CN106960191A - A kind of fingerprint recognition system - Google Patents
A kind of fingerprint recognition system Download PDFInfo
- Publication number
- CN106960191A CN106960191A CN201710180015.4A CN201710180015A CN106960191A CN 106960191 A CN106960191 A CN 106960191A CN 201710180015 A CN201710180015 A CN 201710180015A CN 106960191 A CN106960191 A CN 106960191A
- Authority
- CN
- China
- Prior art keywords
- fingerprint
- initial
- component
- fingerprint image
- image
- 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.)
- Withdrawn
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
-
- 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
- G06V40/1365—Matching; Classification
Abstract
The invention provides a kind of fingerprint recognition system, including finger print acquisition module, fingerprint image processing module, Finger print characteristic abstract module, checking identification module, personnel's fingerprint database and result display module, personnel's fingerprint database is used for the personnel's fingerprint image for storing standard；The finger print acquisition module is used to gather initial fingerprint image；The fingerprint image processing module is used to the initial fingerprint image collected carrying out a High-resolution Processing, obtains high-resolution fingerprint image；The Finger print characteristic abstract module is used to extract the fingerprint characteristic image in high-resolution fingerprint image；The checking identification module is used to carry out matching checking with personnel's fingerprint image of the standard in personnel's fingerprint database by fingerprint characteristic image；The result display module is used to receive the result and be shown.Beneficial effects of the present invention are：High-resolution Processing is carried out to the fingerprint image collected, effectively take the fingerprint characteristic information, improve the accuracy of fingerprint authentication.
Description
Technical field
The present invention relates to fingerprint recognition field, and in particular to a kind of fingerprint recognition system.
Background technology
Fingerprint recognition system in correlation technique is transmitted to fingerprint image using realtime graphic, and fingerprint image is not carried out
High-resolution Processing, the initial fingerprint characteristic fingerprint collected is not obvious, often leads to that target fingerprint can not be carried out accurately
Checking identification.
High-resolution image can provide abundant detailed information, but in actual environment due to existing apart from limited,
The problems such as environmental disturbances, high-resolution image is often difficult to obtain, and is restricted by factors such as technique, cost and environment, allows
High-resolution image is more difficult to extensive acquisition.
The image being typically observed is made up of a variety of different types of basic information sources or composition, each class information source or composition tool
There are different functions.In recent years, Starcket et al. is different in nature and openness according to poor morphology, it is proposed that form PCA
(Morphological Component Analysis, MCA).Because it can effectively solve have different shape special in complicated image
The resolution problem of content is levied, turns into the main stream approach of picture breakdown at present.
The content of usual natural image may be considered to be made up of heterogeneity, and per the element of the first species with unique form
Learn feature, such as common smooth component and texture component, the large-scale structure feature in smooth representation in components image, and line
Manage detailed information in representation in components image.
At present, multiresolution analysis method is the most commonly used character representation method in image super-resolution field, no
Same multiresolution analysis method is suitable for extracting different characteristic in image respectively, and Stationary Wavelet Transform (SWT) is used to represent to scheme
As point-like character, non-down sampling contourlet transform (NSCT) is used for the line and contour feature for representing image.Effectively combine steady small
Wave conversion, non-down sampling contourlet transform complementary and smooth component, the different shape feature of texture component, can rationally be designed
Go out ultra-resolution method, there is provided abundant image detail information by the definition for greatly improving image.
The content of the invention
The purpose of the present invention be overcome in the prior art fingerprint accurately recognize difficult problem, the present invention is intended to provide a kind of essence
True fingerprint recognition system.
