CN106446813A - Calibration method for finger vein identification device - Google Patents

Calibration method for finger vein identification device Download PDF

Info

Publication number
CN106446813A
CN106446813A CN201610820854.3A CN201610820854A CN106446813A CN 106446813 A CN106446813 A CN 106446813A CN 201610820854 A CN201610820854 A CN 201610820854A CN 106446813 A CN106446813 A CN 106446813A
Authority
CN
China
Prior art keywords
vein
section
finger
sigma
identification device
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.)
Granted
Application number
CN201610820854.3A
Other languages
Chinese (zh)
Other versions
CN106446813B (en
Inventor
梁添才
刘建平
金晓峰
黎明
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
GUANGZHOU GRGBANKING INFORMATION TECHNOLOGY Co Ltd
GRG Banking Equipment Co Ltd
Guangdian Yuntong Financial Electronic Co Ltd
Original Assignee
GUANGZHOU GRGBANKING INFORMATION TECHNOLOGY Co Ltd
Guangdian Yuntong Financial Electronic Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by GUANGZHOU GRGBANKING INFORMATION TECHNOLOGY Co Ltd, Guangdian Yuntong Financial Electronic Co Ltd filed Critical GUANGZHOU GRGBANKING INFORMATION TECHNOLOGY Co Ltd
Priority to CN201610820854.3A priority Critical patent/CN106446813B/en
Publication of CN106446813A publication Critical patent/CN106446813A/en
Priority to PCT/CN2017/087841 priority patent/WO2018049858A1/en
Application granted granted Critical
Publication of CN106446813B publication Critical patent/CN106446813B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/20—Image preprocessing
    • G06V10/22—Image preprocessing by selection of a specific region containing or referencing a pattern; Locating or processing of specific regions to guide the detection or recognition
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/20—Image preprocessing
    • G06V10/26—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion
    • G06V10/267—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion by performing operations on regions, e.g. growing, shrinking or watersheds
    • 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
    • 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
    • G06V40/14—Vascular patterns

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Multimedia (AREA)
  • Theoretical Computer Science (AREA)
  • Human Computer Interaction (AREA)
  • Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
  • Image Input (AREA)

Abstract

The invention discloses a calibration method for a finger vein identification device. The method comprises the steps of: acquiring at least one finger vein image of each of multiple persons in the current state by using the finger vein identification device; calculating a Gaussian distribution parameter of a sectional pixel gray value of the acquired finger vein image; calculating a gamma mapping coefficient of an infrared image sensor of the finger vein identification device according to the Gaussian distribution parameter of the sectional pixel gray value of the finger vein image; adjusting a pixel input value and pixel output value mapping relation acquired by the infrared image sensor in the current state according to the gamma mapping coefficient, thus identifying the adjusted pixel output value to obtain an identification result. The calibration method for the finger vein identification device can effectively reduce the influence of factors such as different regional climates, temperatures, illumination and the like on the identification success rate of the finger vein device.

