CN106204958A - A kind of ATM input equipment being identified by iris - Google Patents
A kind of ATM input equipment being identified by iris Download PDFInfo
- Publication number
- CN106204958A CN106204958A CN201610548331.8A CN201610548331A CN106204958A CN 106204958 A CN106204958 A CN 106204958A CN 201610548331 A CN201610548331 A CN 201610548331A CN 106204958 A CN106204958 A CN 106204958A
- Authority
- CN
- China
- Prior art keywords
- iris
- submodule
- iris image
- input equipment
- 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.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G07—CHECKING-DEVICES
- G07F—COIN-FREED OR LIKE APPARATUS
- G07F19/00—Complete banking systems; Coded card-freed arrangements adapted for dispensing or receiving monies or the like and posting such transactions to existing accounts, e.g. automatic teller machines
- G07F19/20—Automatic teller machines [ATMs]
- G07F19/201—Accessories of ATMs
-
- 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
- G06V40/18—Eye characteristics, e.g. of the iris
- G06V40/193—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
- G06V40/18—Eye characteristics, e.g. of the iris
- G06V40/197—Matching; Classification
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Human Computer Interaction (AREA)
- Theoretical Computer Science (AREA)
- Multimedia (AREA)
- Health & Medical Sciences (AREA)
- Ophthalmology & Optometry (AREA)
- Business, Economics & Management (AREA)
- Finance (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Accounting & Taxation (AREA)
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
- Collating Specific Patterns (AREA)
- Image Input (AREA)
- Image Analysis (AREA)
Abstract
A kind of ATM input equipment being identified by iris of the present invention, including ATM input equipment and the iris identification device that is connected with the ATM input equipment signal of telecommunication, described iris identification device includes: (1) sampling module;(2) pretreatment module;(3) feature coding module, for the feature of iris image is extracted and encoded, it includes that LBP operator processes submodule, for the second time LBP operator for the first time and processes submodule, for the third time LBP operator process submodule and the 4th LBP operator process submodule;(4) codes match module.Invention increases the relatedness of central point and other neighborhood of surrounding, disclosure satisfy that the image texture of different scale and frequency, after repeatedly LBP operator processes submodule process, under not affecting the central point relatedness with surrounding neighbors, constantly reduce code length, saved memory space, decrease amount of calculation, improve recognition speed, enhance recognition accuracy, obtain higher robustness.
Description
Technical field
The present invention relates to ATM input equipment design field, be specifically related to a kind of ATM being identified by iris defeated
Enter device.
Background technology
In correlation technique, the ATM input equipment being identified by iris generally uses basic LBP (local binary mould
Formula) iris image feature extracted and encodes by operator, and LBP operator is a kind of to describe textural characteristics in the range of gradation of image
Method, has the strongest robustness for illumination variation, thus is widely used in the texture feature extraction of image.
Basic LBP operator is commonly defined as: by central point n in 3 × 3 windowsc8 neighborhood n about0,...n7Group
Becoming, defined in it, texture T is: T=(n0-nc,n1-nc,...,n7-nc), it is carried out binary conversion treatment, with ncFor threshold value, neighborhood
8 points and ncRelatively, if being labeled as 1 more than the value of central point, 0 otherwise it is labeled as.Texture T after binaryzation is: T=(sgn
(n0-nc),sgn(n1-nc),...,sgn(n7-nc)), whereinThrough calculating, will obtain with ncCentered by
8 binary numbers, then be weighted different pixels position suing for peace just obtaining the LBP value of central point, the wherein meter of LBP value
Calculation formula is:Pixel each in image is carried out LBP computing, just can obtain figure
The LBP texture description of picture.
But, owing to basic LBP operator cover only 8 neighborhood territory pixels of central point so that it is with other neighborhood of surrounding
Relatedness is the most comprehensive, it is impossible to meet the image texture of different scale and frequency.
Summary of the invention
For the problems referred to above, the present invention provides the one that a kind of recognition speed is fast, identification range is wide to be known by iris
Other ATM input equipment, solves to use basic LBP operator that iris image feature is extracted and encoded in correlation technique
The problem that ATM input equipment system can not process the image texture of different scale and frequency.
The purpose of the present invention realizes by the following technical solutions:
A kind of ATM input equipment being identified by iris, is filled including ATM input equipment with ATM input
Putting the iris identification device that the signal of telecommunication connects, described ATM input equipment includes:
Housing;
Touch control screen module includes a chip, in housing described in described touch control screen Module-embedding;
Randomized blocks includes the set of number of 0-9 and period in order to randomly generate, and is defined as random array, described random number
Period in group and each numeral in 0-9 at least occur once;
Processor is in order to process data, and described processor is connected to described randomized blocks and described touch-control by wire
The described chip of formula screen module.
Preferably, it is characterized in that, above-mentioned hookup mechanism has:
A keyboard region, described random array is provided the most all to show in described keyboard region in described touch control screen module
Out, period described in keyboard region and the position stochastic generation of these 10 symbols of 0-9.
Preferably, it is characterized in that, it is arranged in arrays that numeral in described random array and period are scattered in four row.
