CN101866420B - Image preprocessing method for optical volume holographic iris recognition - Google Patents
Image preprocessing method for optical volume holographic iris recognition Download PDFInfo
- Publication number
- CN101866420B CN101866420B CN201010191412.XA CN201010191412A CN101866420B CN 101866420 B CN101866420 B CN 101866420B CN 201010191412 A CN201010191412 A CN 201010191412A CN 101866420 B CN101866420 B CN 101866420B
- Authority
- CN
- China
- Prior art keywords
- iris
- image
- region
- radius
- pixel
- 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.)
- Expired - Fee Related
Links
Images
Landscapes
- Image Analysis (AREA)
- Collating Specific Patterns (AREA)
- Image Processing (AREA)
Abstract
The invention relates to the technical field of iris recognition, in particular to an image preprocessing method for optical volume holographic iris recognition, which is used for processing an acquired iris image. The method comprises the following steps of: (11) positioning an iris area of the acquired iris image and determining an inner edge and an outer edge of an iris and an upper eyelid and a lower eyelid of a human eye; (12) after determining the iris area, performing feature extraction and encoding on the iris; and (13) performing optical volume holographic recognition and secondary encoding on encoded data. The method is used in preprocessing procedure of the optical volume holographic iris recognition, and due to the method, the technology for the optical volume holographic iris recognition is applied in the field of the iris recognition. The technology for the optical volume holographic iris recognition is the application of the optical volume holographic iris recognition to the iris recognition and also has the function of parallelly recognizing the iris. The image preprocessing method for the optical volume holographic iris recognition has the advantages of quickly positioning the iris, reducing the amount of calculation, improving the calculation accuracy, and simultaneously improving the iris recognition ratio.
Description
Technical field
The present invention relates to iris recognition technology field, particularly a kind of image preprocessing method for optical volume holographic iris identification.
Background technology
Iris recognition has pin-point accuracy, is not easy the advantages such as forgery, non-infringement, is considered to the promising biological identification technology of tool.Since iris recognition has started business-like process, through development for many years, reach its maturity algorithm field is technical.Optical volume holographic image recognition technology can make iris recognition technology have further development in the application that reaches iris recognition field, allows iris recognition have and identifies several iris images simultaneously, and raising iris recognition is processed the ability of a large amount of iris images.
Iris image identification pretreatment process comprised location, feature extraction and coding.At present the method for reliable Iris Location mainly contains the method that calculus method and edge extracting are combined with Hough transformation; Feature extracting method has the method such as gabor small echo and log-gabor filtering.
But prior art is not applied to iris recognition field optical volume holographic image recognition technology.
Summary of the invention
The invention provides a kind of image preprocessing method for optical volume holographic iris identification, optical volume holographic image recognition technology is not applied to the technical matters in iris recognition field to solve prior art.
The technical solution used in the present invention is as follows:
For an image preprocessing method for optical volume holographic iris identification, the iris image gathering to be processed, described method comprises:
(11) iris image gathering is carried out to the location of iris region, determine palpebra inferior on iris outer edge and human eye;
(12) determine after iris region, iris is carried out to feature extraction and coding;
(13) data after coding are carried out to optical volume holographic identification secondary coding.