Description

A kind of calibration steps referring to vein identification device
Technical field
The present invention relates to IMAQ field, more particularly, to a kind of calibration steps referring to vein identification device.
Background technology
With the development in epoch, personal information security is more and more important.Rational authentication techniques are selected to be to ensure that information is pacified Full necessary factor.Traditional authentication techniques are based on personal identification number, and the probability more and more higher that password is cracked.Biological identification The significant change of a few years from now on information industry will be become, more and more individual, consumer, company or even government organs all hold Recognize, existing is far from being enough based on the identification system of smart card, identification card number and password, biometrics identification technology Consequence will be occupied in terms of following offer safety certification.
Vein identification technology is to reach authentication purpose by carrying out vivo identification to finger or palm medium sized vein image, tool There are highly false proof, In vivo detection, pin-point accuracy, strong adaptability and easy easy-to-use characteristic.
Refer to vein identification device and be related to a series of content such as light source, optical lens, photosensor chip, the application of this identifying device Each region of region overlay China, extraneous weather and temperature can affect the luminous intensity of equipment near-infrared LED lamp, optical frames Head light transmittance, the light that the direct interference photosensor chip of exterior light note simultaneously receives, and then have influence on the image collecting, Lead to eventually refer to the reduction of the recognition success rate of vein identification device.In order to effectively improve the identification success referring to vein identification device Rate, needs to carry out respective alignment to finger vein identification device before use.
Content of the invention
The present invention propose a kind of refer to vein identification device calibration steps, can effectively reduce different geographical weather, The impact to the recognition success rate referring to vein identification device for the factor such as temperature and illumination.
For achieving the above object, the embodiment of the present invention proposes a kind of calibration steps referring to vein identification device, including step Suddenly：
S1, using refer to vein identification device infrared image sensor gather current state under N number of people each L width refer to vein Image, wherein N >=2, L >=1；
S2, calculate collect described N*L refer to vein image section grey scale pixel value Gaussian Distribution Parameters (μ, σ)；Wherein, μ is average, and σ is standard deviation, and each described section grey scale pixel value referring to vein image meets Gaussian Profile；
S3, calculate the gamma mapping coefficient of described infrared image sensor by below equation：
Wherein, (μs,σs)、γsFor the described finger default standard value of vein identification device, (μ, σ) obtains for above-mentioned steps Gaussian Distribution Parameters, sign (for x) sign function, γ is the gamma mapping coefficient of described infrared image sensor；
S4, according to calculated described gamma mapping coefficient γ, adjust that described infrared image sensor obtains is current Pixel input values f under state and pixel output f1Mapping relations, then after described finger vein identification device is to adjustment Pixel output f1It is identified to obtain recognition result.
The recognizer of the finger vein identification device that the present invention provides is directly to be identified on the image collecting, and adopts The uniformity of the image collecting can directly influence device recognition success rate, improves distinct device in difference by above-mentioned steps Gather the uniformity of image under environment, can effectively reduce IMAQ uniformity to the recognizer referring to vein identification device Impact, improves recognition success rate.
As the improvement of such scheme, calculate the section of the described N*L finger vein image collecting by following steps The Gaussian Distribution Parameters (μ, σ) of grey scale pixel value；
S21, to each described finger vein image carry out binaryzation to mark off finger venosomes, based on described finger vein area Domain obtains referring to vein lines；
S22, to each described finger vein lines take equidistant T section, B=N*L*T section is obtained；Take each The gray value of section all pixels, is designated asWhereinFor the gray value of the ith pixel in b-th section, MbNumber of pixels corresponding to b-th section, T >=3；
S23, it is calculated the Gaussian Distribution Parameters (μ, σ) of described B section grey scale pixel value by below equation:
As the improvement of such scheme, by formulaAdjust what described infrared image sensor obtained Pixel input values f under current state and pixel output f1Mapping relations.
Understand that in vein image, the section pixel grey scale of every radicular vein substantially conforms to Gauss and divides according to experimental data statistics Cloth, refers to vein lines by above-mentioned steps to every and takes multiple sections to enrich statistics, make statistics more accurate.
The invention allows for another kind refers to the calibration steps of vein identification device, including step：
S1, using refer to vein identification device infrared image sensor gather current state under N number of people each L width refer to vein Image, wherein N >=2, L >=1；
S2, calculate respectively collect each described finger vein image section grey scale pixel value Gaussian Distribution Parameters； Wherein, each described section grey scale pixel value referring to vein image meets Gaussian Profile, and described Gaussian Distribution Parameters include pixel Gray average and standard deviation；
S3, it is calculated each institute according to each described Gaussian Distribution Parameters of section grey scale pixel value referring to vein image State the width referring to vein image, then calculate the mean value h of all width referring to vein imagew, and it is quiet to choose described N*L finger In arteries and veins image, width is in hwAll finger vein images between (1 ± A%)；Wherein, 0 ＜ A≤50, each described finger vein image Width refer to each described refer to vein image on section grey scale pixel value be less than its corresponding pixel grey scale average pixel Number；
Gaussian Distribution Parameters (the μ of the selected section grey scale pixel value of all finger vein images of S4, calculatingr,σr)；
S5, calculate the gamma mapping coefficient of described infrared image sensor by below equation：
Wherein, (μs,σs)、γsFor the described finger default standard value of vein identification device, (μr,σr) obtain for above-mentioned steps Gaussian Distribution Parameters, γ is the gamma mapping coefficient of described infrared image sensor；
S6, according to calculated described gamma mapping coefficient γ, adjust that described infrared image sensor obtains is current Pixel input values f under state and pixel output f1Mapping relations, then after described finger vein identification device is to adjustment Pixel output f1It is identified to obtain recognition result.
As the improvement of such scheme, each described finger vein figure collecting can be calculated using following steps and formula The Gaussian Distribution Parameters of the section grey scale pixel value of picture：