To achieve these goals, the present invention provides a kind of fingerprint recognition system, and the fingerprint recognition system includes：Fingerprint is adopted
Collect module, fingerprint image processing module, Finger print characteristic abstract module, checking identification module, personnel's fingerprint database and result aobvious
Show module, personnel's fingerprint database is used for the personnel's fingerprint image for storing standard；The finger print acquisition module is used to gather
Initial fingerprint image；The fingerprint image processing module is connected to the finger print acquisition module, for initial by what is collected
Fingerprint image carries out a High-resolution Processing, obtains high-resolution fingerprint image；The Finger print characteristic abstract module, which is connected to, to be tested
Identification module is demonstrate,proved, for extracting the fingerprint characteristic image in high-resolution fingerprint image, and by fingerprint characteristic image transmitting to institute
State checking identification module；The checking identification module is connected to Finger print characteristic abstract module and personnel's fingerprint database, for inciting somebody to action
Fingerprint characteristic image carries out matching checking with personnel's fingerprint image of personnel's fingerprint database Plays；The result display module
The checking identification module is connected to, for receiving the result and being shown.
Beneficial effects of the present invention are：High-resolution Processing is carried out to the fingerprint image collected, effectively taken the fingerprint
Characteristic information, improves the accuracy of fingerprint authentication.
Brief description of the drawings
Using accompanying drawing, the invention will be further described, but the embodiment in accompanying drawing does not constitute any limit to the present invention
System, for one of ordinary skill in the art, on the premise of not paying creative work, can also be obtained according to the following drawings
Other accompanying drawings.
The frame construction drawing of Fig. 1 present invention；
Fig. 2 is the frame construction drawing of fingerprint image processing module of the present invention.
Reference：
Finger print acquisition module 1, fingerprint image processing module 2, Finger print characteristic abstract module 3, checking identification module 4, personnel
Fingerprint database 5, result display module 6, fingerprint image component submodule 20, component image processing submodule 21, fingerprint image
Synthesize submodule 22.
Embodiment
With reference to following application scenarios, the invention will be further described.
Referring to Fig. 1, Fig. 2, a kind of fingerprint recognition system of the present embodiment, including the processing of finger print acquisition module 1, fingerprint image
Module 2, Finger print characteristic abstract module 3, checking identification module 4, personnel's fingerprint database 5 and result display module 6, the personnel
Fingerprint database 5 is used for the personnel's fingerprint image for storing standard；The finger print acquisition module 1 is used to gather initial fingerprint image；
The fingerprint image processing module 2 is connected to the finger print acquisition module 1, for the initial fingerprint image collected to be carried out
High-resolution Processing, obtains high-resolution fingerprint image；The Finger print characteristic abstract module 3 is connected to checking identification module 2, uses
In extracting the fingerprint characteristic image in high-resolution fingerprint image, and by fingerprint characteristic image transmitting to the checking identification module
4；The checking identification module 4 is connected to Finger print characteristic abstract module 3 and personnel's fingerprint database 5, for by fingerprint characteristic figure
As carrying out matching checking with personnel's fingerprint image of personnel's fingerprint database Plays；The result display module 6 is connected to institute
Checking identification module is stated, for receiving the result and being shown.
Preferably, gathered during the collection of finger print acquisition module 1 initial fingerprint image using common camera, camera
Focus alignment finger pressing lens centre, post anti-fingerprint pad pasting on the outside of lens.
Preferably, the checking identification module 4 refers to the personnel of fingerprint characteristic image and the Plays of personnel's fingerprint database 5
Print image carries out contrast verification, personnel's fingerprint image phase that can be with personnel's fingerprint database Plays in fingerprint characteristic image
Timing, is shown by display screen and is proved to be successful, in fingerprint characteristic image and personnel's fingerprint image of personnel's fingerprint database Plays
As when can not match, authentication failed is shown by display screen.
The above embodiment of the present invention, fingerprint characteristic is carried out after carrying out High-resolution Processing using the initial fingerprint to collection again
Extract to extract effective fingerprint feature information, improve the accuracy of fingerprint authentication.