As a kind of preferred version, the concrete steps of described step (11) are as follows:
First carry out iris outer edge location, concrete steps are as follows:
(211) by scaled the iris image gathering;
(212) carry out morphologic filtering;
(213) approach near minimum value and the position center in image as according to by image binaryzation according to brightness;
(214), according to the statistical relationship between the size of pupil region and iris radius, determine region of search and the search radius of iris outward flange and inward flange;
(215) image after morphologic filtering described in (212) is used to canny operator extraction edge, obtain outline map, and the marginal point beyond iris outward flange region of search is removed, all possible radius is carried out to Hough transformation, in the Hough transformation figure that each possibility radius obtains, choose the peak value of three brightness maximums as the best Hough of candidate peak;
(217) after finding outward flange radius, outline map described in (215) is amplified, and the marginal point beyond iris inner boundary region of search is removed, all possible radius is carried out to Hough transformation, in the Hough transformation figure that each possibility radius obtains, choose the peak value of three brightness maximums as the best Hough of candidate peak, in the best Houghs of all candidates peak, choose the peak point that meets minimum basis for estimation;
Carry out palpebra inferior location on human eye, concrete steps are as follows:
(221) adopting Hough transformation to carry out circle search, is upper eyelid from the nearest circular arc in the pupil center of circle;
(222) Imo is carried out to 180 degree rotations, the acquisition methods of described Imo is:
First iris image is carried out to morphologic filtering and obtain positioning image Imr, remove the details of reflective spot and eyelashes, and then carry out binaryzation, find out the approximate region that near low brightness area picture centre is pupil, then find out this regional barycenter C and its transverse and longitudinal width of this area measure and get again the approximate diameter d of both mean values as pupil; Imr is scaled, then to choose centered by C, the border circular areas that diameter is 3d is as the target area of iris outward flange location; Use cann
ythe edge of the target area of operator extraction Imr, other parts first mask, and obtain Imo;
(223) adopting Hough transformation to carry out circle search, is palpebra inferior from the nearest circular arc in the pupil center of circle
(224) obtain positioning result;
As further preferred version, the described basis for estimation of described step (217) judges according to following mode:
The brightness of determining i peak value of j possibility radius is I
ij, this peak value is D to the distance of pupil center
ij, the high-high brightness of all peak values is I
max, be D to the bee-line of pupil center
min, the basis for estimation S at i peak
ijfor:
wherein p
ithe weight of luminance standard, p
dit is the weight of criterion distance.Preferred pI=0.3 in the present invention, pD=0.7.
As a kind of preferred version,
The positioning result that above-mentioned steps (224) is obtained checks, if the standard of not meeting thinks that original iris image shooting quality is not good, output re-starts the prompting of iris image acquiring, and the described concrete steps that positioning result is checked are as follows:
(41) in compute location result, pupil, with interior mean flow rate, if higher than the threshold value of setting, export the prompting that re-starts iris image acquiring, otherwise performs step (42);
(42) mean flow rate in the annulus of the outer certain radius of calculating pupil region, if lower than the threshold value of setting, export the prompting that re-starts iris image acquiring, otherwise execution step (43);
(43) check whether iris outward flange center and inward flange off-centring exceed threshold value, if exceed threshold value, export the prompting that re-starts iris image acquiring, otherwise execution step (44);
(44) area that in calculating, palpebra inferior covers accounts for the number percent of iris annulus area, exceedes threshold value and exports the prompting that re-starts iris image acquiring.
As a kind of preferred version, the concrete steps of described step (12) are as follows:
(51) upper and lower people's eye pattern eyelid part is got up as noise region mark, and by its deletion;
(52) use affined transformation that iris region and noise region are transformed into rectangle;
(53) image of step (52) is done to Fourier transform, use log-gabor filtering to carry out feature extraction at frequency domain;
(54) do inverse fourier transform, the real part of getting image carries out binaryzation, and noise region is deleted, and obtains last feature coding;
As a kind of preferred version, the concrete steps of described step (13) are as follows:
(61) step (12) is encoded each pixel of the image obtaining, be mapped in secondary coding figure, mapping method is that the corresponding width of a pixel on former figure is 1 pixel, it is highly the region more than 1 pixel, this region is called secondary pixel, and secondary pixel is arranged in the middle of image from the two ends up and down of image.After having arranged all pixels, keep secondary coding figure approximately 1/3 zone line there is no secondary pixel;
(62) line-spacing of the image obtaining in step (61) is expanded to more than 1 pixel.
As further preferred version:
Described step (61) is by encode each pixel of the image obtaining of step (12), and being mapped to width is 1 pixel, is highly the region of 10～15 pixels;
Described step (62) expands the line-spacing of the image obtaining in step (61) to 15～20 pixels.
The present invention, for the pretreatment process of optical volume holographic iris identification, has realized optical volume holographic image recognition technology has been applied to iris recognition field.Optical volume holographic iris recognition technology is that optical volume holographic image is identified in the application in iris recognition, has the function of parallel identification iris.The feature that the present invention is directed to optical volume holographic image identification is carried out pre-treatment to iris image, has the function of general iris recognition pre-treatment, meets again the requirement of optical volume holographic image identification simultaneously.Adopt disposal route of the present invention, realize the quick location of iris, reduce calculated amount, improve accuracy in computation, improve iris recognition rate simultaneously.
