposition in the fingerprint image of the present invention comprises: an input unit for inputting an original fingerprint image; and an calculus unit for decomposing an original fingerprint image by using a singular value decomposition method to obtain a a gain fingerprint image; the calculation unit performs energy conversion on the gain fingerprint image to obtain an energy distribution map; the calculation unit uses the energy distribution map to search for a fingerprint contour to obtain a plurality of demarcations, which form a cut polygon; An output unit is configured to output a cut fingerprint boundary image according to the cut polygon.

In the preferred embodiment of the present invention, when the fingerprint profile is sought, the demarcation is obtained by horizontal projection.

In the preferred embodiment of the present invention, when the fingerprint profile is sought, the demarcation is obtained by the left and right horizontal projection methods.

A preferred embodiment of the present invention utilizes the demarcation to cut a block of interest fingerprints.

In order to fully understand the present invention, the preferred embodiments of the present invention are described in detail below and are not intended to limit the invention.

The method and system for enhancing and cutting a fingerprint image using singular value decomposition according to a preferred embodiment of the present invention are applicable to various fingerprint identification devices and related application devices, such as various types of computer systems or security systems, but are not limited thereto. The scope of the invention.

FIG. 1 is a schematic diagram showing an enhancement and cutting method using a singular value decomposition to a fingerprint image and a system of the original fingerprint image in an unprocessed embodiment of the preferred embodiment of the present invention. Referring to FIG. 1 , an enhancement and cutting system using a singular value decomposition to a fingerprint image according to a preferred embodiment of the present invention includes an input unit, a calculation unit, and an output unit. The method for enhancing and cutting a fingerprint image using the singular value decomposition is performed on the input unit for inputting at least one original fingerprint image, as shown in FIG.

FIG. 2 is a schematic diagram showing a method for enhancing and cutting a fingerprint image by using a singular value decomposition according to a preferred embodiment of the present invention, and a system for obtaining a gain fingerprint image by performing singular value decomposition processing on the original fingerprint image. The original fingerprint image shown in Figure 1. Referring to FIG. 2, the method for enhancing and cutting a fingerprint image using the singular value decomposition is performed in the calculation unit, which uses a singular value decomposition method to decompose the original fingerprint image to obtain a gain fingerprint image. As shown in Figure 2.

In the preferred embodiment of the present invention, the singular value decomposition is applied to the fingerprint image enhancement and cutting method using the singular value decomposition of the m × n matrix A ( m n ) for,

A = U Σ V ^t

Where U = [ u ₁ , u ₂ , .., u _m ] and V = [ v ₁ , v ₂ ,..., v _n ] are all orthogonal matrices, Σ = [ D , 0 ] ^t contains a singular value on the diagonal; D = diag ( λ ₁ , λ ₂ ,..., λ _k ) which has a singular value λ _{i , i =1,...} in a non-increasing [non-increasing] sequence _{... , k} , 0 is a zero matrix; matrix Σ is the intensity information indicating the fingerprint image [intensity information].

The normalized image of the fingerprint image is a probability density function (PDF) with a mean and a variation of the Gaussian distribution. Decompose the m × n matrix A into

The k < r positive integer taken from the above equation is reduced to

Expand the above equation to

Where k is the reserved singular value, and rank (Σ) = min ( m , n ), λ _i u _i v _i ^t contains the vector outer product. The Frobenius norm of matrix A is [Frobenius norm],

The energy of the matrix A from the above equation can be segmented by the Frobenius norm.

In order to eliminate background noise, the maximum singular value of the normal matrix is set in the following manner.

Where Σ _G is the mean and variation of the Gaussian image for each database in Table 1.

Table 1: Mean and variation values of Gaussian distribution functions for each database.

The equation of the synthetic intensity image produced by the present invention is

A _eq = U (αΣ _G ) V ^t

A synthetic intensity image having the original fingerprint image size is obtained from the above equation, and the background noise of the original fingerprint image is eliminated.

FIG. 3 is a schematic diagram showing an energy distribution of a fingerprint image obtained by performing energy conversion processing on a gain fingerprint image by using a singular value decomposition on a fingerprint image enhancement and cutting method according to a preferred embodiment of the present invention, which corresponds to the first and 2 The original fingerprint image and the gain fingerprint image shown in FIG. Referring to FIG. 3, the enhancement and cutting method using the singular value decomposition in the fingerprint image of the present invention is performed by performing the energy conversion processing on the calculation unit to obtain an energy distribution map. The definition of the local standard average μ and the conversion energy ε of 7×7 pixels is expressed as

Where N = 49 is the normalization constant.

In order to convert the grayscale intensity image, the threshold image of the matrix A is

The pixel value 255 is the object of interest [interest object], and the pixel value 0 is the block of no interest.

Referring again to Figures 1 and 3, the energy distribution fingerprint image of Figure 3 has better contrast than the original fingerprint image of Figure 1.