Preferably, the fingerprint image component submodule 20, form component is passed through to the initial fingerprint image collected
Analysis (MCA) method is handled, and the different shape in initial fingerprint image is separated, and obtains corresponding initial smooth point
Amount and initial texture component, set the algebraically repeatly in MCA methods as 60, iteration threshold is 10-7。
This preferred embodiment, sets fingerprint image component submodule 20, and the different shape in initial fingerprint image is carried out
Separation, it is to avoid the situation that the artifact and grain details that fingerprint image smooth region is produced are smoothed, is carried out to component processing
Optimization, maintains preferable component performance, is that follow-up fingerprint image processing reduces amount of calculation, lifts the calculating speed of total system
Degree.
Preferably, the component image processing submodule 21, the initial texture obtained after being separated to initial fingerprint image point
Amount carries out non-down sampling contourlet transform (NSCT) processing, and initial smooth component is carried out being based on Stationary Wavelet Transform (SWT) place
Reason, be specially：
(1) initial high resolution texture component is obtained after the Bicubic interpolation that 2n times is carried out to initial texture component
Then non-down sampling contourlet transform (NSCT) is carried out, the low pass subband of an initial high resolution texture component is obtained, correspondence
Low pass subband coefficient be2 are included with i different scale and each yardstickiThe band logical directional subband of individual different directions,
The band logical directional subband coefficient in corresponding i-th of yardstick, s-th of direction is
(2) weighted value and maximum of coefficient in all band logical directional subbands under current scale are calculated：
In formula,Represent that self-defined weighted value calculates function, PmaxRepresent maximum value calculation function, wsFor weight factor,For the band logical directional subband coefficient in i-th of yardstick, s-th of direction；
(3) to each pixel in whole band logical directional subbandsStrong edge is classified as according to row (j) and row (k)
Or weak edge, defining strong and weak marginal classification decision criteria is：
Wherein μ is the classification control parameter of setting,For the standard deviation of noise under current scale i；
(4) to each pixel of strong edge in whole band logical directional subbandsCoefficient carry out enhancing processing, definition
Enhancing handles function：
In formula,For pixelCorresponding NSCT conversion coefficients,For after processing NSCT convert it is strong
Edge pixel coefficient；
(5) to including fuzzy and deformation each pixel in weak edge in whole band logical directional subbandsCoefficient carry out
Decrease processing, definition weakens processing function and is：
In formula,For pixelCorresponding NSCT conversion coefficients,Weak side is converted for NSCT after processing
Edge pixel coefficient；
Finally, whole band logical directional subband coefficients of texture component after being handledWithAnd low pass
Sub-band coefficientsAfterwards, high-resolution texture component is obtained by NSCT inverse transformations (INSCT) reconstruct.
(5) Stationary Wavelet Transform (SWT) is based on, High-resolution Processing is carried out to initial smooth component, will be initially smooth
Component is decomposed into the low pass with initial fingerprint image size formed objects, horizontal direction, vertical direction and diagonally opposed four sons
Band, directly replaces low pass subband with initial smooth component, then to initial smooth component, horizontal direction, vertical direction and diagonal side
2n times of interpolation is carried out to four son bands, finally, SWT inverse transformations (ISWT) reconstruct is carried out to four son bands after interpolation and obtains high score
The smooth component of resolution.
In this preferred embodiment, initial fingerprint image different shape is separated, by the initial texture component after separation
Non-down sampling contourlet transform is carried out, Stationary Wavelet Transform processing is carried out to initial smooth component, and use customized power
Marginal classification decision criteria accurately distinguishes the strong and weak edge of each pixel, and carries out different processing to it, there is defined
Enhancing processing function and decrease processing function formula needed for processing, are conducive to highlighting the contour feature of strong edge, weaken weak
The fuzzy and metaboly at edge, it is to avoid occur the situation of distortion at signal breakpoint, the enhancing of fingerprint image noise removal capability is more accurate
Really represent fingerprint image details.
Preferably, the fingerprint image synthesis submodule 22, to the smooth component of high-resolution and high-resolution texture component
Synthesis is overlapped, defining Superposition Formula is：
In formula,For high-resolution fingerprint image,For the smooth component of high-resolution,For high-resolution texture component, δ
For customized parameter, 0<δ<2, adjustable δ strengthen or reduce texture component, and A is the initial fingerprint image collected, AaFor
Initial smooth component, AbInitial texture component.
In this preferred embodiment, customized parameter is introduced in Superposition Formula, can be to high-resolution according to the difference of environment
The smooth component of fingerprint image and texture component are adjusted, and strengthen the adaptability of sensor, can be under various circumstances to difference
Details in fingerprint display selected.
Finally it should be noted that the above embodiments are merely illustrative of the technical solutions of the present invention, rather than to present invention guarantor
The limitation of scope is protected, although being explained with reference to preferred embodiment to the present invention, one of ordinary skill in the art should
Work as understanding, technical scheme can be modified or equivalent, without departing from the reality of technical solution of the present invention
Matter and scope.
Claims (7)
1. a kind of fingerprint recognition system, it is characterized in that, including finger print acquisition module, fingerprint image processing module, fingerprint characteristic carry
Modulus block, checking identification module, personnel's fingerprint database and result display module, personnel's fingerprint database, which is used to store, to be marked
Accurate personnel's fingerprint image；The finger print acquisition module is used to gather initial fingerprint image；The fingerprint image processing module is used
In the initial fingerprint image collected is carried out into High-resolution Processing, high-resolution fingerprint image is obtained；The fingerprint characteristic
Extraction module is used to extract the fingerprint characteristic image in high-resolution fingerprint image；The checking identification module is used for fingerprint is special
Image is levied to carry out matching checking with personnel's fingerprint image of the standard in personnel's fingerprint database；The result display module is used for
Receive the result and shown.
2. a kind of fingerprint recognition system according to claim 1, it is characterized in that, the finger print acquisition module collection initially refers to
Gathered, the lens centre of the focus alignment finger pressing of camera, posted on the outside of lens using common camera during print image
Anti-fingerprint pad pasting.
3. a kind of fingerprint recognition system according to claim 1, it is characterized in that, it is described to verify identification module by fingerprint characteristic
Personnel's fingerprint image of image and personnel's fingerprint database Plays carries out contrast verification, can be with personnel in fingerprint characteristic image
When personnel's fingerprint image of fingerprint database Plays matches, shown and be proved to be successful by display screen, in fingerprint characteristic image
When can not be matched with personnel's fingerprint image of personnel's fingerprint database Plays, authentication failed is shown by display screen.
4. a kind of fingerprint recognition system according to claim 1, it is characterized in that, the fingerprint image processing module includes three
Individual submodule, be respectively：Fingerprint image component submodule, component image processing submodule, fingerprint image synthesis submodule；
(1) the fingerprint image component submodule is used to carry out initial fingerprint image by form PCA (MCA) method
Component processing, obtains the initial smooth component and texture component of initial fingerprint image；
(2) the component image processing submodule is used to carry out based on Stationary Wavelet Transform (SWT) processing initial smooth component,
High-resolution smooth component is obtained, carrying out non-down sampling contourlet transform (NSCT) to initial texture component is handled, and is obtained
High-resolution texture component；
(3) the fingerprint image synthesis submodule is used to be overlapped the smooth component of high-resolution and high-resolution texture component
Synthesis, obtains high-resolution fingerprint image.
5. a kind of fingerprint recognition system according to claim 4, it is characterized in that, it is described that form is passed through to initial fingerprint image
PCA (MCA) method carries out component processing, and the different shape in initial fingerprint image is separated, and obtains corresponding first
Begin smooth component and initial texture component.
6. a kind of fingerprint recognition system according to claim 5, it is characterized in that, obtained after the initial fingerprint image separation
Initial texture component carry out non-down sampling contourlet transform (NSCT) processing, initial smooth component is carried out based on stationary wavelet
(SWT) processing is converted, including：
(1) initial high resolution texture component is obtained after the Bicubic interpolation that 2n times is carried out to initial texture componentThen enter
Row non-down sampling contourlet transform (NSCT), obtains the low pass subband of an initial high resolution texture component, corresponding low pass
Sub-band coefficients are2 are included with i different scale and each yardstickiThe band logical directional subband of individual different directions, it is corresponding
The band logical directional subband coefficient in i-th of yardstick, s-th of direction is
(2) weighted value and maximum of coefficient in all band logical directional subbands under current scale are calculated：
In formula,Represent that self-defined weighted value calculates function, PmaxRepresent maximum value calculation function, wsFor weight factorFor the band logical directional subband coefficient in i-th of yardstick, s-th of direction；
(3) to each pixel in whole band logical directional subbandsStrong edge is classified as according to row (j) and row (k) or weak
Edge, defining strong and weak marginal classification decision criteria is：
Wherein μ is the classification control parameter of setting,For the standard deviation of noise under current scale i；
(4) to each pixel of strong edge in whole band logical directional subbandsCoefficient carry out enhancing processing, definition enhancing
Handling function is：
In formula,For pixelCorresponding NSCT conversion coefficients,The strong edge converted for NSCT after processing
Pixel coefficient；
(5) to including fuzzy and deformation each pixel in weak edge in whole band logical directional subbandsCoefficient weakened
Processing, definition weakens processing function and is：
In formula,For pixelCorresponding NSCT conversion coefficients,Weak edge picture is converted for NSCT after processing
Prime system number；
Finally, whole band logical directional subband coefficients of texture component after being handledWithAnd low pass subband
CoefficientAfterwards, high-resolution texture component is obtained by NSCT inverse transformations (INSCT) reconstruct.
(6) Stationary Wavelet Transform (SWT) is based on, will initial smooth component to initial smooth component progress High-resolution Processing
The low pass with initial fingerprint image size formed objects, horizontal direction, vertical direction and diagonally opposed four subbands are decomposed into, directly
Connect and low pass subband is replaced with initial smooth component, then to initial smoothly component, horizontal direction, vertical direction and diagonally opposed four
Subband carries out 2n times of interpolation, finally, and obtaining high-resolution to four son band progress SWT inverse transformations (ISWT) reconstruct after interpolation puts down
Sliding component.
7. a kind of fingerprint recognition system according to claim 6, it is characterized in that, the smooth component of high-resolution and high score
Resolution texture component is overlapped synthesis, defines Superposition Formula and is：
In formula,For high-resolution fingerprint image,For the smooth component of high-resolution,For high-resolution texture component, δ is can
Regulation parameter, 0<δ<2, adjustable δ strengthen or reduce texture component, and A is the initial fingerprint image collected, AaTo be initial
Smooth component, AbInitial texture component.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710180015.4A CN106960191A (en) | 2017-03-23 | 2017-03-23 | A kind of fingerprint recognition system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710180015.4A CN106960191A (en) | 2017-03-23 | 2017-03-23 | A kind of fingerprint recognition system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN106960191A true CN106960191A (en) | 2017-07-18 |
Family
ID=59470960
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710180015.4A Withdrawn CN106960191A (en) | 2017-03-23 | 2017-03-23 | A kind of fingerprint recognition system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106960191A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111666807A (en) * | 2020-04-20 | 2020-09-15 | 浙江工业大学 | Multi-source fingerprint image fusion method based on convolution sparse representation |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101499130A (en) * | 2008-01-30 | 2009-08-05 | 深圳市普罗巴克科技股份有限公司 | Fingerprint recognition method and fingerprint recognition system |
| CN102708363A (en) * | 2012-05-15 | 2012-10-03 | 江苏群雄智能科技有限公司 | Self-adaptive fingerprint identification system |
| CN104616001A (en) * | 2015-03-04 | 2015-05-13 | 上海箩箕技术有限公司 | Fingerprint recognition system and fingerprint recognition method |
| CN104732194A (en) * | 2014-11-12 | 2015-06-24 | 国家电网公司 | Fingerprint recognition system |
-
2017
- 2017-03-23 CN CN201710180015.4A patent/CN106960191A/en not_active Withdrawn
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101499130A (en) * | 2008-01-30 | 2009-08-05 | 深圳市普罗巴克科技股份有限公司 | Fingerprint recognition method and fingerprint recognition system |
| CN102708363A (en) * | 2012-05-15 | 2012-10-03 | 江苏群雄智能科技有限公司 | Self-adaptive fingerprint identification system |
| CN104732194A (en) * | 2014-11-12 | 2015-06-24 | 国家电网公司 | Fingerprint recognition system |
| CN104616001A (en) * | 2015-03-04 | 2015-05-13 | 上海箩箕技术有限公司 | Fingerprint recognition system and fingerprint recognition method |
Non-Patent Citations (1)
| Title |
|---|
| 刘微容: "面向汽车主动安全的图像处理关键技术研究", 《中国博士学位论文全文数据库 信息科技辑》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111666807A (en) * | 2020-04-20 | 2020-09-15 | 浙江工业大学 | Multi-source fingerprint image fusion method based on convolution sparse representation |
| CN111666807B (en) * | 2020-04-20 | 2023-06-30 | 浙江工业大学 | Multi-source fingerprint image fusion method based on convolution sparse representation |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107393542B (en) | Bird species identification method based on two-channel neural network | |
| CN101430759B (en) | Optimized recognition pretreatment method for human face | |
| CN103116763B (en) | A kind of living body faces detection method based on hsv color Spatial Statistical Character | |
| CN100458831C (en) | Human face model training module and method, human face real-time certification system and method | |
| CN101615292B (en) | Accurate positioning method for human eye on the basis of gray gradation information | |
| CN110969124A (en) | Two-dimensional human body posture estimation method and system based on lightweight multi-branch network | |
| CN107066961B (en) | Fingerprint method for registering and device | |
| CN110647820B (en) | Low-resolution face recognition method based on feature space super-resolution mapping | |
| CN105138983B (en) | The pedestrian detection method divided based on weighting block model and selective search | |
| CN107958443A (en) | A kind of fingerprint image joining method based on crestal line feature and TPS deformation models | |
| CN107169471A (en) | A kind of fingerprint recognition system based on image co-registration | |
| CN106960192B (en) | Based on the preceding roadmarking extraction system to camera in automatic Pilot | |
| CN111291669A (en) | Two-channel depression angle human face fusion correction GAN network and human face fusion correction method | |
| CN113158833B (en) | Unmanned vehicle control command method based on human body posture | |
| CN110895697A (en) | Transformer nameplate information acquisition method and intelligent acquisition system | |
| CN103714331A (en) | Facial expression feature extraction method based on point distribution model | |
| CN114155445A (en) | SAR image target detection method based on improved YOLOv3 | |
| CN113392728B (en) | Target detection method based on SSA sharpening attention mechanism | |
| CN106960191A (en) | A kind of fingerprint recognition system | |
| CN103136530A (en) | Method for automatically recognizing target images in video images under complex industrial environment | |
| CN105138886A (en) | Robot biometric identification system | |
| CN103984965A (en) | Pedestrian detection method based on multi-resolution character association | |
| CN104166840A (en) | Focusing realization method based on video conference system | |
| CN114998708B (en) | Tea variety identification method and device based on map signals | |
| CN104731324B (en) | A kind of gesture inner plane rotation detection model generation method based on HOG+SVM frameworks |
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 | ||
| WW01 | Invention patent application withdrawn after publication |
Application publication date: 20170718 |
|
| WW01 | Invention patent application withdrawn after publication |