Brief description of the drawings
Fig. 1 is the people's eye pattern after morphologic filtering;
Fig. 2 is pupil approximate region;
Fig. 3 is the outline map after restriction;
Fig. 4 carries out the Hough transformation figure after Hough transformation to Fig. 3;
Fig. 5 is iris-encoding schematic diagram;
Fig. 6 is original eye image;
Fig. 7 adopts the Iris Location of the present invention to Fig. 6 image;
Fig. 8 is Iris Location process flow diagram;
Fig. 9 is the output map after secondary coding.
Embodiment
Below in conjunction with the drawings and specific embodiments, the present invention is described in more detail.
The present invention need to complete three functions: location, iris feature extraction and the coding of iris region, optical volume holographic identification secondary coding.
1. the location of iris outer edge and upper palpebra inferior
Iris-orientation Algorithm of the present invention adopts morphologic filtering and hough conversion to combine, and the quantity that reduces marginal point by coarse positioning reduces operation time to reach, and improves the object of arithmetic speed.Iris Location has comprised the location of iris outer edge and upper palpebra inferior.
Iris outer edge is located as shown in Figure 1 to 4.As shown in Figure 1, before carrying out hough conversion, first image is dwindled according to a certain percentage, then use morphologic filtering to remove the unnecessary details (reflective spot in eyelashes, pupil, the reflective spot of skin, iris texture etc.), morphologic filtering need to first carry out that image is opened and then carry out image and close, and these operations all need a structural elements usually to complete.Here adopt circular structural element, suitably select the radius of this structural element, just can remove the detailed structure below certain size in image, the edge of these structures can be weakened, brightness meeting and near pixel are close, can not become edge in the time of edge extracting.
Then as shown in Figure 2, approach near minimum value and the position center in image as according to by image binaryzation according to brightness, find the approximate region of pupil.Pupil region brightness approaches minimum value, and is not positioned at four corners (brightness of four corners is generally very low, also approaches minimum value) of eye image, as long as meet this two conditions through after morphologic filtering, can ensure it is the approximate region of pupil.
According to pupil region and size and iris radius between statistical relationship (radius of pupil is 0.1～0.8 of iris radius, iris region is positioned at the longitudinal 1/3 of full figure center and horizontal 1/3 center), so just can roughly determine the outer peripheral region of search of iris and search radius.The pupil region obtaining is done to following processing: obtain the average transverse and longitudinal coordinate (being center of gravity) of pupil region as the approximate centre of pupil, then obtain each pixel in pupil region from the ultimate range of pupil center the roughly radius as pupil
According to Daugman professor's result of study, the radius of pupil is 0.1～0.8 of iris radius.So, there is the large roughly radius of above-mentioned pupil to obtain the roughly radius of iris, the multiple integers within the scope of this are exactly multiple possible radiuses.
After finding outward flange radius, image is amplified, find region of search and the search radius of inward flange according to outward flange.This is in order to allow search have identical precision when outer edge.The radius ratio outward flange of inward flange is little, so adopt larger ratio in the time searching inward flange.The general ratio that the iris image of collection is enlarged into twice that adopts.The approximate range of inside radius is exactly the approximate range of pupil.This is roughly worth 0.5 as may inward flange the lower limit of radius, 1.5 times as the upper limit, adopts the method search inward flange identical with search outward flange.
Image after morphologic filtering is used to canny operator extraction edge, all possible radius is carried out to Hough transformation, in the Hough transformation figure that each possibility radius obtains, choose the peak value of three brightness maximums as the best Hough of candidate peak;
Suppose to have 10 possible radiuses, can carry out 10 times Hough transformation, use a different radius at every turn, obtain 3 peak values the brightest, always have 3 × 10=30 peak value.Then among these 30 peak values, find out most suitable peak value according to following basis for estimation.
