CN109063681A - Direction of fingerprint information acquisition method based on fingerprint phase gradient - Google Patents
Direction of fingerprint information acquisition method based on fingerprint phase gradient Download PDFInfo
- Publication number
- CN109063681A CN109063681A CN201810983874.1A CN201810983874A CN109063681A CN 109063681 A CN109063681 A CN 109063681A CN 201810983874 A CN201810983874 A CN 201810983874A CN 109063681 A CN109063681 A CN 109063681A
- Authority
- CN
- China
- Prior art keywords
- fingerprint
- information
- signal
- phase
- single base
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/12—Fingerprints or palmprints
- G06V40/1347—Preprocessing; Feature extraction
- G06V40/1359—Extracting features related to ridge properties; Determining the fingerprint type, e.g. whorl or loop
Landscapes
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Human Computer Interaction (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Collating Specific Patterns (AREA)
Abstract
The present invention relates to fingerprint pretreatment technologies and direction of fingerprint technology for information acquisition, specially propose a kind of direction of fingerprint information acquisition method based on fingerprint phase gradient.The gradient information and direction of fingerprint information of fingerprint phase are orthogonal, and fingerprint phase information is monotonic increase in one cycle, therefore the present invention obtains reliable direction of fingerprint information from fingerprint phase gradient.Wherein, the phase information of fingerprint signal is the angle that real number signal rotates in single base signal complex space.Fingerprint phase information can be obtained by parsing single base signal, can remove fingerprint signal amplitude information at the same time.Wherein single base signal is by pretreated fingerprint signal and successively to be done line integral and function in the direction indicated to pretreated fingerprint signal
Description
Technical field:
The present invention relates to fingerprint pretreatment technologies and direction of fingerprint technology for information acquisition, propose a kind of based on fingerprint phase ladder
The direction of fingerprint information acquisition method of degree.
Background technique:
Fingerprint identification technology is the earliest biological identification technology of research, its main feature is that convenient, safe and low in cost, identification
Degree is high, therefore fingerprint recognition has become authentication mode the most commonly used in biological identification technology.Fingerprint recognition system is one
The modules such as a typical pattern recognition system, including fingerprint image acquisition, pretreatment, feature extraction and matching.Wherein, fingerprint is special
Sign extraction module is the important component of fingerprint recognition system, affects the degree of reliability of entire identifying system.Fingerprint characteristic
Extract the extraction comprising fingerprints important feature information such as direction of fingerprint, phase, frequency, singular point, minutiae points.Believe in the direction of fingerprint
Breath describes the direction mode of fingerprint ridge line, valley line, as the global reliable feature of fingerprint, Fingerprint enhancement, classification and
Matching etc. plays an important role.Therefore, direction of fingerprint acquisition of information is both the important step in fingerprint identification process,
It is the key link for influencing fingerprint recognition system performance.
The wide direction of fingerprint information extracting method of application is time domain gray grads method at present: when this method calculates first
The corresponding gradient of image and gray scale of each pixel obtains gradient vector field in the fingerprint image of domain, further according to fingerprint ridge direction and ladder
Degree direction mutually orthogonal relationship acquires direction of fingerprint information.Time domain gray grads method still has two sides in application process
The deficiency in face:
On the one hand, the gradient vector length acquired is bigger, and the direction that vector is directed toward is to finally determining neighborhood direction shadow
Sound is bigger, then wrong way pointed by the biggish gradient vector of length is necessarily to the accuracy in finally determining neighborhood direction
It has a negative impact；
On the other hand, in extremal region i.e. in fingerprint ridge intermediate region, since fingerprint grey scale change is slow, gradient
Direction is not often mutually perpendicular to fingerprint ridge direction or even random distribution, this will cause finally determining direction of fingerprint information
There is discontinuous expression, and increase subsequent processing link not only and will increase the time cost of whole system, can also further influence
System identification precision.
Summary of the invention:
The purpose of the present invention is to propose to a kind of direction of fingerprint information acquisition methods based on fingerprint phase gradient, can both solve
Direction of fingerprint information is discontinuous and influenced problem by amplitude information in crestal line intermediate region, and can reduce computation complexity, obtains
Obtain relatively reliable direction of fingerprint information.The present invention regards fingerprint as a two-dimentional amplitude-modulation frequency-modulation signal, since fingerprint signal exists
There are extreme point region in fingerprint ridge intermediate region, conventional method can not be handled in the region, so as to cause tradition
There is discontinuous expression in the region in the directional information that method obtains.The gradient information and direction of fingerprint information of fingerprint phase be
It is orthogonal, and fingerprint phase information is monotonic increase in one cycle, extreme point region is not present, therefore the present invention is not
Direction of fingerprint information is obtained by the gradient of fingerprint signal, but reliable direction of fingerprint letter is obtained from fingerprint phase gradient
Breath.Wherein, the phase information of fingerprint signal is the angle that real number signal rotates in single base vector signal complex space, i.e., real number is believed
Angle number between real axis.Fingerprint phase information can be obtained by parsing single base vector signal, obtain phase at the same time
Fingerprint signal amplitude information can be removed in the process.Wherein single base vector signal is by pretreated fingerprint signal and successively right
Pretreated fingerprint signal is done line integral and function in the direction indicatedTwo components obtained after convolution operation
Composition.