Binaryzation is carried out to mark off finger venosomes to each described finger vein image, is obtained based on described finger venosomes To finger vein lines；
Equidistant T section is taken to each described finger vein lines, takes the gray value of each section all pixels, be designated asWhereinFor the gray value of the ith pixel in t-th section, MtPixel corresponding to t-th section Number, T >=3；
As the improvement of such scheme, can be calculated using below equation and meet the selected of finger vein image width requirement The section grey scale pixel value of all finger vein images Gaussian Distribution Parameters (μr,σr)：
μr=(μ1+μ2+μ3+...+μp)/P
Wherein, P is the total number of selected all finger vein images, μ1,μ2,μ3,...,μpEvery selected by being respectively The pixel grey scale average of the section grey scale pixel value of one finger vein image.
As the improvement of such scheme, the section of selected all finger vein images also can be calculated using below equation Gaussian Distribution Parameters (the μ of grey scale pixel valuer,σr)：
Equidistant T is taken to each described finger vein lines2Individual section, is obtained B2=P*T2Individual section；Take each section The gray value of all pixels, is designated asWhereinFor b2The gray value of the ith pixel in individual section, Mb2For b2Number of pixels corresponding to individual section, P is the total number of selected all finger vein images, T2≥3.
As the improvement of such scheme, using NIBLACK image binaryzation method, each described finger vein image is carried out Binary conversion treatment, thus extract refer to venosomes；After the described finger venosomes extracting is split, carry out skeletal extraction, Obtain referring to vein lines.
As the improvement of such scheme, A=10.
As the improvement of such scheme, by formulaAdjust what described infrared image sensor obtained Pixel input values f under current state and pixel output f1Mapping relations.
In sum, the calibration steps referring to vein identification device of the present invention, according to the vein referring to venous collection image The grey scale change in region, obtains the adjusting parameter that this refers to the corresponding infrared image sensor of vein identification device, can effectively drop The impact of the factors such as low different geographical weather, temperature and illumination.
Brief description
Fig. 1 is a kind of schematic flow sheet of the calibration steps referring to vein identification device in the embodiment of the present invention one.
Fig. 2 is the idiographic flow schematic diagram of step S102 in Fig. 1.
Fig. 3 is that the embodiment of the present invention one medium sized vein extracts schematic diagram.
Fig. 4 is the embodiment of the present invention one medium sized vein schematic cross-section.
Fig. 5 is a kind of schematic flow sheet of the calibration steps referring to vein identification device in the embodiment of the present invention two.
Fig. 6 is the idiographic flow schematic diagram of step S202 in Fig. 5.
Fig. 7 is a kind of schematic flow sheet of the calibration steps referring to vein identification device in the embodiment of the present invention three.
Fig. 8 is the idiographic flow schematic diagram of step S302 in Fig. 7.
Fig. 9 is the idiographic flow schematic diagram of step S304 in Fig. 7.
Specific embodiment
Below in conjunction with the accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is carried out clear, complete Site preparation description is it is clear that described embodiment is only a part of embodiment of the present invention, rather than whole embodiments.It is based on Embodiment in the present invention, it is every other that those of ordinary skill in the art are obtained under the premise of not making creative work Embodiment, broadly falls into the scope of protection of the invention.
The present invention provides a kind of calibration steps referring to vein identification device, for using basis before referring to vein identification device Current environment carries out respective alignment to referring to vein identification device.As it was previously stated, external environment (include extraneous weather, temperature and The factors such as illumination) influence whether infrared image sensor acquired image, and then have influence on the identification referring to vein identification device Accuracy rate.Therefore, the present invention provides a kind of calibration steps referring to vein identification device, is to improve different finger hand vein recognition Device, the uniformity of IMAQ in different environments, affected thus reducing and referring to vein identification device by environmental disturbances, from And improve the recognition accuracy referring to vein identification device.Below by the multiple embodiments finger vein identification device to the present invention Calibration steps is specifically described.
Referring to Fig. 1, it is that a kind of flow process of the calibration steps of finger vein identification device that the embodiment of the present invention one provides is illustrated Figure, a kind of calibration steps of finger vein identification device that the embodiment of the present invention one provides includes step S101～S104：
S101, using refer to vein identification device infrared image sensor gather current state under N number of people each L width refer to quiet Arteries and veins image, wherein N >=2, L >=1；
This step refers to vein image for collection.Specifically, in this step, using the infrared figure referring to vein identification device As sensor adopts the finger that vein infrared light supply irradiates N number of picker that refers to of preset strength, collection current state is (in same area Domain environment) under N number of people each L width refer to vein image, thus obtain N*L finger vein image, wherein N >=2, L >=1.
S102：Calculate collect described N*L refer to vein image section grey scale pixel value Gaussian Distribution Parameters (μ, σ)；Wherein, each described section grey scale pixel value referring to vein image meets Gaussian Profile.
This step is used for the gray value of the N*L finger vein image collecting is counted.Specifically, with reference to Fig. 2, should Step can be realized by following steps, including step S1021～S1023, wherein：
S1021：Binaryzation is carried out to mark off finger venosomes to each described finger vein image, based on described finger vein Region obtains referring to vein lines；
Referring to Fig. 3, it is that the embodiment of the present invention one medium sized vein extracts schematic diagram, extracts vein lines from vein image 31 32, specifically, binary conversion treatment is carried out to each described finger vein image using NIBLACK image binaryzation method, thus carrying Fetching venosomes.After the described finger venosomes extracting is split, carry out skeletal extraction, thus obtaining referring to intravenous line Bar.
S1022：Equidistant T section is taken to each described finger vein lines, B=N*L*T section is obtained；Take every The gray value of individual section all pixels, is designated asWhereinGray scale for the ith pixel in b-th section Value, MbNumber of pixels corresponding to b-th section, T >=3；Wherein, each described section grey scale pixel value referring to vein image Meet Gaussian Profile；
Referring to Fig. 4, it is the embodiment of the present invention one medium sized vein schematic cross-section, specifically, to the vein lines 40 extracting Take equidistant T section, obtain vein section 41, vein section 42, vein section 43.