Preferably, it is characterized in that, described iris identification device includes:
(1) sampling module, for obtaining, correcting iris image and gather the information of iris image, due to reality acquisition
In approximately the same plane, understand slightly deviation between iris image and the iris image of standard acquisition, need the iris that reality is obtained
Image carries out plane correction, sets image rectification submodule, and the updating formula that described image rectification submodule uses is:
Wherein, and I (x, y)AThe iris image that expression reality obtains, and I (x, y)BRepresent the iris image of standard acquisition, actual
Standard deviation between the iris image and each pixel point value of the iris image of standard acquisition that obtain;
(2) pretreatment module, for positioning the iris image obtained and normalized, it includes that light speckle is filled out
Filling submodule, described smooth speckle is filled submodule and is used for being filled with each hot spot point detected in iris image, fills
The gray value of four the envelope points up and down in the non-spot area that Shi Liyong is adjacent with light speckle calculates the ash of light speckle
Angle value, a light speckle in definition iris image is P0(x0,y0), described four envelope points are followed successively by P1(x1,y1)、P2(x2,
y2)、P3(x3,y3)、P4(x4,y4), the gray value computing formula of definition light speckle is:
Preferably, it is characterized in that, described iris identification device also includes:
(3) feature coding module, for the feature of iris image is extracted and is encoded, including:
A, for the first time LBP operator process submodule: for any point n in iris imagecWith the K in 5 × 5 windows
Pixel is compared to calculate LBP value, and described K pixel is with a ncCentered by be distributed in a ncPeriphery, if ncCoordinate be
(xc,yc), the computing formula of LBP value is:
Wherein, described K pixel is labeled as n0~nK, the span of K is [20,24], 1st-LBP (xc,yc) take
Value scope is [0, K];
B, for the second time LBP operator process submodule, for strengthening described some n on the premise of ensureing code lengthcWith week
Enclosing the relatedness of neighborhood, it is with a nc8 neighborhood territory pixel points as sub-center point, be denoted as nvc0,nvc1,...,nvc7, use 3
× 3 windows, by the average of entire pixels in windowReplace the value of sub-center point, re-use LBP operator to central point
ncCalculating, computing formula is:
C, for the third time LBP operator process submodule, process the square after submodule processes for shortening through second time LBP operator
The feature coding length of shape image, it is with a ncCentered by, according to self-defining function { n in the window of 3 × 3vcj,|nvcj-nc|
=rank4(nvci-nc|, i=0,1 ..., 7), j=0,1,2,3} selects 4 sub-center points to calculate, and computing formula is:
Wherein, rank4(|nvci-nc|, i=0,1 ..., 7) represent 7 | nvci-nc| value arrange from small to large
After take front 4 numbers, nvcjRepresent 4 the sub-center points chosen;
D, the 4th LBP operator process submodule: on the basis of processing after submodule processes at third time LBP operator
Continuing to reduce code length, computing formula is:
After having calculated, output represents the coding of iris image feature;
(4) codes match module, for receiving the coding of described expression iris image feature and by itself and data base
Feature coding is compared, and completes the identification to identity.
Wherein, described pretreatment module also includes:
(1) coarse positioning submodule: fill submodule with light speckle and be connected, is used for carrying out iris image cutting and the most fixed
Position pupil position, during cutting centered by described pupil position, the iris image after filling hot spot cut by the radius of 5 times
Cut;
(2) fine positioning submodule: be connected with coarse positioning submodule, is used for being accurately positioned iris region;
(3) normalization submodule, for being launched into the iris image of fixed resolution by the iris region behind location.
Wherein, described fine positioning submodule includes the downsampling unit being sequentially connected with, first positioning unit and again positions
Unit, described downsampling unit is for carrying out down-sampling to the iris image after cutting, and described first positioning unit is used for passing through
Iris inside and outside circle is positioned by the Canny edge detection operator and the Hough loop truss that improve, and described positioning unit again is used for
It is accurately positioned on iris image with the parameter that first positioning unit positions.
Wherein, the Canny edge detection operator of described improvement is the suppression that vertical direction only carries out non-maximum
Canny edge detection operator.
Wherein, the Canny edge detection operator of described improvement is the Canny limit carrying out strong rim detection only with high threshold
Edge detective operators.
The invention have the benefit that
1, image rectification submodule is set, and defines updating formula, improve the precision of image procossing;
2, light speckle is set and fills submodule, and define the gray value computing formula of light speckle, remain rainbow well
The structural information of film image, the iris image after filling can position effectively;
3, the first positioning unit arranged, its Canny edge detection operator passing through improvement and Hough loop truss are to iris
Inside and outside circle positions, it is simple to the speed positioning and improve iris of iris;
4, the first time LBP operator arranged processes submodule, adds the relatedness of central point and other neighborhood of surrounding, energy
Enough meet the image texture of different scale and frequency;
5, the second time LBP operator arranged processes submodule, for the third time LBP operator and processes submodule and the 4th LBP calculation
Son processes submodule, under not affecting the central point relatedness with surrounding neighbors, constantly reduces code length, has saved storage sky
Between, decrease amount of calculation, improve recognition speed, enhance recognition accuracy, obtain higher robustness.
Accompanying drawing explanation
The invention will be further described to utilize accompanying drawing, but the embodiment in accompanying drawing does not constitute any limit to the present invention
System, for those of ordinary skill in the art, on the premise of not paying creative work, it is also possible to obtain according to the following drawings
Other accompanying drawing.
Fig. 1 is the iris identification device connection diagram of the present invention.
Detailed description of the invention
The invention will be further described with the following Examples.
Embodiment 1
Seeing Fig. 1, a kind of ATM input equipment being identified by iris of the present embodiment, including ATM input equipment
And the iris identification device being connected with the ATM input equipment signal of telecommunication, described ATM input equipment includes:
Housing;
Touch control screen module includes a chip, in housing described in described touch control screen Module-embedding;
Randomized blocks includes the set of number of 0-9 and period in order to randomly generate, and is defined as random array, described random number
Period in group and each numeral in 0-9 at least occur once;
Processor is in order to process data, and described processor is connected to described randomized blocks and described touch-control by wire
The described chip of formula screen module.
Preferably, it is characterized in that, above-mentioned hookup mechanism has:
A keyboard region, described random array is provided the most all to show in described keyboard region in described touch control screen module
Out, period described in keyboard region and the position stochastic generation of these 10 symbols of 0-9.
Preferably, it is characterized in that, it is arranged in arrays that numeral in described random array and period are scattered in four row.
Preferably, it is characterized in that, described iris identification device includes:
(1) sampling module, for obtaining iris image and gathering the information of iris image;
(2) pretreatment module, for obtaining, correcting iris image and gather the information of iris image, owing to reality obtains
Iris image and the iris image of standard acquisition between in approximately the same plane can slightly deviation, need the rainbow that reality is obtained
Film image carries out plane correction, sets image rectification submodule, and the updating formula that described image rectification submodule uses is:
Wherein, and I (x, y)AThe iris image that expression reality obtains, and I (x, y)BRepresent the iris image of standard acquisition, actual
Standard deviation between the iris image and each pixel point value of the iris image of standard acquisition that obtain;
Preferably, it is characterized in that, described iris identification device also includes:
(3) feature coding module, for the feature of iris image is extracted and is encoded, including:
A, for the first time LBP operator process submodule: for any point n in iris imagecWith 20 in 5 × 5 windows
Individual pixel is compared to calculate LBP value, and described 20 pixels are with a ncCentered by be distributed in a ncPeriphery, if ncSeat
It is designated as (xc,yc), the computing formula of LBP value is:
Wherein, described 20 pixels are labeled as n0~n20, 1st-LBP (xc,yc) span be [0,20];
B, for the second time LBP operator process submodule, for strengthening described some n on the premise of ensureing code lengthcWith week
Enclosing the relatedness of neighborhood, it is with a nc8 neighborhood territory pixel points as sub-center point, be denoted as nvc0,nvc1,...,nvc7, use 3
× 3 windows, by the average of entire pixels in windowReplace the value of sub-center point, re-use LBP operator to central point
ncCalculating, computing formula is:
C, for the third time LBP operator process submodule, process the square after submodule processes for shortening through second time LBP operator
The feature coding length of shape image, it is with a ncCentered by, according to self-defining function { n in the window of 3 × 3vcj,|nvcj-nc|
=rank4(nvci-nc|, i=0,1 ..., 7), j=0,1,2,3} selects 4 sub-center points to calculate, and computing formula is:
Wherein, rank4(|nvci-nc|, i=0,1 ..., 7) represent 7 | nvci-nc| value arrange from small to large
After take front 4 numbers, nvcjRepresent 4 the sub-center points chosen;
D, the 4th LBP operator process submodule: on the basis of processing after submodule processes at third time LBP operator
Continuing to reduce code length, computing formula is:
After having calculated, output represents the coding of iris image feature;
(4) codes match module, for receiving the coding of described expression iris image feature and by itself and data base
Feature coding is compared, and completes the identification to identity.
Wherein, described pretreatment module includes:
(1) light speckle fills submodule: for being filled with each hot spot point detected in iris image, during filling
The gray value utilizing four the envelope points up and down in the non-spot area adjacent with light speckle calculates the gray scale of light speckle
Value, a light speckle in definition iris image is P0(x0,y0), described four envelope points are followed successively by P1(x1,y1)、P2(x2,
y2)、P3(x3,y3)、P4(x4,y4), the gray value computing formula of definition light speckle is:
(2) coarse positioning submodule: fill submodule with light speckle and be connected, is used for carrying out iris image cutting and the most fixed
Position pupil position, during cutting centered by described pupil position, the iris image after filling hot spot cut by the radius of 5 times
Cut;
(3) fine positioning submodule: be connected with coarse positioning submodule, is used for being accurately positioned iris region;
(4) normalization submodule, for being launched into the iris image of fixed resolution by the iris region behind location.
Wherein, described fine positioning submodule includes the downsampling unit being sequentially connected with, first positioning unit and again positions
Unit, described downsampling unit is for carrying out down-sampling to the iris image after cutting, and described first positioning unit is used for passing through
Iris inside and outside circle is positioned by the Canny edge detection operator and the Hough loop truss that improve, and described positioning unit again is used for
It is accurately positioned on iris image with the parameter that first positioning unit positions.
Wherein, the Canny edge detection operator of described improvement is the suppression that vertical direction only carries out non-maximum
Canny edge detection operator.
Wherein, the Canny edge detection operator of described improvement is the Canny limit carrying out strong rim detection only with high threshold
Edge detective operators.
The present embodiment arranges light speckle and fills submodule, remains the structural information of iris image well, after filling
Iris image can position effectively;Arrange first positioning unit, its by improve Canny edge detection operator and
Iris inside and outside circle is positioned by Hough loop truss, it is simple to the speed positioning and improve iris of iris;The first time arranged
LBP operator processes submodule, adds the relatedness of central point and other neighborhood of surrounding, it is possible to meet different scale and frequency
Image texture;The second time LBP operator arranged processes submodule, for the third time LBP operator and processes submodule and the 4th LBP operator
Process submodule, under not affecting the central point relatedness with surrounding neighbors, constantly reduce code length, save storage sky
Between, decrease amount of calculation, improve recognition speed, enhance recognition accuracy, obtained higher robustness, use CASIA
When V1.0 iris storehouse is tested, result is as follows:
Embodiment 2
Seeing Fig. 1, a kind of ATM input equipment being identified by iris of the present embodiment, including ATM input equipment
And the iris identification device being connected with the ATM input equipment signal of telecommunication, described ATM input equipment includes:
Housing;
Touch control screen module includes a chip, in housing described in described touch control screen Module-embedding;
Randomized blocks includes the set of number of 0-9 and period in order to randomly generate, and is defined as random array, described random number
Period in group and each numeral in 0-9 at least occur once;
Processor is in order to process data, and described processor is connected to described randomized blocks and described touch-control by wire
The described chip of formula screen module.
Preferably, it is characterized in that, above-mentioned hookup mechanism has:
A keyboard region, described random array is provided the most all to show in described keyboard region in described touch control screen module
Out, period described in keyboard region and the position stochastic generation of these 10 symbols of 0-9.
Preferably, it is characterized in that, it is arranged in arrays that numeral in described random array and period are scattered in four row.
Preferably, it is characterized in that, described iris identification device includes:
(1) sampling module, for obtaining iris image and gathering the information of iris image;
(2) pretreatment module, for obtaining, correcting iris image and gather the information of iris image, owing to reality obtains
Iris image and the iris image of standard acquisition between in approximately the same plane can slightly deviation, need the rainbow that reality is obtained
Film image carries out plane correction, sets image rectification submodule, and the updating formula that described image rectification submodule uses is:
Wherein, and I (x, y)AThe iris image that expression reality obtains, and I (x, y)BRepresent the iris image of standard acquisition, actual
Standard deviation between the iris image and each pixel point value of the iris image of standard acquisition that obtain;
Preferably, it is characterized in that, described iris identification device also includes:
(3) feature coding module, for the feature of iris image is extracted and is encoded, including:
A, for the first time LBP operator process submodule: for any point n in iris imagecWith 21 in 5 × 5 windows
Individual pixel is compared to calculate LBP value, and described 21 pixels are with a ncCentered by be distributed in a ncPeriphery, if ncSeat
It is designated as (xc,yc), the computing formula of LBP value is:
Wherein, described 21 pixels are labeled as n0~n21, 1st-LBP (xc,yc) span be [0,21];
B, for the second time LBP operator process submodule, for strengthening described some n on the premise of ensureing code lengthcWith week
Enclosing the relatedness of neighborhood, it is with a nc8 neighborhood territory pixel points as sub-center point, be denoted as nvc0,nvc1,...,nvc7, use 3
× 3 windows, by the average of entire pixels in windowReplace the value of sub-center point, re-use LBP operator to central point
ncCalculating, computing formula is:
C, for the third time LBP operator process submodule, process the square after submodule processes for shortening through second time LBP operator
The feature coding length of shape image, it is with a ncCentered by, according to self-defining function { n in the window of 3 × 3vcj,|nvcj-nc|
=rank4(|nvci-nc|, i=0,1 ..., 7), j=0,1,2,3} selects 4 sub-center points to calculate, and computing formula is:
Wherein, rank4(|nvci-nc|, i=0,1 ..., 7) represent 7 | nvci-nc| value arrange from small to large
After take front 4 numbers, nvcjRepresent 4 the sub-center points chosen;
D, the 4th LBP operator process submodule: on the basis of processing after submodule processes at third time LBP operator
Continuing to reduce code length, computing formula is:
After having calculated, output represents the coding of iris image feature;
(4) codes match module, for receiving the coding of described expression iris image feature and by itself and data base
Feature coding is compared, and completes the identification to identity.
Wherein, described pretreatment module includes:
(1) light speckle fills submodule: for being filled with each hot spot point detected in iris image, during filling
The gray value utilizing four the envelope points up and down in the non-spot area adjacent with light speckle calculates the gray scale of light speckle
Value, a light speckle in definition iris image is P0(x0,y0), described four envelope points are followed successively by P1(x1,y1)、P2(x2,
y2)、P3(x3,y3)、P4(x4,y4), the gray value computing formula of definition light speckle is:
(2) coarse positioning submodule: fill submodule with light speckle and be connected, is used for carrying out iris image cutting and the most fixed
Position pupil position, during cutting centered by described pupil position, the iris image after filling hot spot cut by the radius of 5 times
Cut;
(3) fine positioning submodule: be connected with coarse positioning submodule, is used for being accurately positioned iris region;
(4) normalization submodule, for being launched into the iris image of fixed resolution by the iris region behind location.
Wherein, described fine positioning submodule includes the downsampling unit being sequentially connected with, first positioning unit and again positions
Unit, described downsampling unit is for carrying out down-sampling to the iris image after cutting, and described first positioning unit is used for passing through
Iris inside and outside circle is positioned by the Canny edge detection operator and the Hough loop truss that improve, and described positioning unit again is used for
It is accurately positioned on iris image with the parameter that first positioning unit positions.
Wherein, the Canny edge detection operator of described improvement is the suppression that vertical direction only carries out non-maximum
Canny edge detection operator.
Wherein, the Canny edge detection operator of described improvement is the Canny limit carrying out strong rim detection only with high threshold
Edge detective operators.
The present embodiment arranges light speckle and fills submodule, remains the structural information of iris image well, after filling
Iris image can position effectively;Arrange first positioning unit, its by improve Canny edge detection operator and
Iris inside and outside circle is positioned by Hough loop truss, it is simple to the speed positioning and improve iris of iris;The first time arranged
LBP operator processes submodule, adds the relatedness of central point and other neighborhood of surrounding, it is possible to meet different scale and frequency
Image texture;The second time LBP operator arranged processes submodule, for the third time LBP operator and processes submodule and the 4th LBP operator
Process submodule, under not affecting the central point relatedness with surrounding neighbors, constantly reduce code length, save storage sky
Between, decrease amount of calculation, improve recognition speed, enhance recognition accuracy, obtained higher robustness, use CASIA
When V1.0 iris storehouse is tested, result is as follows:
Embodiment 3
Seeing Fig. 1, a kind of ATM input equipment being identified by iris of the present embodiment, including ATM input equipment
And the iris identification device being connected with the ATM input equipment signal of telecommunication, described ATM input equipment includes:
Housing;
Touch control screen module includes a chip, in housing described in described touch control screen Module-embedding;
Randomized blocks includes the set of number of 0-9 and period in order to randomly generate, and is defined as random array, described random number
Period in group and each numeral in 0-9 at least occur once;
Processor is in order to process data, and described processor is connected to described randomized blocks and described touch-control by wire
The described chip of formula screen module.
Preferably, it is characterized in that, above-mentioned hookup mechanism has:
A keyboard region, described random array is provided the most all to show in described keyboard region in described touch control screen module
Out, period described in keyboard region and the position stochastic generation of these 10 symbols of 0-9.
Preferably, it is characterized in that, it is arranged in arrays that numeral in described random array and period are scattered in four row.
Preferably, it is characterized in that, described iris identification device includes:
(1) sampling module, for obtaining, correcting iris image and gather the information of iris image, due to reality acquisition
In approximately the same plane, understand slightly deviation between iris image and the iris image of standard acquisition, need the iris that reality is obtained
Image carries out plane correction, sets image rectification submodule, and the updating formula that described image rectification submodule uses is:
Wherein, and I (x, y)AThe iris image that expression reality obtains, and I (x, y)BRepresent the iris image of standard acquisition, actual
Standard deviation between the iris image and each pixel point value of the iris image of standard acquisition that obtain;
(2) pretreatment module, for positioning and normalized the iris image obtained;
Preferably, it is characterized in that, described iris identification device also includes:
(3) feature coding module, for the feature of iris image is extracted and is encoded, including:
A, for the first time LBP operator process submodule: for any point n in iris imagecWith 22 in 5 × 5 windows
Individual pixel is compared to calculate LBP value, and described 22 pixels are with a ncCentered by be distributed in a ncPeriphery, if ncSeat
It is designated as (xc,yc), the computing formula of LBP value is:
Wherein, described 22 pixels are labeled as n0~n21, 1st-LBP (xc,yc) span be [0,22];
B, for the second time LBP operator process submodule, for strengthening described some n on the premise of ensureing code lengthcWith week
Enclosing the relatedness of neighborhood, it is with a nc8 neighborhood territory pixel points as sub-center point, be denoted as nvc0,nvc1,...,nvc7, use 3
× 3 windows, by the average of entire pixels in windowReplace the value of sub-center point, re-use LBP operator to central point
ncCalculating, computing formula is:
C, for the third time LBP operator process submodule, process the square after submodule processes for shortening through second time LBP operator
The feature coding length of shape image, it is with a ncCentered by, according to self-defining function { n in the window of 3 × 3vcj,|nvcj-nc|
=rank4(nvci-nc|, i=0,1 ..., 7), j=0,1,2,3} selects 4 sub-center points to calculate, and computing formula is:
Wherein, rank4(|nvci-nc|, i=0,1 ..., 7) represent 7 | nvci-nc| value arrange from small to large
After take front 4 numbers, nvcjRepresent 4 the sub-center points chosen;
D, the 4th LBP operator process submodule: on the basis of processing after submodule processes at third time LBP operator
Continuing to reduce code length, computing formula is:
After having calculated, output represents the coding of iris image feature;
(4) codes match module, for receiving the coding of described expression iris image feature and by itself and data base
Feature coding is compared, and completes the identification to identity.
Wherein, described pretreatment module includes:
(1) light speckle fills submodule: for being filled with each hot spot point detected in iris image, during filling
The gray value utilizing four the envelope points up and down in the non-spot area adjacent with light speckle calculates the gray scale of light speckle
Value, a light speckle in definition iris image is P0(x0,y0), described four envelope points are followed successively by P1(x1,y1)、P2(x2,
y2)、P3(x3,y3)、P4(x4,y4), the gray value computing formula of definition light speckle is:
(2) coarse positioning submodule: fill submodule with light speckle and be connected, is used for carrying out iris image cutting and the most fixed
Position pupil position, during cutting centered by described pupil position, the iris image after filling hot spot cut by the radius of 5 times
Cut;
(3) fine positioning submodule: be connected with coarse positioning submodule, is used for being accurately positioned iris region;
(4) normalization submodule, for being launched into the iris image of fixed resolution by the iris region behind location.
Wherein, described fine positioning submodule includes the downsampling unit being sequentially connected with, first positioning unit and again positions
Unit, described downsampling unit is for carrying out down-sampling to the iris image after cutting, and described first positioning unit is used for passing through
Iris inside and outside circle is positioned by the Canny edge detection operator and the Hough loop truss that improve, and described positioning unit again is used for
It is accurately positioned on iris image with the parameter that first positioning unit positions.
Wherein, the Canny edge detection operator of described improvement is the suppression that vertical direction only carries out non-maximum
Canny edge detection operator.
Wherein, the Canny edge detection operator of described improvement is the Canny limit carrying out strong rim detection only with high threshold
Edge detective operators.
The present embodiment arranges light speckle and fills submodule, remains the structural information of iris image well, after filling
Iris image can position effectively;Arrange first positioning unit, its by improve Canny edge detection operator and
Iris inside and outside circle is positioned by Hough loop truss, it is simple to the speed positioning and improve iris of iris;The first time arranged
LBP operator processes submodule, adds the relatedness of central point and other neighborhood of surrounding, it is possible to meet different scale and frequency
Image texture;The second time LBP operator arranged processes submodule, for the third time LBP operator and processes submodule and the 4th LBP operator
Process submodule, under not affecting the central point relatedness with surrounding neighbors, constantly reduce code length, save storage sky
Between, decrease amount of calculation, improve recognition speed, enhance recognition accuracy, obtained higher robustness, use CASIA
When V1.0 iris storehouse is tested, result is as follows:
Embodiment 4
Seeing Fig. 1, a kind of ATM input equipment being identified by iris of the present embodiment, including ATM input equipment
And the iris identification device being connected with the ATM input equipment signal of telecommunication, described ATM input equipment includes:
Housing;
Touch control screen module includes a chip, in housing described in described touch control screen Module-embedding;
Randomized blocks includes the set of number of 0-9 and period in order to randomly generate, and is defined as random array, described random number
Period in group and each numeral in 0-9 at least occur once;
Processor is in order to process data, and described processor is connected to described randomized blocks and described touch-control by wire
The described chip of formula screen module.
Preferably, it is characterized in that, above-mentioned hookup mechanism has:
A keyboard region, described random array is provided the most all to show in described keyboard region in described touch control screen module
Out, period described in keyboard region and the position stochastic generation of these 10 symbols of 0-9.
Preferably, it is characterized in that, it is arranged in arrays that numeral in described random array and period are scattered in four row.
Preferably, it is characterized in that, described iris identification device includes:
(1) sampling module, for obtaining, correcting iris image and gather the information of iris image, due to reality acquisition
In approximately the same plane, understand slightly deviation between iris image and the iris image of standard acquisition, need the iris that reality is obtained
Image carries out plane correction, sets image rectification submodule, and the updating formula that described image rectification submodule uses is:
Wherein, and I (x, y)AThe iris image that expression reality obtains, and I (x, y)BRepresent the iris image of standard acquisition, actual
Standard deviation between the iris image and each pixel point value of the iris image of standard acquisition that obtain;
(2) pretreatment module, for positioning and normalized the iris image obtained;
Preferably, it is characterized in that, described iris identification device also includes:
(3) feature coding module, for the feature of iris image is extracted and is encoded, including:
A, for the first time LBP operator process submodule: for any point n in iris imagecWith 23 in 5 × 5 windows
Individual pixel is compared to calculate LBP value, and described 23 pixels are with a ncCentered by be distributed in a ncPeriphery, if ncSeat
It is designated as (xc,yc), the computing formula of LBP value is:
Wherein, described 23 pixels are labeled as n0~n21, 1st-LBP (xc,yc) span be [0,23];
B, for the second time LBP operator process submodule, for strengthening described some n on the premise of ensureing code lengthcWith week
Enclosing the relatedness of neighborhood, it is with a nc8 neighborhood territory pixel points as sub-center point, be denoted as nvc0,nvc1,...,nvc7, use 3
× 3 windows, by the average of entire pixels in windowReplace the value of sub-center point, re-use LBP operator to central point
ncCalculating, computing formula is:
C, for the third time LBP operator process submodule, process the square after submodule processes for shortening through second time LBP operator
The feature coding length of shape image, it is with a ncCentered by, according to self-defining function { n in the window of 3 × 3vcj,|nvcj-nc|
=rank4(nvci-nc|, i=0,1 ..., 7), j=0,1,2,3} selects 4 sub-center points to calculate, and computing formula is:
Wherein, rank4(|nvci-nc|, i=0,1 ..., 7) represent 7 | nvci-nc| value arrange from small to large
After take front 4 numbers, nvcjRepresent 4 the sub-center points chosen;
D, the 4th LBP operator process submodule: on the basis of processing after submodule processes at third time LBP operator
Continuing to reduce code length, computing formula is:
After having calculated, output represents the coding of iris image feature;
(4) codes match module, for receiving the coding of described expression iris image feature and by itself and data base
Feature coding is compared, and completes the identification to identity.
Wherein, described pretreatment module includes:
(1) light speckle fills submodule: for being filled with each hot spot point detected in iris image, during filling
The gray value utilizing four the envelope points up and down in the non-spot area adjacent with light speckle calculates the gray scale of light speckle
Value, a light speckle in definition iris image is P0(x0,y0), described four envelope points are followed successively by P1(x1,y1)、P2(x2,
y2)、P3(x3,y3)、P4(x4,y4), the gray value computing formula of definition light speckle is:
(2) coarse positioning submodule: fill submodule with light speckle and be connected, is used for carrying out iris image cutting and the most fixed
Position pupil position, during cutting centered by described pupil position, the iris image after filling hot spot cut by the radius of 5 times
Cut;
(3) fine positioning submodule: be connected with coarse positioning submodule, is used for being accurately positioned iris region;
(4) normalization submodule, for being launched into the iris image of fixed resolution by the iris region behind location.
Wherein, described fine positioning submodule includes the downsampling unit being sequentially connected with, first positioning unit and again positions
Unit, described downsampling unit is for carrying out down-sampling to the iris image after cutting, and described first positioning unit is used for passing through
Iris inside and outside circle is positioned by the Canny edge detection operator and the Hough loop truss that improve, and described positioning unit again is used for
It is accurately positioned on iris image with the parameter that first positioning unit positions.
Wherein, the Canny edge detection operator of described improvement is the suppression that vertical direction only carries out non-maximum
Canny edge detection operator.
Wherein, the Canny edge detection operator of described improvement is the Canny limit carrying out strong rim detection only with high threshold
Edge detective operators.
The present embodiment arranges light speckle and fills submodule, remains the structural information of iris image well, after filling
Iris image can position effectively;Arrange first positioning unit, its by improve Canny edge detection operator and
Iris inside and outside circle is positioned by Hough loop truss, it is simple to the speed positioning and improve iris of iris;The first time arranged
LBP operator processes submodule, adds the relatedness of central point and other neighborhood of surrounding, it is possible to meet different scale and frequency
Image texture;The second time LBP operator arranged processes submodule, for the third time LBP operator and processes submodule and the 4th LBP operator
Process submodule, under not affecting the central point relatedness with surrounding neighbors, constantly reduce code length, save storage sky
Between, decrease amount of calculation, improve recognition speed, enhance recognition accuracy, obtained higher robustness, use CASIA
When V1.0 iris storehouse is tested, result is as follows:
Embodiment 5
Seeing Fig. 1, a kind of ATM input equipment being identified by iris of the present embodiment, including ATM input equipment
And the iris identification device being connected with the ATM input equipment signal of telecommunication, described ATM input equipment includes:
Housing;
Touch control screen module includes a chip, in housing described in described touch control screen Module-embedding;
Randomized blocks includes the set of number of 0-9 and period in order to randomly generate, and is defined as random array, described random number
Period in group and each numeral in 0-9 at least occur once;
Processor is in order to process data, and described processor is connected to described randomized blocks and described touch-control by wire
The described chip of formula screen module.
Preferably, it is characterized in that, above-mentioned hookup mechanism has:
A keyboard region, described random array is provided the most all to show in described keyboard region in described touch control screen module
Out, period described in keyboard region and the position stochastic generation of these 10 symbols of 0-9.
Preferably, it is characterized in that, it is arranged in arrays that numeral in described random array and period are scattered in four row.
Preferably, it is characterized in that, described iris identification device includes:
(1) sampling module, for obtaining, correcting iris image and gather the information of iris image, due to reality acquisition
In approximately the same plane, understand slightly deviation between iris image and the iris image of standard acquisition, need the iris that reality is obtained
Image carries out plane correction, sets image rectification submodule, and the updating formula that described image rectification submodule uses is:
Wherein, and I (x, y)AThe iris image that expression reality obtains, and I (x, y)BRepresent the iris image of standard acquisition, actual
Standard deviation between the iris image and each pixel point value of the iris image of standard acquisition that obtain;
(2) pretreatment module, for positioning and normalized the iris image obtained;
Preferably, it is characterized in that, described iris identification device also includes:
(3) feature coding module, for the feature of iris image is extracted and is encoded, including:
A, for the first time LBP operator process submodule: for any point n in iris imagecWith 24 in 5 × 5 windows
Individual pixel is compared to calculate LBP value, and described 24 pixels are with a ncCentered by be distributed in a ncPeriphery, if ncSeat
It is designated as (xc,yc), the computing formula of LBP value is:
Wherein, described 24 pixels are labeled as n0~n21, 1st-LBP (xc,yc) span be [0,24];
B, for the second time LBP operator process submodule, for strengthening described some n on the premise of ensureing code lengthcWith week
Enclosing the relatedness of neighborhood, it is with a nc8 neighborhood territory pixel points as sub-center point, be denoted as nvc0,nvc1,...,nvc7, use 3
× 3 windows, by the average of entire pixels in windowReplace the value of sub-center point, re-use LBP operator to central point
ncCalculating, computing formula is:
C, for the third time LBP operator process submodule, process the square after submodule processes for shortening through second time LBP operator
The feature coding length of shape image, it is with a ncCentered by, according to self-defining function { n in the window of 3 × 3vcj,|nvcj-nc|
=rank4(|nvci-nc|, i=0,1 ..., 7), j=0,1,2,3} selects 4 sub-center points to calculate, and computing formula is:
Wherein, rank4(|nvci-nc|, i=0,1 ..., 7) represent 7 | nvci-nc| value arrange from small to large
After take front 4 numbers, nvcjRepresent 4 the sub-center points chosen;
D, the 4th LBP operator process submodule: on the basis of processing after submodule processes at third time LBP operator
Continuing to reduce code length, computing formula is:
After having calculated, output represents the coding of iris image feature;
(4) codes match module, for receiving the coding of described expression iris image feature and by itself and data base
Feature coding is compared, and completes the identification to identity.
Wherein, described pretreatment module includes:
(1) light speckle fills submodule: for being filled with each hot spot point detected in iris image, during filling
The gray value utilizing four the envelope points up and down in the non-spot area adjacent with light speckle calculates the gray scale of light speckle
Value, a light speckle in definition iris image is P0(x0,y0), described four envelope points are followed successively by P1(x1,y1)、P2(x2,
y2)、P3(x3,y3)、P4(x4,y4), the gray value computing formula of definition light speckle is:
(2) coarse positioning submodule: fill submodule with light speckle and be connected, is used for carrying out iris image cutting and the most fixed
Position pupil position, during cutting centered by described pupil position, the iris image after filling hot spot cut by the radius of 5 times
Cut;
(3) fine positioning submodule: be connected with coarse positioning submodule, is used for being accurately positioned iris region;
(4) normalization submodule, for being launched into the iris image of fixed resolution by the iris region behind location.
Wherein, described fine positioning submodule includes the downsampling unit being sequentially connected with, first positioning unit and again positions
Unit, described downsampling unit is for carrying out down-sampling to the iris image after cutting, and described first positioning unit is used for passing through
Iris inside and outside circle is positioned by the Canny edge detection operator and the Hough loop truss that improve, and described positioning unit again is used for
It is accurately positioned on iris image with the parameter that first positioning unit positions.
Wherein, the Canny edge detection operator of described improvement is the suppression that vertical direction only carries out non-maximum
Canny edge detection operator.
Wherein, the Canny edge detection operator of described improvement is the Canny limit carrying out strong rim detection only with high threshold
Edge detective operators.
The present embodiment arranges light speckle and fills submodule, remains the structural information of iris image well, after filling
Iris image can position effectively;Arrange first positioning unit, its by improve Canny edge detection operator and
Iris inside and outside circle is positioned by Hough loop truss, it is simple to the speed positioning and improve iris of iris;The first time arranged
LBP operator processes submodule, adds the relatedness of central point and other neighborhood of surrounding, it is possible to meet different scale and frequency
Image texture;The second time LBP operator arranged processes submodule, for the third time LBP operator and processes submodule and the 4th LBP operator
Process submodule, under not affecting the central point relatedness with surrounding neighbors, constantly reduce code length, save storage sky
Between, decrease amount of calculation, improve recognition speed, enhance recognition accuracy, obtained higher robustness, use CASIA
When V1.0 iris storehouse is tested, result is as follows:
Last it should be noted that, above example is only in order to illustrate technical scheme, rather than the present invention is protected
Protecting the restriction of scope, although having made to explain to the present invention with reference to preferred embodiment, those of ordinary skill in the art should
Work as understanding, technical scheme can be modified or equivalent, without deviating from the reality of technical solution of the present invention
Matter and scope.
Claims (9)
1. the ATM input equipment being identified by iris, including ATM input equipment and with ATM input equipment
The iris identification device that the signal of telecommunication connects, described ATM input equipment includes:
Housing;
Touch control screen module includes a chip, in housing described in described touch control screen Module-embedding;
Randomized blocks includes the set of number of 0-9 and period in order to randomly generate, and is defined as random array, in described random array
Period and 0-9 in each numeral at least occur once;
Processor is in order to process data, and described processor is connected to described randomized blocks and described touch screen by wire
The described chip of curtain module.
A kind of ATM input equipment being identified by iris the most according to claim 1, be is characterized in that, above-mentioned company
Knot device has:
A keyboard region, described random array is provided the most all to demonstrate in described keyboard region in described touch control screen module
Come, period described in keyboard region and the position stochastic generation of these 10 symbols of 0-9.
A kind of ATM input equipment being identified by iris the most according to claim 2, be is characterized in that, described with
It is arranged in arrays that numeral in machine array and period are scattered in four row.
A kind of ATM input equipment being identified by iris the most according to claim 3, be is characterized in that, described rainbow
Film evaluator includes:
(1) sampling module, for obtaining, correcting iris image and gather the information of iris image, the iris obtained due to reality
In approximately the same plane, understand slightly deviation between image and the iris image of standard acquisition, need the iris image that reality is obtained
Carrying out plane correction, set image rectification submodule, the updating formula that described image rectification submodule uses is:
Wherein, and I (x, y)AThe iris image that expression reality obtains, and I (x, y)BRepresent the iris image of standard acquisition, actual acquisition
Iris image and standard acquisition iris image each pixel point value between standard deviation;
(2) pretreatment module, for positioning the iris image obtained and normalized, it includes that light speckle fills son
Module, described smooth speckle is filled submodule and is used for being filled with each hot spot point detected in iris image, sharp during filling
The gray value of light speckle is calculated with the gray value of four the envelope points up and down in the non-spot area adjacent with light speckle,
A light speckle in definition iris image is P0(x0,y0), described four envelope points are followed successively by P1(x1,y1)、P2(x2,y2)、P3
(x3,y3)、P4(x4,y4), the gray value computing formula of definition light speckle is:
A kind of ATM input equipment being identified by iris the most according to claim 4, be is characterized in that, described rainbow
Film evaluator also includes:
(3) feature coding module, for the feature of iris image is extracted and is encoded, including:
A, for the first time LBP operator process submodule: for any point n in iris imagecWith K pixel in 5 × 5 windows
Point is compared to calculate LBP value, and described K pixel is with a ncCentered by be distributed in a ncPeriphery, if ncCoordinate be (xc,
yc), the computing formula of LBP value is:
Wherein, described K pixel is labeled as n0~nK, the span of K is [20,24], 1st-LBP (xc, yc) value model
Enclose for [0, K];
B, for the second time LBP operator process submodule, for strengthening described some n on the premise of ensureing code lengthcWith surrounding neighbors
Relatedness, it is with a nc8 neighborhood territory pixel points as sub-center point, be denoted as nvc0,nvc1,...,nvc7, use 3 × 3 windows,
By the average of entire pixels in windowReplace the value of sub-center point, re-use LBP operator to central point ncCarry out
Calculating, computing formula is:
C, for the third time LBP operator process submodule, process the iris figure after submodule processes for shortening through second time LBP operator
The feature coding length of picture, it is with a ncCentered by, according to self-defining function { n in the window of 3 × 3vcj,|nvcj-nc|=
rank4(|nvci-nc|, i=0,1 ..., 7), j=0,1,2,3} selects 4 sub-center points to calculate, and computing formula is:
Wherein, rank4(|nvci-nc|, i=0,1 ..., 7) represent 7 | nvci-nc| value arrange from small to large after take
Front 4 numbers, nvcjRepresent 4 the sub-center points chosen;
D, the 4th LBP operator process submodule: continue on the basis of processing after submodule processes at third time LBP operator
Reducing code length, computing formula is:
After having calculated, output represents the coding of iris image feature;
(4) codes match module, for receiving the coding of described expression iris image feature and by itself and the feature in data base
Coding is compared, and completes the identification to identity.
A kind of ATM input equipment being identified by iris the most according to claim 5, be is characterized in that, described pre-
Processing module also includes:
(1) coarse positioning submodule: fill submodule with light speckle and be connected, for carrying out cutting Primary Location pupil to iris image
Hole site, during cutting centered by described pupil position, the iris image after filling hot spot cuts by the radius of 5 times;
(2) fine positioning submodule: be connected with coarse positioning submodule, is used for being accurately positioned iris region;
(3) normalization submodule, for being launched into the iris image of fixed resolution by the iris region behind location.
A kind of ATM input equipment being identified by iris the most according to claim 6, be is characterized in that, described essence
Locator module includes the downsampling unit being sequentially connected with, first positioning unit and positioning unit, described downsampling unit again
For the iris image after cutting is carried out down-sampling, described first positioning unit is for the Canny rim detection by improving
Iris inside and outside circle is positioned by operator and Hough loop truss, and described positioning unit again is for positioning with first positioning unit
Parameter be accurately positioned on iris image.
A kind of ATM input equipment being identified by iris the most according to claim 7, be is characterized in that, described in change
The Canny edge detection operator entered is the Canny edge detection operator of the suppression that vertical direction only carries out non-maximum.
A kind of ATM input equipment being identified by iris the most according to claim 8, be is characterized in that, described in change
The Canny edge detection operator entered is the Canny edge detection operator carrying out strong rim detection only with high threshold.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610548331.8A CN106204958B (en) | 2016-07-08 | 2016-07-08 | A kind of ATM machine input unit being identified by iris |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610548331.8A CN106204958B (en) | 2016-07-08 | 2016-07-08 | A kind of ATM machine input unit being identified by iris |
Publications (2)
Publication Number | Publication Date |
---|---|
CN106204958A true CN106204958A (en) | 2016-12-07 |
CN106204958B CN106204958B (en) | 2018-10-02 |
Family
ID=57477192
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610548331.8A Expired - Fee Related CN106204958B (en) | 2016-07-08 | 2016-07-08 | A kind of ATM machine input unit being identified by iris |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106204958B (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107273834A (en) * | 2017-06-06 | 2017-10-20 | 宋友澂 | A kind of iris identification method and identifier |
CN107424347A (en) * | 2017-04-12 | 2017-12-01 | 深圳大图科创技术开发有限公司 | A kind of ATM cash withdrawal system based on iris recognition |
CN107609370A (en) * | 2017-08-29 | 2018-01-19 | 深圳怡化电脑股份有限公司 | A kind of identification system, method and its device |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2012108796A (en) * | 2010-11-18 | 2012-06-07 | Hitachi Omron Terminal Solutions Corp | Automatic transaction system, automatic transaction method and automatic transaction machine |
CN104063965A (en) * | 2013-11-25 | 2014-09-24 | 苏州天鸣信息科技有限公司 | Input device of ATM (automatic teller machine) |
CN104281316A (en) * | 2013-07-12 | 2015-01-14 | 三星电子株式会社 | Input device for electronic device and input method using the same |
-
2016
- 2016-07-08 CN CN201610548331.8A patent/CN106204958B/en not_active Expired - Fee Related
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2012108796A (en) * | 2010-11-18 | 2012-06-07 | Hitachi Omron Terminal Solutions Corp | Automatic transaction system, automatic transaction method and automatic transaction machine |
CN104281316A (en) * | 2013-07-12 | 2015-01-14 | 三星电子株式会社 | Input device for electronic device and input method using the same |
CN104063965A (en) * | 2013-11-25 | 2014-09-24 | 苏州天鸣信息科技有限公司 | Input device of ATM (automatic teller machine) |
Non-Patent Citations (1)
Title |
---|
李欢利: "虹膜特征表达与识别算法研究", 《中国博士学位论文全文数据库 信息科技辑》 * |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107424347A (en) * | 2017-04-12 | 2017-12-01 | 深圳大图科创技术开发有限公司 | A kind of ATM cash withdrawal system based on iris recognition |
CN107424347B (en) * | 2017-04-12 | 2018-12-21 | 深圳市创智成科技股份有限公司 | A kind of ATM cash withdrawal system based on iris recognition |
CN107273834A (en) * | 2017-06-06 | 2017-10-20 | 宋友澂 | A kind of iris identification method and identifier |
CN107609370A (en) * | 2017-08-29 | 2018-01-19 | 深圳怡化电脑股份有限公司 | A kind of identification system, method and its device |
Also Published As
Publication number | Publication date |
---|---|
CN106204958B (en) | 2018-10-02 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109934121B (en) | Orchard pedestrian detection method based on YOLOv3 algorithm | |
EP3702957A1 (en) | Target detection method and apparatus, and computer device | |
CN109146892A (en) | A kind of image cropping method and device based on aesthetics | |
CN111079683A (en) | Remote sensing image cloud and snow detection method based on convolutional neural network | |
CN111985374B (en) | Face positioning method and device, electronic equipment and storage medium | |
CN106204958A (en) | A kind of ATM input equipment being identified by iris | |
CN115393597B (en) | Semantic segmentation method and device based on pulse neural network and laser radar point cloud | |
KR20220125719A (en) | Method and equipment for training target detection model, method and equipment for detection of target object, electronic equipment, storage medium and computer program | |
CN111507359A (en) | Self-adaptive weighting fusion method of image feature pyramid | |
CN115861823B (en) | Remote sensing change detection method and device based on self-supervision deep learning | |
CN116994140A (en) | Cultivated land extraction method, device, equipment and medium based on remote sensing image | |
CN112989995A (en) | Text detection method and device and electronic equipment | |
CN116309348A (en) | Lunar south pole impact pit detection method based on improved TransUnet network | |
CN116778346B (en) | Pipeline identification method and system based on improved self-attention mechanism | |
CN107886093B (en) | Character detection method, system, equipment and computer storage medium | |
CN106248070A (en) | A kind of navigator started based on iris identification | |
CN116704324A (en) | Target detection method, system, equipment and storage medium based on underwater image | |
CN105959121B (en) | A kind of mobile terminal with identification verification function | |
CN106022320A (en) | Automatic control device based on iris recognition | |
CN106204842A (en) | A kind of door lock being identified by iris | |
CN106206112A (en) | A kind of control switch by iris identification | |
CN106203345A (en) | A kind of image control device with identification verification function | |
CN106203348A (en) | A kind of high-tension switch cabinet with identification verification function | |
Deb et al. | Multiplet based technique to derive atmospheric winds from Kalpana-1 | |
CN106203347A (en) | A kind of multi-stage regulating switch opened based on iris identification |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
TA01 | Transfer of patent application right | ||
TA01 | Transfer of patent application right |
Effective date of registration: 20180815 Address after: 224000 5, Blaupunkt road 199, Yancheng City environmental protection science and Technology City, Jiangsu (28) Applicant after: Yancheng Kuang Zhi Technology Co., Ltd. Address before: No. 32, Zhenhai District, Zhejiang Province, Zhenhai District, Drum Tower East Road, Ningbo, Zhejiang Applicant before: Zhong Linchao |
|
GR01 | Patent grant | ||
GR01 | Patent grant | ||
CF01 | Termination of patent right due to non-payment of annual fee | ||
CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20181002 Termination date: 20210708 |