It is compared with the traditional method, on the one hand, the present invention obtains direction of fingerprint letter from the gradient information of fingerprint phase information
The operation of breath can directly eliminate the influence of the amplitude information in fingerprint signal, obtain relatively reliable direction of fingerprint information.Another party
Face, the discontinuous problem of directional information invention also avoids time domain gray grads method in fingerprint ridge intermediate region are reduced
Subsequent processing link.
Above-mentioned purpose is achieved through the following technical solutions:
A kind of direction of fingerprint information acquisition method based on fingerprint phase gradient.The method of the present invention includes the following steps:
(1) fingerprint image preprocessing: the present invention regards fingerprint as a two-dimentional amplitude-modulation frequency-modulation signal, accurately to be referred to
Line directional information should first eliminate noise and grey level compensation value in fingerprint signal.The present invention is filtered out using gauss low frequency filter
Noise, filter parameter are determined by fingerprint ridge mean breadth；Fingerprint grey level compensation value is calculated using mean filter and is incited somebody to action
It is filtered out.
(2) phase information obtains: the present invention first does line integral to fingerprint signal in the direction indicated, following and functionThen convolution obtains two components of single base vector signal in complex plane according to Fourier's Slice Theorem, and believes with fingerprint
Number constitute three-dimensional single base vector signal.Again by the property of single base signal it is found that during signal modulation, any complex signal exists
The value at certain moment can by real number signal complex number space inward turning transfer Lai.The complex plane projection of gained list base signal and horizontal seat
The angle of parameter is the local direction information of fingerprint signal, and the local phase information of fingerprint signal is real number signal in complex space
Angle between the angle of middle rotation, i.e. complex signal and real axis.Last this method carries out the list base signal under polar coordinates
It solves, solves fingerprint local direction information and phase information.
(3) reliable direction acquisition of information: the present invention obtains relatively reliable finger by calculating the gradient of fingerprint local phase
Line directional information, while the influence of amplitude information in fingerprint signal is eliminated, it avoids solving direction of fingerprint in time domain gray grads method
The discontinuous problem of information reduces subsequent processing link.
The utility model has the advantages that
1. conventional method thinks that direction of fingerprint information can only be obtained in the gradient of fingerprint grayscale image, but due to fingerprint ash
Spending image, there are extreme point information, unreliable so as to cause acquisition directional information；And the present invention has broken the think of of conventional method
Think, and obtain direction of fingerprint information in the gradient of fingerprint phase, and fingerprint phase information is monotonic increase within the period, no
There are extreme point information, therefore are continuously, not generate noise, obtained directional information in the direction that fingerprint ridge middle obtains
It is relatively reliable, without post-treatment operations are increased, to improve fingerprint recognition system efficiency.
2. the present invention eliminates fingerprint amplitude information, the direction of each pixel of acquisition while obtaining phase information
Information be to the Block direction informational influence of fingerprint it is identical, also prevent noise disperse.Under same scale, present invention acquisition side
It is more preferable to information noiseproof feature.
Specific embodiment:
1. fingerprint image preprocessing
The present invention regards fingerprint image as a two-dimentional FM amplitude modulation signal, i.e. fingerprint signal can indicate in two-dimensional space
Are as follows:
Wherein, a (x, y) is the grey level compensation value of fingerprint image, and b (x, y) is the amplitude of modulated signal,For fingerprint
Phase information, n (x, y) are the high-frequency noise in image.
To given fingerprint signal T (x, y), the height in signal is filtered out respectively using gauss low frequency filter and mean filter
Frequency noise n (x, y) and grey level compensation item a (x, y) obtains fingerprint signal only comprising amplitude and phase information:
2. phase information obtains
Line integral is done in the direction indicated to fingerprint signal to obtain:
R { f } (t, θ)=∫ f (x, y) δ (xcos θ+ysin θ-t) d (x, y) (3)
Wherein θ ∈ [0, π), i.e. direction of fingerprint information, t be with horizontal direction angle be θ straight line it is nearest to coordinate origin
Distance, δ is impulse function.