S1023：It is calculated the Gaussian Distribution Parameters (μ, σ) of described B section grey scale pixel value by below equation:
S103：By formula：Calculate described infrared image sensor Gamma mapping coefficient, wherein, (μs,σs)、γsFor the described finger default standard value of vein identification device, (μ, σ) is above-mentioned steps The Gaussian Distribution Parameters obtaining, γ is the gamma mapping coefficient of described infrared image sensor.
The present embodiment by gamma mapping coefficient GAMMA infrared image sensor is corrected effect be in order that Distinct device, when different temperatures, region, gather same finger when, the image of output is as consistent as possible, from And ensure equipment recognition effect.
According to calculated described gamma mapping coefficient γ, by formulaAdjust described infrared figure As the pixel input values f under the current state that sensor obtains and pixel output f1Mapping relations, then pass through described finger quiet Arteries and veins identifying device is to the pixel output f after adjustment1It is identified to obtain recognition result.
In this step, according to step S103 calculated gamma mapping coefficient γ, you can public by default adjustment FormulaTo the image pixel referring under the current state that vein identification device is obtained by infrared image sensor Calibrated, and the image after pixel alignment is just carried out refer to hand vein recognition, to be identified result.Wherein, identify institute Using algorithmic procedure can be using well known to a person skilled in the art mode, here omits description.
Embodiments described above one, using refer to vein identification device infrared image sensor gather many people, everyone at least One width refers to vein image, and takes multiple sections to every finger vein lines, enriches statistics, makes statistics more accurate, Meanwhile, by gamma mapping coefficient in embodiment one to the correction of infrared image sensor so that difference refer to vein identification device, When different temperatures, region, when gathering same finger, the image of output is as consistent as possible, improves finger vein The recognition success rate of identifying device.
Referring to Fig. 5, it is that a kind of flow process of the calibration steps of finger vein identification device that the embodiment of the present invention two provides is illustrated Figure, a kind of calibration steps of finger vein identification device that the embodiment of the present invention two provides includes step S201～S206：
S201：Referred to quiet using N number of people each L width that the infrared image sensor referring to vein identification device gathers under current state Arteries and veins image, wherein N >=2, L >=1；
This step refers to vein image for collection.Specifically, in this step, using the infrared figure referring to vein identification device As sensor adopts the finger that vein infrared light supply irradiates N number of picker that refers to of preset strength, collection current state is (in same area Domain environment) under N number of people each L width refer to vein image, thus obtain N*L finger vein image, wherein N >=2, L >=1.
S202：Calculate the Gaussian Profile ginseng of the section grey scale pixel value of each described finger vein image collecting respectively Number；Wherein, each described section grey scale pixel value referring to vein image meets Gaussian Profile, and described Gaussian Distribution Parameters include picture Plain gray average and standard deviation；
This step is used for the gray value of the N*L finger vein image collecting is counted.Specifically, referring to Fig. 4, should Step can be realized by following steps, including step S2021～S2023, wherein：
S2021：Binaryzation is carried out to mark off finger venosomes to each described finger vein image, based on described finger vein Region obtains referring to vein lines；
Specifically, binary conversion treatment is carried out to each described finger vein image using NIBLACK image binaryzation method, from And extract and refer to venosomes.After the described finger venosomes extracting is split, carry out skeletal extraction, thus obtaining referring to vein Lines.
S2022：Equidistant T section is taken to each described finger vein lines, takes the gray scale of each section all pixels Value, is designated asWhereinFor the gray value of the ith pixel in t-th section, MtCorresponding to t-th section Number of pixels, T >=3；Wherein, each described section grey scale pixel value referring to vein image meets Gaussian Profile.
S2023：Divided by the Gauss that below equation is calculated each described section grey scale pixel value referring to vein image Cloth parameter (μ, σ):
S203：It is calculated each according to the Gaussian Distribution Parameters of each described section grey scale pixel value referring to vein image The described width referring to vein image, then calculates the mean value h of all width referring to vein imagew, and choose described N*L finger In vein image, width is in hwAll finger vein images between (1 ± A%)；Wherein, 0 ＜ A≤50, each described finger vein figure The width of picture refers to that each described section grey scale pixel value referring on vein image is less than the picture of its corresponding pixel grey scale average Plain number；
This step is used for the finger vein image collecting further is chosen, and chooses the satisfactory finger of width quiet Arteries and veins image.Specifically, in this step, calculate the width that each refers to vein image first, then obtain width mean value hw, in the present embodiment, A=10, that is, choose the finger vein image of 0.9 times to 1.1 times that width is mean breadth.
S204：Calculate the Gaussian Distribution Parameters (μ of the selected section grey scale pixel value of all finger vein imagesr,σr)；
This step is used for the gray value of the finger vein image chosen is counted.Specifically, use in this step with Lower formula calculates the Gaussian Distribution Parameters (μ of the selected section grey scale pixel value of all finger vein imagesr,σr)：
μr=(μ1+μ2+μ3+...+μp)/P
Wherein, P is the total number of selected all finger vein images, μ1,μ2,μ3,...,μpEvery selected by being respectively The pixel grey scale average of the section grey scale pixel value of one finger vein image.
S205：By formula：Calculate described infrared image sensor Gamma mapping coefficient, wherein, (μs,σs)、γsFor the described finger default standard value of vein identification device, (μr,σr) it is above-mentioned steps The Gaussian Distribution Parameters referring to vein image section grey scale pixel value of the calculated selection of S204, γ is that described infrared image passes The gamma mapping coefficient of sensor.
S206：According to calculated described gamma mapping coefficient γ, by formulaAdjustment is described Pixel input values f under the current state that infrared image sensor obtains and pixel output f1Mapping relations, then pass through institute State and refer to vein identification device to the pixel output f after adjustment1It is identified to obtain recognition result.
In this step, according to step S205 calculated gamma mapping coefficient γ, you can public by default adjustment FormulaTo the image pixel referring under the current state that vein identification device is obtained by infrared image sensor Calibrated, and the image after pixel alignment is just carried out refer to hand vein recognition, to be identified result.