FIG. 4 is a schematic diagram showing the energy distribution projection of the energy distribution map of the fingerprint image from the left and right sides by using the singular value decomposition on the fingerprint image enhancement and cutting method and the system thereof according to a preferred embodiment of the present invention; The image on the lower left of Figure 3 is shown by the arrow in Figure 3. Referring to FIG. 4, the method for enhancing and cutting a fingerprint image using the singular value decomposition of the present invention is performed by performing the dicing process. The calculation unit uses the energy distribution map to find a fingerprint profile to obtain a plurality of landmarks [landmark], which surrounds forming a cut polygon [segmented polygon].

Referring to FIG. 4 again, in the preferred embodiment of the present invention, when the fingerprint contour is sought, the boundary is obtained by horizontal projection. In addition, when looking for the fingerprint outline, the boundary is detected by the left and right horizontal projection methods, as shown on the left side of FIG. 4, and the horizontal boundary is used to obtain the boundary, as shown on the right side of FIG.

FIG. 5 is a schematic diagram showing a method for enhancing and cutting a fingerprint image by using a singular value decomposition and a system thereof to form a fingerprint block of interest in fingerprint image cutting according to a preferred embodiment of the present invention. Referring to FIG. 5, the method for enhancing and cutting a fingerprint image using the singular value decomposition is performed by the output unit to output an image according to the cut polygon, so as to output a cut fingerprint boundary image, as shown in FIG. 4 The polygon is shown.

Referring again to FIG. 5, a preferred embodiment of the present invention utilizes the demarcation to cut anImpression of Interest (IOI).

The foregoing preferred embodiments are merely illustrative of the invention and the technical features thereof, and the techniques of the embodiments can be carried out with various substantial equivalent modifications and/or alternatives; therefore, the scope of the invention is subject to the appended claims. The scope defined by the scope shall prevail.

[No component symbol]

FIG. 1 is a schematic diagram of an original fingerprint image of a method for enhancing and cutting a fingerprint image by using a singular value decomposition and a system thereof according to a preferred embodiment of the present invention.

FIG. 2 is a schematic diagram of a method for enhancing and cutting a fingerprint image by using a singular value decomposition according to a preferred embodiment of the present invention, and a system for obtaining a gain fingerprint image after the original fingerprint image is subjected to singular value decomposition processing.

FIG. 3 is a schematic diagram of an energy distribution of a fingerprint image obtained by performing energy conversion processing on a gain fingerprint image by using a singular value decomposition method and a system for enhancing and cutting a fingerprint image according to a preferred embodiment of the present invention.

FIG. 4 is a schematic diagram showing the energy distribution projection of the energy distribution map of the fingerprint image from the left and right sides by using the singular value decomposition to enhance and cut the fingerprint image according to the preferred embodiment of the present invention.

FIG. 5 is a schematic diagram of a method for enhancing and cutting a fingerprint image by using a singular value decomposition according to a preferred embodiment of the present invention and a system for forming a fingerprint block of interest by fingerprint image cutting.

[No component symbol]

Claims (8)

A method for enhancing and cutting a fingerprint image by using a singular value decomposition method comprises: decomposing an original fingerprint image by using a singular value decomposition method to obtain a gain fingerprint image; and performing energy conversion on the gain fingerprint image to obtain an energy distribution. And using the energy profile to find a fingerprint profile to obtain a plurality of demarcations around which a cut polygon is formed. According to the enhancement and cutting method of the fingerprint image, which is decomposed by the singular value according to the first item of the patent application scope, when the fingerprint contour is sought, the boundary is obtained by horizontal projection. According to the enhancement and cutting method of the fingerprint image, which is decomposed by the singular value according to the first item of the patent application scope, when the fingerprint contour is sought, the boundary is obtained by the left and right horizontal projection methods. The method for enhancing and cutting a fingerprint image by using a singular value according to the first aspect of the patent application, wherein the target fingerprint block is cut by the boundary. An enhancement and cutting system for singular value decomposition into a fingerprint image, comprising: an input unit for inputting an original fingerprint image; and an calculus unit for decomposing an original fingerprint image by using a singular value decomposition method to obtain a a gain fingerprint image; the calculation unit performs energy conversion on the gain fingerprint image to obtain an energy distribution map, and the calculation unit uses the energy distribution map to search for a fingerprint contour to obtain a plurality of demarcations, which form a cut polygon; And an output unit for outputting a cut fingerprint boundary image according to the cut polygon. According to the enhancement and cutting system of the fingerprint image which is decomposed by the singular value according to the fifth item of the patent application scope, when the fingerprint contour is sought, the demarcation boundary is obtained by the horizontal projection method. According to the enhancement and cutting system of the fingerprint image which is decomposed by the singular value according to the fifth item of the patent application scope, when the fingerprint contour is sought, the boundary is obtained by the left and right horizontal projection methods. An enhancement and cutting system for decomposing a fingerprint image using the singular value according to item 5 of the patent application scope, wherein the target fingerprint block is cut by the boundary.