Fingerprint is done to the signal and function that line integral obtains in the direction indicatedConvolution then switches in frequency domain,
Law is sliced further according to Fourier to obtain:
Wherein, F1{ } (u, θ) be 2D signal direction θ ∈ [0, π) on one-dimensional Fourier transform.Pass through Fourier
Two components then can be obtained in inverse transformation
Two components of single base signal can be obtained
It can obtain
This step seeks the direction letter of fingerprint image according to the property of two components of single base signal complex field under polar coordinates
Breath:
According to single base signal each vector [f, fx,fy] between relationship, demodulate fingerprint local phase information.
3. reliable direction acquisition of information
Fingerprint local phase describes fingerprint ridge structural information, and direction of fingerprint letter is contained in the gradient information of phase
Breath.
Claims (4)
1. a kind of direction of fingerprint information acquisition method based on fingerprint phase gradient, which is characterized in that by fingerprint image preprocessing,
Phase information obtains and three step compositions of acquisition of reliable direction information.
2. a kind of direction of fingerprint information acquisition method based on fingerprint phase gradient according to claim 1, feature exist
In the fingerprint image preprocessing step specifically: the present invention regards fingerprint as a two-dimentional amplitude-modulation frequency-modulation signal, to obtain
Accurate direction of fingerprint information, should first eliminate the noise and grey level compensation value in fingerprint signal, and the present invention uses Gaussian low pass
Wave device filters out noise, and filter parameter is determined by fingerprint ridge mean breadth；Fingerprint gray scale is calculated using mean filter to mend
It repays value and is filtered out.
3. a kind of direction of fingerprint information acquisition method based on fingerprint phase gradient according to claim 1, feature exist
In the phase information obtaining step specifically: the present invention first does line integral to fingerprint signal in the direction indicated, connects down
Come with 1/ π t convolution of function, two components of single base vector signal in complex plane are then obtained according to Fourier's Slice Theorem,
And three-dimensional single base vector signal is constituted with fingerprint signal, then by the property of single base signal it is found that during signal modulation, arbitrarily
The value of complex signal at a certain moment can be transferred by real number signal in complex number space inward turning Lai the complex plane of gained list base signal is thrown
Shadow and the angle of horizontal axis are the local direction information of fingerprint signal, and the local phase information of fingerprint signal is real number letter
Number angle rotated in complex space, i.e. angle between complex signal and real axis, last this method is under polar coordinates to the list
Base signal is solved, and fingerprint local direction information and phase information are solved.
4. a kind of direction of fingerprint information acquisition method based on fingerprint phase gradient according to claim 1, feature exist
In the obtaining step of the reliable direction information specifically: the present invention is obtained more by calculating the gradient of fingerprint local phase
Add reliable direction of fingerprint information, while eliminating the influence of amplitude information in fingerprint signal, avoids asking in time domain gray grads method
The discontinuous problem of direction of fingerprint information is solved, subsequent processing link is reduced.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810983874.1A CN109063681A (en) | 2018-08-28 | 2018-08-28 | Direction of fingerprint information acquisition method based on fingerprint phase gradient |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810983874.1A CN109063681A (en) | 2018-08-28 | 2018-08-28 | Direction of fingerprint information acquisition method based on fingerprint phase gradient |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109063681A true CN109063681A (en) | 2018-12-21 |
Family
ID=64757249
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810983874.1A Pending CN109063681A (en) | 2018-08-28 | 2018-08-28 | Direction of fingerprint information acquisition method based on fingerprint phase gradient |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109063681A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109858418A (en) * | 2019-01-23 | 2019-06-07 | 上海思立微电子科技有限公司 | The treating method and apparatus of fingerprint image |
| CN110705436A (en) * | 2019-09-27 | 2020-01-17 | 哈尔滨理工大学 | Continuous fingerprint frequency field acquisition method |
| CN112084999A (en) * | 2020-09-22 | 2020-12-15 | 哈尔滨理工大学 | Method for acquiring fingerprint direction information through signal analysis |
| CN112329591A (en) * | 2020-10-30 | 2021-02-05 | 成都凯天电子股份有限公司 | Digital signal processing method for eliminating glitch interference signal |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101241545A (en) * | 2007-02-06 | 2008-08-13 | 西北工业大学 | Phase displacement fuzzy difference binaryzation method in fingerprint digital picture processing |
| CN106022348A (en) * | 2016-07-01 | 2016-10-12 | 昆明理工大学 | Finger retrieving method base on specific point direction field and fingerprint projection |
| US20170083742A1 (en) * | 2015-09-22 | 2017-03-23 | Safran Identity & Security | Method for extracting morphological characteristics from a sample of biological material |
| CN107463916A (en) * | 2017-08-16 | 2017-12-12 | 哈尔滨理工大学 | Fingerprint minutiae feature acquisition methods based on spiral operator |
| CN107506742A (en) * | 2017-09-05 | 2017-12-22 | 哈尔滨理工大学 | Direction of fingerprint characteristic-acquisition method based on fringe |
-
2018
- 2018-08-28 CN CN201810983874.1A patent/CN109063681A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101241545A (en) * | 2007-02-06 | 2008-08-13 | 西北工业大学 | Phase displacement fuzzy difference binaryzation method in fingerprint digital picture processing |
| US20170083742A1 (en) * | 2015-09-22 | 2017-03-23 | Safran Identity & Security | Method for extracting morphological characteristics from a sample of biological material |
| CN106022348A (en) * | 2016-07-01 | 2016-10-12 | 昆明理工大学 | Finger retrieving method base on specific point direction field and fingerprint projection |
| CN107463916A (en) * | 2017-08-16 | 2017-12-12 | 哈尔滨理工大学 | Fingerprint minutiae feature acquisition methods based on spiral operator |
| CN107506742A (en) * | 2017-09-05 | 2017-12-22 | 哈尔滨理工大学 | Direction of fingerprint characteristic-acquisition method based on fringe |
Non-Patent Citations (1)
| Title |
|---|
| 严汇莹: "一种免校准指纹相位特征提取算法的研究", 《中国优秀硕士学位论文全文数据库信息科技辑》 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109858418A (en) * | 2019-01-23 | 2019-06-07 | 上海思立微电子科技有限公司 | The treating method and apparatus of fingerprint image |
| CN110705436A (en) * | 2019-09-27 | 2020-01-17 | 哈尔滨理工大学 | Continuous fingerprint frequency field acquisition method |
| CN112084999A (en) * | 2020-09-22 | 2020-12-15 | 哈尔滨理工大学 | Method for acquiring fingerprint direction information through signal analysis |
| CN112329591A (en) * | 2020-10-30 | 2021-02-05 | 成都凯天电子股份有限公司 | Digital signal processing method for eliminating glitch interference signal |
| CN112329591B (en) * | 2020-10-30 | 2024-03-29 | 成都凯天电子股份有限公司 | Digital signal processing method for eliminating burr interference signals |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109063681A (en) | Direction of fingerprint information acquisition method based on fingerprint phase gradient | |
| CN106022315B (en) | A kind of pupil center's localization method for iris recognition | |
| CN107301661B (en) | High-resolution remote sensing image registration method based on edge point features | |
| Kim et al. | Efficient classification of ISAR images | |
| CN107563377A (en) | It is a kind of to detect localization method using the certificate key area of edge and character area | |
| CN107564006B (en) | Circular target detection method utilizing Hough transformation | |
| CN102661708B (en) | High-density packaged element positioning method based on speeded up robust features (SURFs) | |
| CN109919960A (en) | A kind of image continuous boundary detection method based on Multiscale Gabor Filters device | |
| Gao et al. | A novel method of fingerprint minutiae extraction based on Gabor phase | |
| CN107506742A (en) | Direction of fingerprint characteristic-acquisition method based on fringe | |
| CN110852992B (en) | Ship trail detection method based on top hat transformation and Radon transformation | |
| CN102254163A (en) | Template size self-adaptable Gabor fingerprint image enhancement method | |
| CN105225216A (en) | Based on the Iris preprocessing algorithm of space apart from circle mark rim detection | |
| CN103679721A (en) | Image contour simplifying method adopting nearest neighbor method Hough transform | |
| CN104091145A (en) | Human palm vein feature image acquisition method | |
| CN100342392C (en) | Coarse positioning method for remote sensing image based on Fourier-Mellin transformation | |
| Cui et al. | 2-D phase demodulation for deformable fingerprint registration | |
| CN107463916A (en) | Fingerprint minutiae feature acquisition methods based on spiral operator | |
| Wang et al. | An edge detection algorithm based on improved Canny operator | |
| CN113989308A (en) | Polygonal target segmentation method based on Hough transform and template matching | |
| KR20050094228A (en) | Fingerprint recognition method | |
| Lin | A novel iris recognition method based on the natural-open eyes | |
| CN110874599A (en) | Ship detection method based on image recognition | |
| Ouyang et al. | Fingerprint matching with rotation-descriptor texture features | |
| CN103177443A (en) | SAR (synthetic aperture radar) target attitude angle estimation method based on randomized hough transformations |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20181221 |
|
| WD01 | Invention patent application deemed withdrawn after publication |