Basis for estimation:
Pass through hou
gh converts each possible radius and has a hough Transformation Graphs, is called one deck.Can obtain so a series of peak point.If only rely on high-high brightness to judge that iris edge is easy to make mistakes.The restrictive condition that adds implantation site in screening peak point will effectively improve the accuracy of location.The brightness at i peak of j layer is I
ij, be D to the distance of Fig. 2 bright pixel central area
ij, the high-high brightness at all peaks is I
max, be D to the bee-line of Fig. 2 bright pixel central area
min.The basis for estimation at i peak is:
wherein p
ithe weight of luminance standard, p
dit is the weight of criterion distance.P in work herein
i=0.3, p
d=0.7.Look for S
ijminimum value can determine the circular center of circle and radius, thereby complete the location at edge.
In above-mentioned process, need not check and also have the not radius of judgement, because the possible range of radius is known, all integer radiuses within the scope of this all will do Hough transformation.
The detection of upper palpebra inferior is also to use hough conversion to carry out circular search, and different with the detection of iris outer edge is the standard that judges best Hough peak.Hough transformation can be known a circular center of circle and radius, so said method, at definite iris inner boundary, has just drawn the center of circle of pupil when pupil.In detecting, eyelid also there is the texture that radian is close with eyelid on due to eyelid, so position is main criterion: be eyelid from the nearest circular arc in the pupil center of circle.
The inspection that finally also will position.Index for examination has 4:
A) in compute location result, pupil, with interior mean flow rate, if higher than the threshold value of setting, shows that pupil radius is looked for too greatly or eyelid has covered pupil subregion;
B) calculate the mean flow rate in the annulus of the outer certain radius of pupil region, if lower than the threshold value of setting, show that pupil radius looks for too littlely, the annulus of above-mentioned certain radius is generally the annulus of 1.0 times～1.3 times of pupil radiuses;
If c) pupil location is correct, but iris outward flange center and inward flange off-centring exceed threshold value, show outward flange location mistake;
D) calculate area that upper palpebra inferior covers and account for the number percent of iris annulus area, exceeding threshold value, to show that eyelid location mistake or eyelid cover iris region excessive.
If these 4 indexs have one defective, will obtain locating failed result, prompting want Resurvey iris image to position.
2. iris feature extracts and coding
Algorithm flow chart is as shown in Figure 5:
(51) upper and lower people's eye pattern eyelid part is got up as noise region mark, and by its deletion;
(52) use affined transformation that iris region and noise region are transformed into rectangle, as described in (51), noise region is upper and lower eyelid part.(51) meaning of in, noise token being got up be generate one with the binary image of former figure same size, on this figure, to be illustrated in the pixel of same position in former figure be noise to white pixel.(52) in, use radiation conversion former figure and above-mentioned noise binary picture need to be transformed into respectively to rectangle;
(53) image of step (52) is done to fast fourier transform, use log-gabor filtering to carry out feature extraction at frequency domain;
(54) do inverse fourier transform, the real part of getting image carries out binaryzation, and noise region is deleted, and obtains last feature coding;
The part that eyelid in iris image is hidden is removed, then iris portion location being found is carried out affined transformation and is obtained Imp, eyelid part is also done to affined transformation simultaneously, generate a corresponding mask, then Imp is carried out to log-Gabor filtering, and get its real part and carry out binaryzation, then do with computing and obtain iris-encoding with mask.
3. optical volume holographic identification secondary coding
The iris-encoding that feature extraction is obtained carries out secondary coding.Picture size after secondary coding is determined by the SLM resolution in volume holographic image identification light path.Here claim that the input picture before secondary coding is imin, the output image after secondary coding is Imout.The pixel count of Imout is larger than imin's.A width in the corresponding Imout of a pixel in imin is 1 pixel, is highly the region of 10 pixels, and this region is called secondary pixel.So a line correspondence in imin just becomes the striped that upper and lower width is 10 pixels in Imout.For the feature of optical volume holographic image identification, reduce vertically harassing between pixel, the fringe spacing in Imout is 15 pixels, and by the Pixel arrangement in imin the two ends up and down to Imout, the pixel of imin is not arranged in the centre of Imout.
Experimental results show that above-mentioned secondary coding method harassing between can lowering between pixel in optical volume holographic image identification, improve discrimination.Adopt the iris database CASI of the Chinese Academy of Sciences version2 device1 in 37 people's iris image as experimental subjects.Identifying with optical volume holographic image recognition methods, is 91% without the image recognition rate of secondary coding, and the discrimination after secondary coding is 100%.As shown in Figure 6, be original eye image, employing figure source be CASI database version2 device1 0055 0055_005.bmp; Fig. 7 adopts the present invention to carry out the design sketch of Iris Location.
Fig. 8 is Iris Location process flow diagram, and the left side is positioning flow, and the right is the design sketch of corresponding step.Iris Location employing morphology and hough convert outer edge and the upper palpebra inferior of the method searching iris combining.First iris image is carried out to morphologic filtering and obtain positioning image Imr, remove the unwanted details such as reflective spot and eyelashes, and then carry out binaryzation, find out the approximate region that near low brightness area picture centre is pupil, then find out this regional barycenter C and its transverse and longitudinal width of this area measure and get again the approximate diameter d of both mean values as pupil.Imr is scaled, then to choose centered by C, the border circular areas that diameter is 3d is as the target area of iris outward flange location.The edge that uses this part region of canny operator extraction Imr, other parts first mask, and obtain Imo, then using d as using hough conversion to find the outward flange of iris with reference to choosing certain radius region of search to Imo.Using Imo as hough, the object of conversion can reduce the quantity of marginal point, reduces the time of computing, has shielded other noises simultaneously, has increased the accuracy of location.Find after iris outward flange the suitable amplification in proportion by Imo, then, using d as carrying out hough conversion with reference to delimiting certain scope as iris inner boundary search radius, find iris inner boundary.Do like this and can ensure that the search precision of inward flange is consistent with outward flange.
For with reference to Imo the first half certain area is intercepted, then as using hough conversion, search radius finds upper eyelid taking the certain multiple of iris outward flange radius according to iris outside radius.Then by Imo Rotate 180 degree, intercept out a region of the containing eyelid search palpebra inferior that makes to use the same method.Fig. 9 is the output map Imout after secondary coding.
When specific coding, using matlab to write Algorithm source code is compiled into dynamic link libraries and calls processing iris image for C++, taking CASI database version2 37 human iris's images in device1 file as experimental subjects, carry out doing optical volume holographic iris identification after image pre-treatment, discrimination is 100%, and average every width figure processing time is 3.6 seconds.
Claims (6)
1. for an image preprocessing method for optical volume holographic iris identification, the iris image gathering is processed, be it is characterized in that, described method comprises:
(11) iris image gathering is carried out to the location of iris region, determine palpebra inferior on iris outer edge and human eye;
(12) determine after iris region, iris is carried out to feature extraction and coding;
(13) data after coding are carried out to optical volume holographic identification secondary coding;
The concrete steps of described step (11) are as follows:
First carry out iris outer edge location, concrete steps are as follows:
(211) by scaled the iris image gathering;
(212) carry out morphologic filtering;
(213) approach near minimum value and the position center in image as according to by image binaryzation according to brightness;
(214), according to the statistical relationship between the size of pupil region and iris radius, determine region of search and the search radius of iris outward flange and inward flange;
(215) image after morphologic filtering described in (212) is used to canny operator extraction edge, obtain outline map, and the marginal point beyond iris outward flange region of search is removed, all possible radius is carried out to Hough transformation, in the Hough transformation figure that each possibility radius obtains, choose the peak value of three brightness maximums as the best Hough of candidate peak;
(217) after finding outward flange radius, outline map described in (215) is amplified, and the marginal point beyond iris inner boundary region of search is removed, all possible radius is carried out to Hough transformation, in the Hough transformation figure that each possibility radius obtains, choose the peak value of three brightness maximums as the best Hough of candidate peak, in the best Houghs of all candidates peak, choose the peak point that meets minimum basis for estimation;
Carry out palpebra inferior location on human eye, concrete steps are as follows:
(221) adopting Hough transformation to carry out circle search, is upper eyelid from the nearest circular arc in the pupil center of circle;
(222) Imo is carried out to 180 degree rotations, the acquisition methods of described Imo is:
First iris image is carried out to morphologic filtering and obtain positioning image Imr, remove the details of reflective spot and eyelashes, and then carry out binaryzation, find out the approximate region that near low brightness area picture centre is pupil, then find out this regional barycenter C and its transverse and longitudinal width of this area measure and get again the approximate diameter d of both mean values as pupil; Imr is scaled, then to choose centered by C, the border circular areas that diameter is 3d is as the target area of iris outward flange location; The edge that uses the target area of canny operator extraction Imr, other parts first mask, and obtain Imo;
(223) adopting Hough transformation to carry out circle search, is palpebra inferior from the nearest circular arc in the pupil center of circle;
(224) obtain positioning result.
2. method according to claim 1, is characterized in that, described (217) described basis for estimation judges according to following mode:
The brightness of determining i peak value of j possibility radius is I
ij, this peak value is D to the distance of pupil center
ij, the high-high brightness of all peak values is I
max, be D to the bee-line of pupil center
min, the basis for estimation S at i peak
ijfor:
Wherein p
ibe the weight of luminance standard, pD is the weight of criterion distance.
3. method according to claim 1 and 2, it is characterized in that, the positioning result that above-mentioned steps (224) is obtained checks, if the standard of not meeting, think that original iris image shooting quality is not good, output re-starts the prompting of iris image acquiring, and the described concrete steps that positioning result is checked are as follows:
(41) in compute location result, pupil, with interior mean flow rate, if higher than the threshold value of setting, export the prompting that re-starts iris image acquiring, otherwise performs step (42);
(42) mean flow rate in the annulus of the outer certain radius of calculating pupil region, if lower than the threshold value of setting, export the prompting that re-starts iris image acquiring, otherwise execution step (43);
(43) check whether iris outward flange center and inward flange off-centring exceed threshold value, if exceed threshold value, export the prompting that re-starts iris image acquiring, otherwise execution step (44);
(44) area that in calculating, palpebra inferior covers accounts for the number percent of iris annulus area, exceedes threshold value and exports the prompting that re-starts iris image acquiring.
4. method according to claim 1, is characterized in that, the concrete steps of described step (12) are as follows:
(51) upper and lower people's eye pattern eyelid part is got up as noise region mark, and by its deletion;
(52) use affined transformation that iris region and noise region are transformed into rectangle;
(53) image of step (52) is done to fast fourier transform, use log-gabor filtering to carry out feature extraction at frequency domain;
(54) do inverse fourier transform, the real part of getting image carries out binaryzation, and noise region is deleted, and obtains last feature coding.
5. method according to claim 1, is characterized in that, the concrete steps of described step (13) are as follows:
(61) step (12) is encoded each pixel of the image obtaining, be mapped in secondary coding figure, mapping method is that the corresponding width of a pixel on former figure is 1 pixel, it is highly the region more than 1 pixel, this region is called secondary pixel, and secondary pixel is arranged in the middle of image from the two ends up and down of image; After having arranged all pixels, keep secondary coding figure approximately 1/3 zone line there is no secondary pixel;
(62) line-spacing of the image obtaining in step (61) is expanded to more than 1 pixel.