Embodiments described above two, different from embodiment one is to calculate the section grey scale pixel value referring to vein image Chosen to referring to vein image during Gaussian Distribution Parameters, preferred in all finger vein images collecting so that experiment is tied The width of really more stable finger vein image is the finger vein image of 0.9 times to 1.1 times of mean breadth, quiet according to the finger chosen The Gaussian Distribution Parameters of the section grey scale pixel value of arteries and veins image obtain gamma mapping coefficient, after the calibration of this gamma mapping coefficient Finger vein identification device recognition success rate higher.
Referring to Fig. 7, it is that a kind of flow process of the calibration steps of finger vein identification device that the embodiment of the present invention three provides is illustrated Figure, a kind of calibration steps of finger vein identification device that the embodiment of the present invention three provides includes step S301～S306：
S301：Referred to quiet using N number of people each L width that the infrared image sensor referring to vein identification device gathers under current state Arteries and veins image, wherein N >=2, L >=1；
This step refers to vein image for collection.Specifically, in this step, using the infrared figure referring to vein identification device As sensor adopts the finger that vein infrared light supply irradiates N number of picker that refers to of preset strength, collection current state is (in same area Domain environment) under N number of people each L width refer to vein image, thus obtain N*L finger vein image, wherein N >=2, L >=1.
S302：Calculate the Gaussian Profile ginseng of the section grey scale pixel value of each described finger vein image collecting respectively Number, wherein, each described section grey scale pixel value referring to vein image meets Gaussian Profile, and described Gaussian Distribution Parameters include picture Plain gray average and standard deviation；
This step is used for the gray value of the N*L finger vein image collecting is counted.Specifically, referring to Fig. 6, should Step can be realized by following steps, including step S3021～S3023, wherein：
S3021：Binaryzation is carried out to mark off finger venosomes to each described finger vein image, based on described finger vein Region obtains referring to vein lines；
Specifically, binary conversion treatment is carried out to each described finger vein image using NIBLACK image binaryzation method, from And extract and refer to venosomes.After the described finger venosomes extracting is split, carry out skeletal extraction, thus obtaining referring to vein Lines.
S3022：Equidistant T section is taken to each described finger vein lines, takes the gray scale of each section all pixels Value, is designated asWhereinFor the gray value of the ith pixel in t-th section, MtCorresponding to t-th section Number of pixels, T >=3, wherein, each described refer to vein image section grey scale pixel value meet Gaussian Profile.
S3023：Divided by the Gauss that below equation is calculated each described section grey scale pixel value referring to vein image Cloth parameter (μ, σ):
S303：It is calculated each according to the Gaussian Distribution Parameters of each described section grey scale pixel value referring to vein image The described width referring to vein image, then calculates the mean value h of all width referring to vein imagew, and choose described N*L finger In vein image, width is in hwAll finger vein images between (1 ± A%)；Wherein, 0 ＜ A≤50, each described finger vein figure The width of picture refers to that each described section grey scale pixel value referring on vein image is less than the picture of its corresponding pixel grey scale average Plain number.
This step is used for the finger vein image collecting further is chosen, and chooses the satisfactory finger of width quiet Arteries and veins image.Specifically, in this step, calculate the width that each refers to vein image first, then obtain width mean value hw, in the present embodiment, A=10, that is, choose the finger vein image of 0.9 times to 1.1 times that width is mean breadth.
S304：Calculate the Gaussian Distribution Parameters (μ of the selected section grey scale pixel value of all finger vein imagesr,σr)；
This step is used for the gray value of the finger vein image chosen is counted.Specifically, referring to Fig. 7, this step can To be realized by following steps, including step S3041～S3042, wherein：
S3041：Equidistant T is taken to each described finger vein lines2Individual section, is obtained B2=P*T2Individual section；Take every The gray value of individual section all pixels, is designated asWhereinFor b2The ash of the ith pixel in individual section Angle value, Mb2For b2Number of pixels corresponding to individual section, P is the total number of selected all finger vein images, T2≥3.
S3042：The Gauss being calculated the section grey scale pixel value of selected all finger vein images by below equation is divided Cloth parameter (μr,σr)
S305：By formula：Calculate described infrared image sensor Gamma mapping coefficient, wherein, (μs,σs)、γsFor the described finger default standard value of vein identification device, (μr,σr) it is above-mentioned steps The section grey scale pixel value Gaussian Distribution Parameters of the finger vein image of the calculated selection of S304, γ is that described infrared image passes The gamma mapping coefficient of sensor.
S306：According to calculated described gamma mapping coefficient γ, by formulaAdjustment is described Pixel input values f under the current state that infrared image sensor obtains and pixel output f1Mapping relations, then pass through institute State and refer to vein identification device to the pixel output f after adjustment1It is identified to obtain recognition result.
In this step, according to step S305 calculated gamma mapping coefficient γ, you can public by default adjustment FormulaTo the image pixel referring under the current state that vein identification device is obtained by infrared image sensor Calibrated, and the image after pixel alignment is just carried out refer to hand vein recognition, to be identified result.
Embodiments described above three, different from embodiment one is to calculate the section grey scale pixel value referring to vein image Chosen to referring to vein image during Gaussian Distribution Parameters, preferred in all finger vein images collecting so that experiment is tied The width of really more stable finger vein image is the finger vein image of 0.9 times to 1.1 times of mean breadth, quiet according to the finger chosen The Gaussian Distribution Parameters of the section grey scale pixel value of arteries and veins image obtain gamma mapping coefficient, after the calibration of this gamma mapping coefficient Finger vein identification device recognition success rate higher.Difference with embodiment two is to calculate selected finger vein image During the Gaussian Distribution Parameters of section grey scale pixel value, method is different, and embodiment two is first to obtain the finger vein that each width is chosen The interface grey scale pixel value Gaussian Distribution Parameters of image, then calculate all Gaussian Distribution Parameters referring to vein image, and embodiment Three Gaussian Distribution Parameters being directly to obtain selected all finger vein sections grey scale pixel value.
The above is the preferred embodiment of the present invention it is noted that for those skilled in the art For, under the premise without departing from the principles of the invention, some improvements and modifications can also be made, these improvements and modifications are also considered as Protection scope of the present invention.

Claims (10)

1. a kind of calibration steps referring to vein identification device is it is characterised in that include step：
S1, using refer to vein identification device infrared image sensor gather current state under N number of people each L width refer to vein figure Picture, wherein N >=2, L >=1；
The Gaussian Distribution Parameters (μ, σ) of the section grey scale pixel value of described N*L finger vein image that S2, calculating collect；Its In, μ is average, and σ is standard deviation, and each described section grey scale pixel value referring to vein image meets Gaussian Profile；
S3, calculate the gamma mapping coefficient of described infrared image sensor by below equation：
γ = 1 - s i g n ( μ - μ s ) ( μ - μ s ) 2 1 + ( σ - σ s ) 2 γ s
Wherein, (μs,σs)、γsFor the described finger default standard value of vein identification device, the Gauss that (μ, σ) obtains for above-mentioned steps Distributed constant, γ is the gamma mapping coefficient of described infrared image sensor；
S4, the current state being obtained according to calculated described gamma mapping coefficient γ, the described infrared image sensor of adjustment Under pixel input values f and pixel output f1Mapping relations, then by the described vein identification device that refers to the picture after adjustment Plain output valve f1It is identified to obtain recognition result.
2. the calibration steps referring to vein identification device as claimed in claim 1 is it is characterised in that described step S2 specifically includes Step：
S21, to each described finger vein image carry out binaryzation to mark off finger venosomes, based on described finger venosomes obtain To finger vein lines；
S22, to each described finger vein lines take equidistant T section, B=N*L*T section is obtained；Take each section The gray value of all pixels, is designated asWhereinFor the gray value of the ith pixel in b-th section, MbFor Number of pixels corresponding to b-th section, T >=3；
S23, it is calculated the Gaussian Distribution Parameters (μ, σ) of described B section grey scale pixel value by below equation:
μ = Σ b = 1 B Σ i = 1 M b x i b / Σ b = 1 B M b
σ = 1 Σ b = 1 B M b Σ b = 1 B Σ i = 1 M b ( x i b - μ ) 2 .
3. the calibration steps referring to vein identification device as claimed in claim 1 or 2 it is characterised in that in step s 4, passes through FormulaAdjust pixel input values f under the current state that described infrared image sensor obtains and pixel is defeated Go out value f1Mapping relations.
4. a kind of calibration steps referring to vein identification device is it is characterised in that include step：
S1, using refer to vein identification device infrared image sensor gather current state under N number of people each L width refer to vein figure Picture, wherein N >=2, L >=1；
S2, calculate respectively collect each described finger vein image section grey scale pixel value Gaussian Distribution Parameters；Wherein, Each described section grey scale pixel value referring to vein image meets Gaussian Profile, and it is equal that described Gaussian Distribution Parameters include pixel grey scale Value and standard deviation；
S3, it is calculated each described finger according to each described Gaussian Distribution Parameters of section grey scale pixel value referring to vein image The width of vein image, then calculates the mean value h of all width referring to vein imagew, and choose described N*L finger vein figure In picture, width is in hwAll finger vein images between (1 ± A%)；Wherein, 0 ＜ A≤50, each described width referring to vein image Degree refers to that each described section grey scale pixel value referring on vein image is less than the number of pixels of its corresponding pixel grey scale average；
Gaussian Distribution Parameters (the μ of the selected section grey scale pixel value of all finger vein images of S4, calculatingr,σr)；
S5, calculate the gamma mapping coefficient of described infrared image sensor by below equation：
γ = 1 - s i g n ( μ r - μ s ) ( μ r - μ s ) 2 1 + ( σ r - σ s ) 2 γ s
Wherein, (μs,σs)、γsFor the described finger default standard value of vein identification device, (μr,σr) height that obtains for above-mentioned steps This distributed constant, γ is the gamma mapping coefficient of described infrared image sensor；
S6, the current state being obtained according to calculated described gamma mapping coefficient γ, the described infrared image sensor of adjustment Under pixel input values f and pixel output f1Mapping relations, then by the described vein identification device that refers to the picture after adjustment Plain output valve f1It is identified to obtain recognition result.
5. the calibration steps referring to vein identification device as claimed in claim 4 is it is characterised in that described step S2 specifically includes Step：
S21, to each described finger vein image carry out binaryzation to mark off finger venosomes, based on described finger venosomes obtain To finger vein lines；
S22, to each described finger vein lines take equidistant T section, take the gray value of each section all pixels, be designated asWhereinFor the gray value of the ith pixel in t-th section, MtPixel corresponding to t-th section Number, T >=3；
S23, by below equation be calculated each described refer to vein image section grey scale pixel value Gaussian Distribution Parameters (μ,σ):
μ = Σ t = 1 T Σ i = 1 M t x i t / Σ t = 1 T M t
σ = 1 Σ t = 1 T M t Σ t = 1 T Σ i = 1 M t ( x i t - μ ) 2 .
6. the calibration steps referring to vein identification device as claimed in claim 5 is it is characterised in that in step s 4, by with Lower formula calculates the Gaussian Distribution Parameters (μ of the selected section grey scale pixel value of all finger vein imagesr,σr)：
μr=(μ1+μ2+μ3+...+μp)/P
σ r = 1 P { ( μ 1 - μ r ) 2 + ( μ 2 - μ r ) 2 + ... + ( μ P - μ r ) 2 } ;
Wherein, P is the total number of selected all finger vein images, μ1,μ2,μ3,...,μpIt is respectively selected each finger The pixel grey scale average of the section grey scale pixel value of vein image.
7. the calibration steps referring to vein identification device as claimed in claim 5 is it is characterised in that in step s 4, by with Lower step calculates the Gaussian Distribution Parameters (μ of the selected section grey scale pixel value of all finger vein imagesr,σr)：
S41, to each described finger vein lines take equidistant T2Individual section, is obtained B2=P*T2Individual section；Take each section The gray value of all pixels, is designated asWhereinFor b2The gray value of the ith pixel in individual section, Mb2For b2Number of pixels corresponding to individual section, P is the total number of selected all finger vein images, T2≥3；
S42, calculated by below equation selected all finger vein images section grey scale pixel value Gaussian Distribution Parameters (μr,σr):
μ r = Σ b 2 = 1 B 2 Σ i = 1 M b 2 x i b 2 / Σ b 2 = 1 B 2 M b 2
σ r = 1 Σ b 2 = 1 B 2 M b 2 Σ b 2 = 1 B 2 Σ i = 1 M b 2 ( x i b 2 - μ r ) 2 .
8. the calibration steps referring to vein identification device as claimed in claim 5 is it is characterised in that described step S21 is specifically wrapped Include：
S211, using NIBLACK image binaryzation method to each described finger vein image carry out binary conversion treatment, thus extracting Refer to venosomes；
After S212, the described finger venosomes to extraction are split, carry out skeletal extraction, obtain referring to vein lines.
9. as described in claim 4 or 5 finger vein identification device calibration steps it is characterised in that in step s3, A= 10.
10. the calibration steps of the finger vein identification device as described in claim 4 or 5 it is characterised in that in step s 6, leads to Cross formulaAdjust the pixel input values f under the current state that described infrared image sensor obtains and pixel Output valve f1Mapping relations.
CN201610820854.3A 2016-09-13 2016-09-13 A kind of calibration method referring to vein identification device Active CN106446813B (en)

Priority Applications (2)

Application Number Priority Date Filing Date Title
CN201610820854.3A CN106446813B (en) 2016-09-13 2016-09-13 A kind of calibration method referring to vein identification device
PCT/CN2017/087841 WO2018049858A1 (en) 2016-09-13 2017-06-10 Calibration method for finger vein identification apparatus

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201610820854.3A CN106446813B (en) 2016-09-13 2016-09-13 A kind of calibration method referring to vein identification device

Publications (2)

Publication Number Publication Date
CN106446813A true CN106446813A (en) 2017-02-22
CN106446813B CN106446813B (en) 2019-10-11

Family

ID=58168857

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201610820854.3A Active CN106446813B (en) 2016-09-13 2016-09-13 A kind of calibration method referring to vein identification device

Country Status (2)

Country Link
CN (1) CN106446813B (en)
WO (1) WO2018049858A1 (en)

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2018049858A1 (en) * 2016-09-13 2018-03-22 广州广电运通金融电子股份有限公司 Calibration method for finger vein identification apparatus
CN110188675A (en) * 2019-05-29 2019-08-30 Oppo广东移动通信有限公司 Vein collection method and Related product
CN110826513A (en) * 2019-11-13 2020-02-21 圣点世纪科技股份有限公司 Calibration algorithm for consistency of finger vein equipment
CN117631717B (en) * 2024-01-10 2024-06-28 深圳豪达尔机械有限公司 Sensor control and regulation system of variable-frequency cold dryer

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1921585A (en) * 2005-08-25 2007-02-28 精工爱普生株式会社 Gray scale coefficient curve adjustment device and method of establishing adjustment points
CN101982826A (en) * 2010-11-10 2011-03-02 中国船舶重工集团公司第七一○研究所 Finger vein collection and identification method capable of automatically adjusting brightness of light source
JP2011091595A (en) * 2009-10-22 2011-05-06 Kyocera Mita Corp Image processor, image processing method and image forming apparatus
CN103996209A (en) * 2014-05-21 2014-08-20 北京航空航天大学 Infrared vessel object segmentation method based on salient region detection
CN104688184A (en) * 2014-12-05 2015-06-10 南京航空航天大学 Vein imaging method for visible-light skin images

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6763127B1 (en) * 2000-10-06 2004-07-13 Ic Media Corporation Apparatus and method for fingerprint recognition system
JP5205968B2 (en) * 2005-12-21 2013-06-05 日本電気株式会社 Gradation correction method, gradation correction apparatus, gradation correction program, and image device
CN101650439B (en) * 2009-08-28 2011-12-07 西安电子科技大学 Method for detecting change of remote sensing image based on difference edge and joint probability consistency
CN104794689A (en) * 2015-03-12 2015-07-22 哈尔滨工程大学 Preprocessing method for enhancing sonar image contract
CN106446813B (en) * 2016-09-13 2019-10-11 广州广电运通金融电子股份有限公司 A kind of calibration method referring to vein identification device

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1921585A (en) * 2005-08-25 2007-02-28 精工爱普生株式会社 Gray scale coefficient curve adjustment device and method of establishing adjustment points
JP2011091595A (en) * 2009-10-22 2011-05-06 Kyocera Mita Corp Image processor, image processing method and image forming apparatus
CN101982826A (en) * 2010-11-10 2011-03-02 中国船舶重工集团公司第七一○研究所 Finger vein collection and identification method capable of automatically adjusting brightness of light source
CN103996209A (en) * 2014-05-21 2014-08-20 北京航空航天大学 Infrared vessel object segmentation method based on salient region detection
CN104688184A (en) * 2014-12-05 2015-06-10 南京航空航天大学 Vein imaging method for visible-light skin images

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
HANY FARID: "Blind Inverse Gamma Correction", 《IEEE TRANSACTIONS ON IMAGE PROCESSING》 *
蓝惠英: "基于LabVIEW的手掌静脉身份识别系统研究", 《中国优秀硕士学位论文全文数据库》 *

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2018049858A1 (en) * 2016-09-13 2018-03-22 广州广电运通金融电子股份有限公司 Calibration method for finger vein identification apparatus
CN110188675A (en) * 2019-05-29 2019-08-30 Oppo广东移动通信有限公司 Vein collection method and Related product
CN110188675B (en) * 2019-05-29 2021-04-13 Oppo广东移动通信有限公司 Vein collection method and related products
CN110826513A (en) * 2019-11-13 2020-02-21 圣点世纪科技股份有限公司 Calibration algorithm for consistency of finger vein equipment
CN110826513B (en) * 2019-11-13 2022-04-19 圣点世纪科技股份有限公司 Calibration method for consistency of finger vein equipment
CN117631717B (en) * 2024-01-10 2024-06-28 深圳豪达尔机械有限公司 Sensor control and regulation system of variable-frequency cold dryer

Also Published As

Publication number Publication date
CN106446813B (en) 2019-10-11
WO2018049858A1 (en) 2018-03-22

Similar Documents

Publication Publication Date Title
CN110516576B (en) Near-infrared living body face recognition method based on deep neural network
CN104751108B (en) Facial image identification device and facial image recognition method
CN108563990B (en) Certificate authentication method and system based on CIS image acquisition system
CN106250845A (en) Flame detecting method based on convolutional neural networks and device
Berezhnoy et al. Computer analysis of van Gogh’s complementary colours
CN106790019A (en) The encryption method for recognizing flux and device of feature based self study
CN104700376A (en) Gamma correction and smoothing filtering based image histogram equalization enhancing method
CN104408780A (en) Face recognition attendance system
KR20130048076A (en) Face recognition apparatus and control method for the same
CN110348322A (en) Human face in-vivo detection method and equipment based on multi-feature fusion
CN106446813A (en) Calibration method for finger vein identification device
CN106682473A (en) Method and device for identifying identity information of users
KR20160119932A (en) Method and apparatus for face recognition based quality assessment
CN109740572A (en) A kind of human face in-vivo detection method based on partial color textural characteristics
CN107038416A (en) A kind of pedestrian detection method based on bianry image modified HOG features
CN111259763B (en) Target detection method, target detection device, electronic equipment and readable storage medium
CN101615241B (en) Method for screening certificate photos
CN107798279A (en) Face living body detection method and device
CN116311549A (en) Living body object identification method, apparatus, and computer-readable storage medium
CN111178130A (en) Face recognition method, system and readable storage medium based on deep learning
CN109829905A (en) It is a kind of face beautification perceived quality without reference evaluation method
CN107516083A (en) A kind of remote facial image Enhancement Method towards identification
CN111260645A (en) Method and system for detecting tampered image based on block classification deep learning
CN109522865A (en) A kind of characteristic weighing fusion face identification method based on deep neural network
CN101533475A (en) Method for extracting feature of shape-adaptive neighborhood based remote sensing image

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant