CN107273834A - A kind of iris identification method and identifier - Google Patents
A kind of iris identification method and identifier Download PDFInfo
- Publication number
- CN107273834A CN107273834A CN201710419540.7A CN201710419540A CN107273834A CN 107273834 A CN107273834 A CN 107273834A CN 201710419540 A CN201710419540 A CN 201710419540A CN 107273834 A CN107273834 A CN 107273834A
- Authority
- CN
- China
- Prior art keywords
- image
- iris
- scale map
- gray
- matrix
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/18—Eye characteristics, e.g. of the iris
- G06V40/19—Sensors therefor
-
- 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)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Ophthalmology & Optometry (AREA)
- Human Computer Interaction (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Collating Specific Patterns (AREA)
Abstract
The present invention relates to a kind of iris identification method and identifier, this method includes the iris image of intake eyes of user, the iris image is handled, and obtains iris code；When user carries out authentication, the iris image of eyes of user is absorbed again, the iris image is handled, obtain new iris code；The characteristic matching degree of the iris code and iris code that prestore is calculated using Hamming distances matching method, judges whether subscriber authentication passes through according to the height of the characteristic matching degree.A kind of identifier is further related to, it includes：Image collection module, image processing module, image authentication module.The precision of acquisition and the speed of identification are substantially increased by the present invention, the possibility of user identity fraud is significantly reduced.
Description
Technical field
The invention belongs to iris recognition field, more particularly to a kind of iris identification method and identifier.
Background technology
Iris recognition technology is started late relative to the fingerprint identification technology being most widely used at present, but iris recognition skill
Art possesses the incomparable advantage of the other biological identification technology such as fingerprint, recognition of face, and these advantages cause iris recognition technology
As most promising biological identification technology at present.Abroad, iris recognition technology state of development relative maturity, iris is known
Other technology has been used for security protection, the field such as national defence, and has realized industrialization.At home, Duo Jia research institutions are for iris recognition skill
Art rests on theoretical research stage or volume production stage, and state of development is also immature.But the country exists in the prior art at present
Recognition speed is slow, the relatively low shortcoming of ratio of precision.
The content of the invention
The technical problems to be solved by the invention are：There is recognition speed in the prior art slow, the relatively low shortcoming of ratio of precision.
To solve technical problem above, the invention provides a kind of iris identification method, this method comprises the following steps：
S1, absorbs the iris image of eyes of user, the iris image is pre-processed, gradation of image figure is obtained；
S2, carries out hough-circle transform to described image gray-scale map, the described image gray-scale map after conversion is mapped into establishment
New blank gray-scale map on, obtain rectangle iris image；
S3, processing is filtered to the rectangle iris image, extracts characteristic points all in the rectangle iris image,
And the characteristic point is encoded, obtain iris code；
S4, the characteristic matching degree of the iris code and iris code that prestore is calculated using Hamming distances matching method, according to institute
State characteristic matching degree and be compared with preset value and judge whether subscriber authentication passes through, if more than preset value, passing through, if not
Exceed, then do not pass through, and preserve the iris code of the user.
Beneficial effects of the present invention：Adopt with the aforedescribed process, by the client iris image of acquisition by hough-circle transform and
Filtering process, substantially increases the precision of acquisition, significantly reduce user identity fraud possibility, while using hamming away from
From matching method, the degree of accuracy to subscriber authentication and recognition speed are substantially increased, recognition speed is stable within 1 second.
Further, the S1 includes：
S11, absorbs the iris image of eyes of user, by the affiliated type of height and width and image of the iris image
It is converted into data matrix；
S12, gray-scale map is converted into by the data matrix, obtains described image gray-scale map.
Further, in the S12, including：
Picture element matrix in the data matrix is calculated with color histogram matrix, and it is corresponding according to result of calculation
Pixel in the picture element matrix is converted into black picture element or white pixel by ground, obtains described image gray-scale map.
Further, in the S12, in addition to：Pixel to the picture element matrix carries out Gaussian smoothing.
Above-mentioned further beneficial effect：The noise of pixel is reduced, the precision of pixel is improved, greatly reduces pixel
Interference.
Further, in the S3, the filtering process refers to filter out different frequency in the rectangle iris image
Image, retains the image of identical frequency.
Further, the filtering process also includes：Spatial domain convolutional calculation is carried out to the rectangle iris image.
The invention further relates to a kind of iris recognition device, it is characterised in that the identifier includes：Image collection module, image
Processing module, image authentication module；Described image acquisition module, the iris image for absorbing eyes of user, to the iris
Image is pre-processed, and obtains gradation of image figure；Described image processing module, for carrying out Hough circle to described image gray-scale map
Conversion, the described image gray-scale map after conversion is mapped on the new blank gray-scale map of establishment, rectangle iris image is obtained, and
Processing is filtered to the rectangle iris image, characteristic points all in the rectangle iris image are extracted, and to the spy
Levy and encoded, obtain iris code；Described image authentication module, for calculating the iris prestored using Hamming distances matching method
The characteristic matching degree of code and iris code, judgement subscriber authentication is compared according to the characteristic matching degree and preset value
Whether pass through, if more than preset value, passing through, if being no more than, not passing through, and preserve the iris code of the user.
Beneficial effects of the present invention：Using above-mentioned system, by the client iris image of acquisition by hough-circle transform and
Filtering process, substantially increases the precision of acquisition, significantly reduce user identity fraud possibility, while using hamming away from
From matching method, the degree of accuracy to subscriber authentication is substantially increased.
Further, described image acquisition module includes：Data matrix unit, gradation of image figure unit；The data square
Array element, the iris image for absorbing eyes of user, by the affiliated type of height and width and image of the iris image
It is converted into data matrix；Described image gray-scale map unit, for the data matrix to be converted into gray-scale map, obtains described image
Gray-scale map.
Further, described image gray-scale map unit, specifically for by the picture element matrix and color in the data matrix
Histogram matrix is calculated, and according to result of calculation correspondingly by the pixel in the picture element matrix be converted into black picture element or
Person's white pixel, obtains described image gray-scale map.
Further, described image gray-scale map unit, is additionally operable to carry out at Gaussian smoothing the pixel of the picture element matrix
Reason.
Above-mentioned further beneficial effect：The noise of pixel is reduced, the precision of pixel is improved, greatly reduces pixel
Interference.
Brief description of the drawings
Fig. 1 is a kind of flow chart of iris identification method of the present invention；
Fig. 2 is a kind of structural representation of iris recognition device of the present invention.
Embodiment
The principle and feature of the present invention are described below in conjunction with accompanying drawing, the given examples are served only to explain the present invention, and
It is non-to be used to limit the scope of the present invention.
Embodiment 1
As shown in figure 1, the present embodiment 1 is a kind of iris identification method, this method comprises the following steps：
S1, absorbs the iris image of eyes of user, the iris image is pre-processed, gradation of image figure is obtained；
S2, carries out hough-circle transform to described image gray-scale map, the described image gray-scale map after conversion is mapped into establishment
New blank gray-scale map on, obtain rectangle iris image；
S3, processing is filtered to the rectangle iris image, extracts characteristic points all in the rectangle iris image,
And the characteristic point is encoded, obtain iris code；
S4, the characteristic matching degree of the iris code and iris code that prestore is calculated using Hamming distances matching method, according to institute
State characteristic matching degree and be compared with preset value and judge whether subscriber authentication passes through, if more than preset value, passing through, if not
Exceed, then do not pass through, and preserve the iris code of the user.
It should be noted that step S1 is the iris image for absorbing eyes of user in the present embodiment 1, by acquired rainbow
The height of film image, width and unsteady state operation are stored in a 2D matrix for the byte of character, and the 2D matrixes, which are one, to shift to an earlier date just
The data type set, this data type is one and included：Three integer type parameters, these three parameters are extracted
The iris image height of eyes of user, width and image type.In addition, all matrixes used in the present embodiment 1 are all 2D
Matrix, will be stored with height, and the matrix conversion of width and image type is gradation of image figure.
In addition, step S2 is to carry out hough-circle transform to gradation of image figure in the present embodiment 1, include meter in conversion process
Calculate the pupil center of circle and calculate the center of circle for the circle for including iris portion, wherein the process for calculating the pupil center of circle is as follows：Sat in card Deere
The equation of circle is in mark system
α=x-rcos θ
Wherein a, b are the center of circle, and r is radius, and θ is angle, it is known that on circle equation a little it is identical, it is possible to by time
The center of circle corresponding to cumulative all non-zero points is gone through, i.e., may be the vectorial intersection number of mould on the point in the center of circle by calculating any point
It is how many to judge whether the point is the required center of circle.If result of calculation is more than high threshold, the point is the center of circle, on the contrary then be not.
High threshold in the present embodiment 1 is 200, and center of circle detection threshold value is 0.Therefore the pupil center of circle and radius are determined, then create three it is whole
Digital variable x, y, r, the center of circle (X, Y) and radius (R) are preserved into three variables respectively.
Pixel within the circle value where pupil in gradation of image figure is all rewritten as black, one is used in iterative calculation
Individual nested for circulations, i.e., comprising another for circulations in one for circulation, wherein the iterator circulated twice is respectively input
The line number and columns of image array, if current pixel point in garden, is not skipping current iteration where pupil, execution changes next time
Generation.
Include the center of circle of the circle of iris portion for calculating, be that hough-circle transform is used to gradation of image figure again, herein
The high threshold of the edge function at place is 200, and center of circle detection threshold value is pupil radium * 1.3, creates three integer types variable x`, y`,
R`, the center of circle (X, Y) and radius (R) are preserved into three variables respectively.Then iris image is normalized, will normalized
Gradation of image figure afterwards is mapped on the new blank gray-scale map of establishment, using 1X512 matrix, by the circular chart of iris
The polar-mapping of picture obtains a rectangle iris image, its formula is on card Deere coordinate：
X=ρ cos θ
Y=ρ sin θs
Wherein x, y are rectangular co-ordinate point, and ρ is pole axis, and θ is angle, and the value of pole axis is r`-r difference, so far new ash
Rectangle iris image is saved in degree figure.
Processing is filtered to above-mentioned steps S2 rectangle iris image in the step S3 of the present embodiment 1, rectangle rainbow is extracted
All characteristic points in film image, and the characteristic point is encoded, iris code is obtained, it is necessary to illustrate, in this implementation
Usually using 2D Gabor filters in example 1, its formula is：
Wherein, λ is wavelength, and θ is the angle that filter characteristic is selected, and φ is phase pushing figure, and γ is the two of any point
Dimension coordinate ratio, σ bandwidth for needed for filtering spatial frequency.
The iris code stored before being calculated using Hamming distances matching method in the step S4 of the present embodiment 1 is with specifically obtaining
Iris code characteristic matching degree, Hamming distances refer to two codings digit different on correspondence position.It is exemplified below：101 and 001
First difference, then Hamming distances 1, then with Hamming distances divided by 2048, draw input parameter point in final result, calculation formula
It is not two parts of irises code, and offset, offset value is 0 herein.
Then result of calculation is analyzed again, if result of calculation is less than or equal to 0.25, be verified, if result is big
In 0.25, then verify and do not pass through.
Alternatively, step S1 includes in the present embodiment 1：
S11, absorbs the iris image of eyes of user, by the affiliated type of height and width and image of the iris image
It is converted into data matrix；
S12, gray-scale map is converted into by the data matrix, obtains described image gray-scale map.
Alternatively, in the S12, including：
Picture element matrix in the data matrix is calculated with color histogram matrix, and it is corresponding according to result of calculation
Pixel in the picture element matrix is converted into black picture element or white pixel by ground, obtains described image gray-scale map.
It should be noted that being specifically to improve all white pixel regions in matrix using for circulations in the present embodiment 1
Brightness.An integer type variable is created, and is entered as 127, in this for is circulated, iterator is the pixel in matrix, is followed every time
The difference that ring subtracts 127 with the pixel of current location is multiplied by 255, if result is less than zero, the position pixel value is rewritten as 0, if result
More than 255, the position pixel value is rewritten as 255, if result is equal to zero, the position pixel value is not altered.In the present embodiment 1
It is that the 1X256 color histogram matrixes that all pixels in matrix and attribute are 32SC1 are calculated with for circulations, if worked as
Preceding matrix pixel is multiplied by height more than original image width with current color histogram matrix pixel sum and is multiplied by 0.02, is become
Black picture element, multiplies if current matrix pixel is multiplied by height with current color histogram matrix pixel sum more than original image width
With 0.04, become white pixel.
Alternatively, in the S12, in addition to：Pixel to the picture element matrix carries out Gaussian smoothing.
It should be noted that all pixels point carries out Gaussian smoothing in being used in the present embodiment 1 to image array
Processing is to reduce noise, and its formula is：
Wherein x, y value current pixel point abscissa value and ordinate value, σ value is that 15, π is pi.
In addition, in the present embodiment 1 for ease of extraction subsequently to iris image, it is necessary to image carry out rim detection,
The value of Sobel Operator is calculated first, and formula is as follows：
Wherein GXFor horizontal Sobel Operator, M is initial iris image matrix.
Wherein GYFor vertical Sobel Operator, M is initial iris image matrix.
The gradient approximation size of each pixel in calculation matrix, its formula is:
Wherein GXAnd GYRespectively horizontal and vertical Sobel Operator.
The pixel gradient direction is measured, its formula is：
Alternatively, in the S3, the filtering process refers to the figure for filtering out different frequency in the rectangle iris image
Picture, retains the image of identical frequency.
It should be noted that each wave filter is only conveyed the image texture that it has identical frequency and led in the present embodiment 1
Cross, and the image texture of different frequency is blocked, and that is to say the image of different frequency in rectangle iris image, retains identical frequency
Image.
Alternatively, the filtering process also includes：Spatial domain convolutional calculation is carried out to the rectangle iris image.
It should be noted that being to make spatial domain to each image block in the iris image by wave filter in the present embodiment 1
Convolution, obtains the output of wave filter, and the characteristic point that all wave filters are exported is saved in 64X512 matrix.
Embodiment 2
As shown in Fig. 2 the present embodiment 2 is a kind of iris recognition device, the identifier includes：At image collection module, image
Manage module, image authentication module；Described image acquisition module, the iris image for absorbing eyes of user, to the iris figure
As being pre-processed, gradation of image figure is obtained；Described image processing module, becomes for carrying out Hough circle to described image gray-scale map
Change, the described image gray-scale map after conversion is mapped on the new blank gray-scale map of establishment, rectangle iris image is obtained, and it is right
The rectangle iris image is filtered processing, extracts characteristic points all in the rectangle iris image, and to the feature
Point is encoded, and obtains iris code；Described image authentication module, for calculating the iris prestored code using Hamming distances matching method
With the characteristic matching degree of iris code, it is compared according to the characteristic matching degree with preset value and judges that subscriber authentication is
It is no to pass through, if more than preset value, passing through, if being no more than, not passing through, and preserve the iris code of the user.
Alternatively, include in described image acquisition module：Data matrix unit, gradation of image figure unit；The data square
Array element, the iris image for absorbing eyes of user, by the affiliated type of height and width and image of the iris image
It is converted into data matrix；Described image gray-scale map unit, for the data matrix to be converted into gray-scale map, obtains described image
Gray-scale map.
Alternatively, described image gray-scale map unit, specifically for the picture element matrix in the data matrix and color is straight
Square figure matrix is calculated, and according to result of calculation correspondingly by the pixel in the picture element matrix be converted into black picture element or
White pixel, obtains described image gray-scale map.
It should be noted that being specifically to improve all white pixel regions in matrix using for circulations in the present embodiment 2
Brightness.An integer type variable is created, and is entered as 127.In this for circulations, iterator is the pixel in matrix, is followed every time
The difference that ring subtracts 127 with the pixel of current location is multiplied by 255, if result is less than zero, the position pixel value is rewritten as 0, if result
More than 255, the position pixel value is rewritten as 255, if result is equal to zero, the position pixel value is not altered.In the present embodiment 2
It is that the 1X256 color histogram matrixes that all pixels in matrix and attribute are 32SC1 are calculated with for circulations, if worked as
Preceding matrix pixel is multiplied by height more than original image width with current color histogram matrix pixel sum and is multiplied by 0.02, is become
Black picture element, multiplies if current matrix pixel is multiplied by height with current color histogram matrix pixel sum more than original image width
With 0.04, become white pixel.
Alternatively, described image gray-scale map unit, is additionally operable to carry out Gaussian smoothing to the pixel of the picture element matrix.
It should be noted that all pixels point carries out Gaussian smoothing in being used in the present embodiment 2 to image array
Processing is to reduce noise, and its formula is：
Wherein x, y value current pixel point abscissa value and ordinate value, σ value is that 15, π is pi.
In addition, in the present embodiment 2 for ease of extraction subsequently to iris image, it is necessary to image carry out rim detection,
The value of Sobel Operator is calculated first, and formula is as follows：
Wherein GXFor horizontal Sobel Operator, M is initial iris image matrix.
Wherein GYFor vertical Sobel Operator, M is initial iris image matrix.
The gradient approximation size of each pixel in calculation matrix, its formula is:
Wherein GXAnd GYRespectively horizontal and vertical Sobel Operator.
The pixel gradient direction is measured, its formula is：
Alternatively, the filtering process refers to the image for filtering out different frequency in the rectangle iris image, retains phase
The image of same frequency.
It should be noted that each wave filter is only conveyed the image texture that it has identical frequency and led in the present embodiment 2
Cross, and the image texture of different frequency is blocked, and that is to say the image of different frequency in rectangle iris image, retains identical frequency
Image.
Alternatively, the filtering process also includes：Spatial domain convolutional calculation is carried out to the rectangle iris image.
It should be noted that being to make spatial domain to each image block in the iris image by wave filter in the present embodiment 2
Convolution, obtains the output of wave filter, and the characteristic point that all wave filters are exported is saved in 64X512 matrix.
In this manual, identical embodiment or example are necessarily directed to the schematic representation of above-mentioned term.
Moreover, specific features, structure, material or the feature of description can be in any one or more embodiments or example with suitable
Mode is combined.In addition, in the case of not conflicting, those skilled in the art can be by the difference described in this specification
The feature of embodiment or example and non-be the same as Example or example is combined and combined.
The foregoing is only presently preferred embodiments of the present invention, be not intended to limit the invention, it is all the present invention spirit and
Within principle, any modification, equivalent substitution and improvements made etc. should be included in the scope of the protection.
Claims (10)
1. a kind of iris identification method, it is characterised in that this method comprises the following steps：
S1, absorbs the iris image of eyes of user, the iris image is pre-processed, gradation of image figure is obtained；
S2, carries out hough-circle transform to described image gray-scale map, the described image gray-scale map after conversion is mapped into the new of establishment
Blank gray-scale map on, obtain rectangle iris image；
S3, processing is filtered to the rectangle iris image, extracts characteristic points all in the rectangle iris image, and right
The characteristic point is encoded, and obtains iris code；
S4, the characteristic matching degree of the iris code and iris code that prestore is calculated using Hamming distances matching method, according to the spy
Levy matching degree and be compared with preset value and judge whether subscriber authentication passes through, if more than preset value, passing through, if not surpassing
Cross, then do not pass through, and preserve the iris code of the user.
2. according to the method described in claim 1, it is characterised in that the S1 includes：
S11, absorbs the iris image of eyes of user, and the affiliated type of the height and width and image of the iris image is changed
Into data matrix；
S12, gray-scale map is converted into by the data matrix, obtains described image gray-scale map.
3. method according to claim 2, it is characterised in that in the S12, including：
Picture element matrix in the data matrix is calculated with color histogram matrix, and correspondingly will according to result of calculation
Pixel in the picture element matrix is converted into black picture element or white pixel, obtains described image gray-scale map.
4. method according to claim 3, it is characterised in that in the S12, in addition to：To the picture of the picture element matrix
Element carries out Gaussian smoothing.
5. according to any described methods of claim 1-4, it is characterised in that in the S3, the filtering process refers to filtering
Fall the image of different frequency in the rectangle iris image, retain the image of identical frequency.
6. method according to claim 5, it is characterised in that the filtering process also includes：To the rectangle iris figure
As carrying out spatial domain convolutional calculation.
7. a kind of iris recognition device, it is characterised in that the identifier includes：Image collection module, image processing module and image
Authentication module；
Described image acquisition module, the iris image for absorbing eyes of user, pre-processes to the iris image, obtains
Gradation of image figure；
Described image processing module, for carrying out hough-circle transform to described image gray-scale map, by the described image ash after conversion
Degree figure is mapped on the new blank gray-scale map of establishment, obtains rectangle iris image, and the rectangle iris image is filtered
Ripple processing, extracts characteristic points all in the rectangle iris image, and the characteristic point is encoded, and obtains iris code；
Described image authentication module, the characteristic matching of the iris code and iris code that prestore is calculated using Hamming distances matching method
Degree, is compared with preset value according to the characteristic matching degree and judges whether subscriber authentication passes through, if more than preset value,
Pass through, if being no more than, do not pass through, and preserve the iris code of the user.
8. identifier according to claim 7, it is characterised in that described image acquisition module includes：Data matrix unit
With gradation of image figure unit；
The data matrix unit, the iris image for absorbing eyes of user, by the height and width of the iris image with
And the affiliated type of image is converted into data matrix；
Described image gray-scale map unit, for the data matrix to be converted into gray-scale map, obtains described image gray-scale map.
9. identifier according to claim 8, it is characterised in that described image gray-scale map unit, specifically for by described in
Picture element matrix in data matrix is calculated with color histogram matrix, and according to result of calculation correspondingly by the pixel square
Pixel in battle array is converted into black picture element or white pixel, obtains described image gray-scale map.
10. identifier according to claim 9, it is characterised in that described image gray-scale map unit, is additionally operable to the picture
The pixel of prime matrix carries out Gaussian smoothing.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710419540.7A CN107273834A (en) | 2017-06-06 | 2017-06-06 | A kind of iris identification method and identifier |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710419540.7A CN107273834A (en) | 2017-06-06 | 2017-06-06 | A kind of iris identification method and identifier |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107273834A true CN107273834A (en) | 2017-10-20 |
Family
ID=60064997
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710419540.7A Pending CN107273834A (en) | 2017-06-06 | 2017-06-06 | A kind of iris identification method and identifier |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107273834A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109165587A (en) * | 2018-08-11 | 2019-01-08 | 石修英 | intelligent image information extraction method |
| CN109965843A (en) * | 2019-03-14 | 2019-07-05 | 华南师范大学 | A kind of eye movements system passing picture based on filtering |
| CN112380966A (en) * | 2020-11-12 | 2021-02-19 | 西安电子科技大学 | Monocular iris matching method based on feature point reprojection |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101093538A (en) * | 2006-06-19 | 2007-12-26 | 电子科技大学 | Method for identifying iris based on zero crossing indication of wavelet transforms |
| CN101916362A (en) * | 2010-05-28 | 2010-12-15 | 深圳大学 | Iris positioning method and iris identification system |
| US20150098630A1 (en) * | 2013-10-08 | 2015-04-09 | Sri International | Iris biometric recognition module and access control assembly |
| CN106204958A (en) * | 2016-07-08 | 2016-12-07 | 钟林超 | A kind of ATM input equipment being identified by iris |
-
2017
- 2017-06-06 CN CN201710419540.7A patent/CN107273834A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101093538A (en) * | 2006-06-19 | 2007-12-26 | 电子科技大学 | Method for identifying iris based on zero crossing indication of wavelet transforms |
| CN101916362A (en) * | 2010-05-28 | 2010-12-15 | 深圳大学 | Iris positioning method and iris identification system |
| US20150098630A1 (en) * | 2013-10-08 | 2015-04-09 | Sri International | Iris biometric recognition module and access control assembly |
| CN106204958A (en) * | 2016-07-08 | 2016-12-07 | 钟林超 | A kind of ATM input equipment being identified by iris |
Non-Patent Citations (1)
| Title |
|---|
| 李栩荣: "虹膜识别算法的研究和实现", 《中国优秀博硕士学位论文全文数据库（硕士） 信息科技辑》 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109165587A (en) * | 2018-08-11 | 2019-01-08 | 石修英 | intelligent image information extraction method |
| CN109965843A (en) * | 2019-03-14 | 2019-07-05 | 华南师范大学 | A kind of eye movements system passing picture based on filtering |
| CN109965843B (en) * | 2019-03-14 | 2022-05-24 | 华南师范大学 | Eye movement system based on filtering image transmission |
| CN112380966A (en) * | 2020-11-12 | 2021-02-19 | 西安电子科技大学 | Monocular iris matching method based on feature point reprojection |
| CN112380966B (en) * | 2020-11-12 | 2023-06-02 | 西安电子科技大学 | Monocular iris matching method based on feature point re-projection |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108765476B (en) | Polarized image registration method | |
| CN108694393A (en) | A kind of certificate image text area extraction method based on depth convolution | |
| CN107273834A (en) | A kind of iris identification method and identifier | |
| CN110473242A (en) | A kind of texture characteristic extracting method, texture feature extraction device and terminal device | |
| US7103234B2 (en) | Method for blind cross-spectral image registration | |
| CN106326841A (en) | Quick iris recognition algorithm | |
| CN105488791B (en) | The localization method and device of image border in natural background | |
| CN104951940A (en) | Mobile payment verification method based on palmprint recognition | |
| Huang et al. | Data matrix code location based on finder pattern detection and bar code border fitting | |
| CN107180436A (en) | A kind of improved KAZE image matching algorithms | |
| CN110378893A (en) | Image quality evaluating method, device and electronic equipment | |
| CN106599891A (en) | Remote sensing image region-of-interest rapid extraction method based on scale phase spectrum saliency | |
| WO2022199395A1 (en) | Facial liveness detection method, terminal device and computer-readable storage medium | |
| CN110516661A (en) | U.S. pupil detection method and device applied to iris recognition | |
| CN110309814A (en) | A kind of iris identification method and device based on edge detection | |
| CN109815772A (en) | Fingerprint enhancement, recognition methods, device and Fingerprint enhancement identifying system | |
| CN102393905B (en) | Extraction method of venous mode texture on back of hand | |
| CN112633070A (en) | High-resolution remote sensing image building extraction method and system | |
| CN116883464A (en) | Registration method for large-viewing-angle difference optics and SAR remote sensing image | |
| Aksakallı et al. | A novel approach for copy-move forgery detection using bilateral filtering | |
| Li et al. | Visual saliency detection based on gradient contrast and color complexity | |
| CN111160290A (en) | Method for extracting effective region of palm vein | |
| CN106327423B (en) | Remote sensing image registration method and system based on directed line segment | |
| CN110264531A (en) | A kind of catching for X-comers takes method, apparatus, system and readable storage medium storing program for executing | |
| Ma et al. | Improving KAZE feature matching algorithm with alternative image gray method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20171020 |