6. method according to claim 5, is characterized in that:
Described step (61) is by encode each pixel of the image obtaining of step (12), and being mapped to width is 1 pixel, is highly the region of 10～15 pixels;
Described step (62) expands the line-spacing of the image obtaining in step (61) to 15～20 pixels.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201010191412.XA CN101866420B (en) | 2010-05-28 | 2010-05-28 | Image preprocessing method for optical volume holographic iris recognition |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201010191412.XA CN101866420B (en) | 2010-05-28 | 2010-05-28 | Image preprocessing method for optical volume holographic iris recognition |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN101866420A CN101866420A (en) | 2010-10-20 |
| CN101866420B true CN101866420B (en) | 2014-06-04 |
Family
ID=42958141
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201010191412.XA Expired - Fee Related CN101866420B (en) | 2010-05-28 | 2010-05-28 | Image preprocessing method for optical volume holographic iris recognition |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101866420B (en) |
Families Citing this family (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| AU2016245332A1 (en) * | 2015-04-08 | 2017-10-19 | Wavefront Biometric Technologies Pty Limited | Multi-biometric authentication |
| CN107844736B (en) * | 2016-09-19 | 2021-01-01 | 北京眼神科技有限公司 | Iris positioning method and device |
| CN110276229A (en) * | 2018-03-14 | 2019-09-24 | 京东方科技集团股份有限公司 | Target object regional center localization method and device |
| CN109190447A (en) * | 2018-07-06 | 2019-01-11 | 深圳虹识技术有限公司 | A kind of iris identification method and equipment based on chip |
| US11877800B2 (en) * | 2018-07-27 | 2024-01-23 | Kaohsiung Medical University | Method and system for detecting blepharoptosis |
| JP7137746B2 (en) * | 2019-01-30 | 2022-09-15 | 株式会社Ｊｖｃケンウッド | VIDEO PROCESSING DEVICE, VIDEO PROCESSING METHOD, AND VIDEO PROCESSING PROGRAM |
| CN110680275A (en) * | 2019-10-21 | 2020-01-14 | 徐明娜 | Binocular multispectral pupil light reflex quantitative measuring instrument |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101416407A (en) * | 2003-08-04 | 2009-04-22 | 霍洛威夫公司 | Enhanced holographic communications apparatus and method |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7512249B2 (en) * | 2004-04-26 | 2009-03-31 | Graphic Security Systems Corporation | System and method for decoding digital encoded images |
-
2010
- 2010-05-28 CN CN201010191412.XA patent/CN101866420B/en not_active Expired - Fee Related
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101416407A (en) * | 2003-08-04 | 2009-04-22 | 霍洛威夫公司 | Enhanced holographic communications apparatus and method |
Non-Patent Citations (3)
| Title |
|---|
| 周雁 等.基于体全息图像库的模式识别算法.《中国激光》.2002,第29卷(第4期), * |
| 孟凡伟 等.基于小波变换的体全息相关识别系统.《压电与声光》.2009,第31卷(第1期), * |
| 黄欢 等.基于光学体全息和小波包分解的虹膜识别实现.《光电工程》.2007,第34卷(第5期), * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN101866420A (en) | 2010-10-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101866420B (en) | Image preprocessing method for optical volume holographic iris recognition | |
| CN105956578B (en) | A kind of face verification method of identity-based certificate information | |
| Gottschlich | Curved-region-based ridge frequency estimation and curved Gabor filters for fingerprint image enhancement | |
| CN100373397C (en) | Pre-processing method for iris image | |
| CN101266645B (en) | Iris positioning method based on multi-resolution analysis | |
| Puhan et al. | Efficient segmentation technique for noisy frontal view iris images using Fourier spectral density | |
| CN101201893A (en) | Iris recognizing preprocessing method based on grey level information | |
| JP2007188504A (en) | Method for filtering pixel intensity in image | |
| CN107273812B (en) | A kind of living body iris method for anti-counterfeit for authentication | |
| CN105261015A (en) | Automatic eyeground image blood vessel segmentation method based on Gabor filters | |
| CN110555380A (en) | Finger vein identification method based on Center Loss function | |
| Zulfira et al. | Segmentation technique and dynamic ensemble selection to enhance glaucoma severity detection | |
| CN104361319A (en) | Fake fingerprint detection method based on SVM-RFE (support vector machine-recursive feature elimination) | |
| CN102332098B (en) | Method for pre-processing iris image | |
| AlShemmary | Towards accurate pupil detection based on morphology and Hough transform | |
| Xu et al. | Improving the performance of iris recogniton system using eyelids and eyelashes detection and iris image enhancement | |
| KR20030066512A (en) | Iris Recognition System Robust to noises | |
| CN106778499A (en) | A kind of method of quick positioning people's eye iris during iris capturing | |
| CN110148125A (en) | Adaptive skin oil and fat detection method based on color detection | |
| CN105975960A (en) | Iris identification method based on texture-direction energy characteristic | |
| Noah Akande et al. | Implementation of a framework for healthy and diabetic retinopathy retinal image recognition | |
| Ahmed et al. | Retina based biometric authentication using phase congruency | |
| Ali et al. | Segmenting retinal blood vessels with gabor filter and automatic binarization | |
| Ashame et al. | Abnormality Detection in Eye Fundus Retina | |
| Khan et al. | Fast and efficient iris segmentation approach based on morphology and geometry operation |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20140604 Termination date: 20200528 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |