CN110826384A - System and method for enhancing iris recognition accuracy - Google Patents
System and method for enhancing iris recognition accuracy Download PDFInfo
- Publication number
- CN110826384A CN110826384A CN201910255205.7A CN201910255205A CN110826384A CN 110826384 A CN110826384 A CN 110826384A CN 201910255205 A CN201910255205 A CN 201910255205A CN 110826384 A CN110826384 A CN 110826384A
- Authority
- CN
- China
- Prior art keywords
- iris
- module
- image
- characteristic
- exclusive
- 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
- 238000000034 method Methods 0.000 title claims abstract description 35
- 230000002708 enhancing effect Effects 0.000 title claims abstract description 30
- 230000011218 segmentation Effects 0.000 claims abstract description 54
- 238000013441 quality evaluation Methods 0.000 claims abstract description 38
- 238000000605 extraction Methods 0.000 claims abstract description 23
- 238000004458 analytical method Methods 0.000 claims description 38
- 239000000284 extract Substances 0.000 claims description 23
- 238000007781 pre-processing Methods 0.000 claims description 16
- 238000012545 processing Methods 0.000 claims description 8
- 230000000877 morphologic effect Effects 0.000 claims description 6
- 238000006243 chemical reaction Methods 0.000 claims description 4
- YTAHJIFKAKIKAV-XNMGPUDCSA-N [(1R)-3-morpholin-4-yl-1-phenylpropyl] N-[(3S)-2-oxo-5-phenyl-1,3-dihydro-1,4-benzodiazepin-3-yl]carbamate Chemical compound O=C1[C@H](N=C(C2=C(N1)C=CC=C2)C1=CC=CC=C1)NC(O[C@H](CCN1CCOCC1)C1=CC=CC=C1)=O YTAHJIFKAKIKAV-XNMGPUDCSA-N 0.000 claims description 3
- 230000003014 reinforcing effect Effects 0.000 claims 1
- 210000000554 iris Anatomy 0.000 description 593
- 238000005516 engineering process Methods 0.000 description 6
- 206010004950 Birth mark Diseases 0.000 description 5
- 210000001747 pupil Anatomy 0.000 description 4
- GNFTZDOKVXKIBK-UHFFFAOYSA-N 3-(2-methoxyethoxy)benzohydrazide Chemical compound COCCOC1=CC=CC(C(=O)NN)=C1 GNFTZDOKVXKIBK-UHFFFAOYSA-N 0.000 description 2
- 238000011156 evaluation Methods 0.000 description 2
- 208000014674 injury Diseases 0.000 description 2
- 238000001303 quality assessment method Methods 0.000 description 2
- 230000008733 trauma Effects 0.000 description 2
- FGUUSXIOTUKUDN-IBGZPJMESA-N C1(=CC=CC=C1)N1C2=C(NC([C@H](C1)NC=1OC(=NN=1)C1=CC=CC=C1)=O)C=CC=C2 Chemical compound C1(=CC=CC=C1)N1C2=C(NC([C@H](C1)NC=1OC(=NN=1)C1=CC=CC=C1)=O)C=CC=C2 FGUUSXIOTUKUDN-IBGZPJMESA-N 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000002035 prolonged effect Effects 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 238000001356 surgical procedure Methods 0.000 description 1
Images
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)
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
Abstract
The invention discloses a system and a method for enhancing iris recognition accuracy, which comprises an iris traditional characteristic module, an iris exclusive characteristic module and an iris recognition module, wherein the iris traditional characteristic module is used for acquiring iris traditional characteristics and comprises the following steps: the device comprises a data acquisition module, a quality evaluation module, an iris segmentation module and an iris coding module; the exclusive characteristic module of iris is used for acquireing the exclusive characteristic of iris, includes: an iris exclusive feature extraction module and an iris exclusive feature coding module; the iris recognition module includes: the iris traditional characteristic comparison module is used for comparing and identifying the iris image to be identified and the iris traditional texture characteristics of the registered iris image; the iris exclusive characteristic comparison module is used for judging whether an iris registration image which completes the traditional iris characteristic matching has the iris exclusive characteristic or not, and if the iris registration image has the iris exclusive characteristic, matching iris exclusive characteristic codes; and the identity output result is used for outputting the comparison result.
Description
Technical Field
The invention relates to the technical field of biological feature recognition, in particular to a system and a method for enhancing iris recognition accuracy.
Background
At present, iris recognition technology dates back to 80 s in the 19 th century, and scientists tried to utilize the characteristics of uniqueness and stability of irises as a means of identity recognition. In the 21 st century, the development of iris recognition technology is rapidly promoted by strong computing speed and data storage capacity and advanced image acquisition technology. As one of the safest biometric techniques, the iris recognition technique is widely applied to the fields of security, anti-terrorism, finance, and the like.
Although iris recognition technology is almost the safest biometric technology, there is still a possibility of misidentification in the face of a large number of users. There are many possible reasons for false recognition, including iris image quality, iris recognition algorithms, and external environmental interference.
Like the birthmarks on the skin, a large portion of the population carries birthmarks and large plaques in the iris, and a significant portion of the population forms permanent unique marks on the iris due to surgery and trauma, and these birthmarks, plaques and marks become the unique identity of these individuals. The unique identity marks are taken as the iris exclusive characteristics and are integrated into the traditional iris recognition technology, so that the identity recognition precision of the part of people carrying the exclusive characteristics can be improved, and the precision of the whole iris recognition system is improved.
Disclosure of Invention
The technical problem to be solved by the invention is to provide a system and a method for enhancing the iris recognition precision aiming at users with unique marks in irises, and further improve the iris recognition precision under the existing hardware and software conditions.
In order to achieve the technical effect, the invention discloses a system for enhancing iris recognition accuracy, which comprises an iris traditional characteristic module, an iris exclusive characteristic module and an iris recognition module, wherein the iris traditional characteristic module comprises:
the data acquisition module is used for acquiring an iris registration image or an iris identification image;
the quality evaluation module is used for evaluating whether the iris registration image or the iris identification image meets the quality requirement;
the iris segmentation module is used for segmenting the inner circle and the outer circle of the iris from the iris registration image or the iris identification image;
the iris coding module is used for coding the segmented iris registration image or iris identification image to form an iris traditional characteristic code;
the iris-specific feature module includes:
the iris exclusive characteristic extraction module is used for processing and analyzing the iris registration image acquired by the iris traditional characteristic module and extracting the iris exclusive characteristic of the iris registration image;
the special iris feature coding module is used for coding the special iris features of the iris registration image and the iris identification image to form special iris feature codes;
the iris recognition module includes:
the iris traditional characteristic comparison module is used for comparing and identifying the iris identification image and the iris traditional characteristic code of the iris registration image;
the iris exclusive characteristic comparison module is used for judging whether an iris registration image successfully matched with the iris identification image by the traditional iris characteristic code has the iris exclusive characteristic or not, and if the iris registration image has the iris exclusive characteristic, comparing and identifying the iris identification image and the iris exclusive characteristic code of the iris registration image;
and the identity output result is used for outputting the comparison result.
The system for enhancing the iris recognition accuracy of the present invention is improved in that the quality evaluation module comprises:
the primary quality evaluation module is used for evaluating the quality of the whole iris registration image or iris identification image acquired by the data acquisition module;
and the final quality evaluation module is used for evaluating the quality of the iris area image in the iris registration image or the iris identification image.
The invention also discloses a method for enhancing the iris recognition precision, which comprises the steps of iris registration and iris recognition, wherein the iris registration comprises the following steps:
a. the data acquisition module acquires an iris registration image;
b. the iris segmentation module segments the inner circle and the outer circle of the iris in the iris registration image;
c. the iris coding module codes the segmented iris registration image to form an iris traditional characteristic code;
d. the special iris feature extraction module analyzes the iris registration image and extracts the special iris features of the iris registration image with the special iris features;
e. the special iris feature coding module codes the special iris features extracted from the iris registration image to form special iris feature codes;
the iris recognition comprises the following steps:
f. the data acquisition module acquires an iris recognition image;
g. the iris segmentation module segments the inner circle and the outer circle of the iris in the iris recognition image;
h. the iris coding module codes the segmented iris recognition image to form an iris traditional characteristic code;
i. the iris traditional characteristic comparison module compares and identifies the iris identification image with the iris traditional characteristic codes of the iris registration image;
j. the iris exclusive characteristic comparison module judges whether an iris registration image successfully matched with the iris identification image by traditional characteristic codes has iris exclusive characteristics or not, and if the iris registration image has the iris exclusive characteristics, the iris identification image and the iris exclusive characteristic codes of the iris registration image are compared and identified;
k. and the identity output module outputs a comparison result.
The method for enhancing the iris identification precision is improved in that after the data acquisition module acquires the iris registration image or the iris identification image, the primary quality evaluation module is required to perform primary quality evaluation on the acquired iris registration image or iris identification image.
The method for enhancing the iris recognition accuracy is further improved in that after the iris segmentation module completes segmentation of the inner and outer circles of the iris in the iris registration image or the iris recognition image, a final quality evaluation module is required to evaluate the image quality of the iris region in the iris registration image or the iris recognition image after the segmentation is completed.
The method for enhancing the iris recognition accuracy of the present invention is further improved in that the step d further comprises:
the data pulling module pulls the iris registration image;
the first preprocessing module carries out preprocessing such as noise reduction and enhancement on the iris registration image;
the first segmentation module performs multi-scale segmentation on the iris registration image and extracts a stable characteristic region;
the exclusive characteristic analysis module comprehensively analyzes the gray distribution and morphological characteristics of the iris registration image characteristic region and determines whether the iris exclusive characteristics exist or not;
and the data output module outputs the analysis result of the exclusive characteristic analysis module.
The method for enhancing the iris recognition accuracy of the present invention is further improved in that the step d further comprises:
the data pulling module pulls the iris registration image;
the first preprocessing module carries out preprocessing such as noise reduction and enhancement on the iris registration image;
the iris expansion module completes the conversion from the space coordinate to the polar coordinate and expands the iris annular region image in the iris registration image to form a rectangular image;
the first segmentation module performs multi-scale segmentation on the unfolded iris rectangular image and extracts a stable characteristic region;
the exclusive characteristic analysis module comprehensively analyzes the gray distribution and morphological characteristics of the iris registration image characteristic region and determines whether the iris exclusive characteristics exist or not;
and the data output module outputs the analysis result of the exclusive characteristic analysis module.
The method for enhancing the iris recognition accuracy of the present invention is further improved in that the step e further comprises:
extracting the exclusive iris feature of the iris registration image;
extracting characteristic points of the exclusive characteristic profile of the iris;
calculating the invariant moment of the exclusive characteristic profile of the iris;
and coding the exclusive characteristics of the iris according to the characteristic points and the invariant moment of the exclusive characteristic outline of the iris to form exclusive characteristic codes of the iris.
The method for enhancing the iris recognition accuracy is further improved in that the step j further comprises the following steps:
the iris exclusive characteristic judgment module judges whether an iris registration image matched with the iris identification image by the iris traditional characteristic code has the iris exclusive characteristic or not, and if the iris exclusive characteristic does not exist, the comparison result of the iris traditional characteristic code is transmitted to the identity output module;
if the iris registration image has the exclusive iris feature, the second preprocessing module carries out local processing on the corresponding position of the iris identification image according to the position information of the exclusive iris feature in the iris registration image;
the second segmentation module performs segmentation on the corresponding position of the iris identification image according to the scale of the exclusive features of the iris in the iris registration image, and extracts stable exclusive features of the iris;
the special iris feature coding module codes the special iris features of the extracted iris identification image to form special iris feature codes;
the special iris feature matching module matches the special iris feature code of the iris identification image with the special iris feature code of the iris registration image.
The method for enhancing the iris identification precision is further improved in that the exclusive characteristic analysis module decides at most 2 characteristic regions from the characteristic regions extracted by the first segmentation module as the iris exclusive characteristics of the iris registration image.
The system and the method for enhancing the iris identification precision judge whether the iris areas of the users have special characteristics such as imprints, birthmarks and the like, extract and analyze the special characteristics of the iris areas of the users, and compare the special characteristics of the iris in the iris image during the user identification with the special characteristics of the iris in the iris image of the user extracted during the registration, so that the users can compare the special characteristics of the iris besides the traditional characteristics of the iris during the identity authentication, and the accuracy of the identity identification is improved.
Drawings
FIG. 1 is a flow chart of iris enrollment of the present invention;
FIG. 2 is a flow chart of iris recognition according to the present invention;
FIG. 3 is a flow chart of the extraction of the exclusive features of the iris according to a preferred embodiment of the present invention;
FIG. 4 is a flow chart of the iris-specific feature extraction according to another preferred embodiment of the present invention;
FIG. 5 is a flow chart of the comparison of the exclusive characteristics of the iris according to the present invention.
Detailed Description
The present invention will be described in further detail with reference to the accompanying drawings and specific embodiments.
The traditional characteristics of the iris in the invention refer to the texture characteristics of the iris area in the prior art, and the special characteristics of the iris refer to birthmarks, plaques, permanent unique imprints formed on the iris due to operation or trauma and the like in the iris area, iris characteristics obviously different from the surrounding iris textures in the iris area and the like. Typically, the acquired iris image has iris-specific features with a significantly lower intensity than the surrounding iris area.
The invention relates to a system and a method for enhancing iris recognition precision, which are used for conveniently distinguishing iris registration images which are acquired as registration templates when a user performs iris registration, and iris recognition images which are acquired when the user performs identity authentication, namely iris recognition.
The invention discloses a system for enhancing iris recognition accuracy, which comprises an iris traditional characteristic module, an iris exclusive characteristic module and an iris recognition module. As shown in fig. 1, the iris traditional feature module is configured to acquire, segment, and encode an iris registration image or an iris recognition image to obtain an iris traditional feature code of the iris registration image or the iris recognition image; this traditional characteristic module of iris includes:
the data acquisition module is used for acquiring an iris registration image or an iris identification image;
the quality evaluation module is used for evaluating whether the iris registration image or the iris identification image meets the quality requirement;
the iris segmentation module is used for segmenting the inner circle and the outer circle of the iris from the iris registration image or the iris identification image, namely extracting the outer edge of the pupil and the outer edge of the iris;
and the iris coding module is used for coding the segmented iris registration image or iris identification image to form an iris traditional characteristic code and storing the iris traditional characteristic code into a database.
Preferably, in this embodiment, the quality evaluation module includes:
the primary quality evaluation module is used for preliminarily evaluating the quality of the whole iris registration image or the iris identification image acquired by the data acquisition module, and generally preliminarily judging whether the whole iris registration image or the whole iris identification image of the iris registration image or the iris identification image is overexposed, too dark, defocused, motion blurred or the like;
the final quality evaluation module is used for evaluating the quality of the iris region image in the iris registration image or the iris identification image in a targeted manner, and generally comprises the steps of analyzing whether the image of the iris region is focused or not, whether motion blur exists or not, whether strong light spot interference exists or not, whether the exposure rate is qualified or not and the like.
As shown in fig. 1 and 2, the iris-specific feature module includes:
the iris exclusive characteristic extraction module is used for processing and analyzing the iris registration image acquired by the iris traditional characteristic module and extracting the iris exclusive characteristic of the iris registration image;
the special iris feature coding module is used for coding the special iris features of the iris registration image and the iris identification image to form special iris feature codes;
as shown in fig. 2, the iris recognition module includes:
the iris traditional characteristic comparison module is used for comparing and identifying the iris identification image and the iris traditional characteristic code of the iris registration image;
the iris exclusive characteristic comparison module is used for judging whether an iris registration image successfully matched with the iris identification image through traditional iris characteristic coding has iris exclusive characteristics or not, if the iris registration image has the iris exclusive characteristics, extracting the iris exclusive characteristics of the iris identification image, and comparing and identifying the iris identification image and the iris exclusive characteristic coding of the iris registration image; in this embodiment, if the iris registration image has the exclusive iris feature, local processing and segmentation are performed on the corresponding position of the iris identification image according to the position information and scale of the exclusive iris feature in the iris registration image, and a stable exclusive iris feature is extracted
And the identity output result is used for outputting the comparison result.
The invention relates to a system for enhancing iris recognition accuracy, wherein an iris traditional characteristic module is used for acquiring, segmenting and coding an iris registration image or an iris recognition image; in order to improve the efficiency, the exclusive iris feature of the iris registration image and the exclusive iris feature of the iris identification image are extracted in different modes, so that the exclusive iris feature extraction module only extracts the exclusive iris feature of the iris registration image, and the exclusive iris feature comparison module is responsible for extracting the exclusive iris feature of the iris identification image. The iris exclusive characteristics of the iris registration image and the iris exclusive characteristics of the iris identification image are coded in the same mode, namely, the iris exclusive characteristics are coded through an iris exclusive characteristic coding module.
The invention also discloses a method for enhancing the iris recognition accuracy, which comprises the steps of iris registration and iris recognition, wherein as shown in figure 1, the iris registration comprises the following steps:
a. the data acquisition module acquires an iris registration image;
b. the iris segmentation module segments the inner circle and the outer circle of the iris in the iris registration image, namely extracts the outer edge of the pupil and the outer edge of the iris;
c. the iris coding module codes the segmented iris registration image to form an iris traditional characteristic code, and then stores the iris traditional characteristic code into a database;
d. the special iris feature extraction module analyzes the iris registration image and extracts the special iris features of the iris registration image with the special iris features;
e. the special iris feature coding module codes the special iris features extracted from the iris registration image to form special iris feature codes;
as shown in fig. 2, the iris recognition includes the steps of:
f. the data acquisition module acquires an iris recognition image;
g. the iris segmentation module segments the inner circle and the outer circle of the iris in the iris recognition image, namely extracts the outer edge of the pupil and the outer edge of the iris;
h. the iris coding module codes the segmented iris recognition image to form an iris traditional characteristic code;
i. the iris traditional characteristic comparison module compares and identifies the iris identification image with the iris traditional characteristic code of the iris registration image, namely, the iris traditional characteristic code of the iris identification image is compared and identified with the iris traditional characteristic code of the iris registration image in the database; when the iris traditional characteristic comparison module is used for comparing the traversal database with the sample to be detected, namely comparing the traversal database with the iris traditional characteristic code of the iris identification image, and if the iris traditional characteristic code is not successfully matched with the iris traditional characteristic code, ending the iris identification process; if matched iris registration images are matched in the database, the iris traditional characteristic comparison module outputs registration information of the iris identification images and the iris registration images in the database to the iris exclusive characteristic comparison module;
j. the iris exclusive characteristic comparison module judges whether an iris registration image successfully matched with the iris identification image by iris traditional characteristic coding has iris exclusive characteristics, if the iris registration image has the iris exclusive characteristics, the iris exclusive characteristics of the iris identification image are extracted, the iris exclusive characteristics of the iris identification image are coded by the iris exclusive characteristic coding module to form iris exclusive characteristic codes of the iris identification image, and then the iris identification image and the iris exclusive characteristic codes of the iris registration image are compared and identified, namely the iris exclusive characteristic codes of the iris identification image and the iris exclusive characteristic codes of the iris registration image matched with the iris traditional characteristic codes are compared and identified;
k. and the identity output module outputs the comparison result and feeds back the identification result to the user.
Further, as shown in fig. 1 and fig. 2, after the data acquisition module acquires the iris registration image, that is, between step a and step b, the data acquisition module further includes the following steps:
the primary quality evaluation module is used for carrying out primary quality evaluation on the acquired iris registration image, wherein the primary quality evaluation refers to the primary judgment of whether the whole image of the iris registration image is overexposed, whether the whole image is too dark, whether the whole image is defocused, whether the whole image is motion blurred or not and the like;
after the data acquisition module acquires the iris recognition image, namely the data acquisition module between the step f and the step g further comprises the following steps:
the primary quality evaluation module carries out primary quality evaluation on the acquired iris identification image, wherein the primary quality evaluation refers to primary judgment on whether the whole image of the iris identification image is overexposed, too dark, defocused or motion blurred or not.
Further, as shown in fig. 1 and fig. 2, after the iris segmentation module completes segmentation of the inner and outer circles of the iris in the iris registration image, that is, between step b and step c, the method further includes the following steps:
the final quality evaluation module is used for performing final quality evaluation on the iris registration image subjected to the inner and outer circle segmentation, wherein the final quality evaluation refers to analysis on whether an iris region image in the iris registration image is focused or not, whether motion is fuzzy or not, whether strong light spot interference exists or not, whether the exposure rate is qualified or not and the like;
after the iris segmentation module completes segmentation of the inner circle and the outer circle of the iris in the iris recognition image, the method also comprises the following steps between the step g and the step h:
and the final quality evaluation module is used for performing final quality evaluation on the iris identification image subjected to the inner and outer circle segmentation, wherein the final quality evaluation refers to analysis on whether the iris area image in the iris identification image is focused or not, whether the motion is fuzzy or not, whether strong light spot interference exists or not, whether the exposure rate is qualified or not and the like.
The iris registration image or the iris recognition image is evaluated by the primary quality evaluation module and the final quality evaluation module and then enters the next step, and the iris registration image or the iris recognition image which is not evaluated by the primary quality evaluation module or the final quality evaluation module needs to be acquired again by the data acquisition module.
When the iris image is collected, because the iris segmentation and the extraction of the iris features cannot be accurately carried out due to the unclear image, quality evaluation must be carried out firstly, the traditional method is to evaluate the quality of the whole iris image (the collected iris image is the image of the whole eye region) before the iris segmentation, and if the quality reaches the standard, subsequent segmentation and feature extraction are carried out. Because the feature extraction has high requirements on the quality of the iris region image, the adoption of the traditional quality assessment mode means that the high-precision quality assessment is carried out on all the regions of each acquired image, including the iris and the non-iris, so that the burden of hardware equipment is increased, and the registration or identification time is prolonged.
According to the two-stage evaluation mode, the primary quality evaluation module which is high in speed but not high in fineness is used for evaluating the whole iris image (including the iris and the non-iris area) before iris segmentation, and after segmentation is completed, the final quality module is used for only carrying out high-precision quality evaluation on the segmented iris area, so that the iris image with unclear texture of the iris area is eliminated. And the evaluation time is reduced while the quality of the iris image is ensured.
Furthermore, when the iris exclusive feature extraction module extracts the iris exclusive feature of the iris registration image, there are two extraction modes, one is to extract on the original image of the iris registration image, and the other is to extract on the iris region expansion image of the iris registration image.
As shown in fig. 3, in a preferred embodiment, the iris-specific feature extracting module extracts an iris-specific feature from an original image of the iris registration image, and step d further includes:
the data pulling module pulls the iris registration image which is not processed by the iris exclusive characteristic extraction module from the database;
the first preprocessing module carries out preprocessing such as noise reduction and enhancement on the iris registration image pulled by the data pulling module;
the first segmentation module performs multi-scale segmentation on the preprocessed iris registration image and extracts at least one stable characteristic region;
the exclusive characteristic analysis module comprehensively analyzes the gray distribution and morphological characteristics of the at least one characteristic region, decides at most 2 characteristic regions from the at least one characteristic region as the exclusive characteristics of the iris registration image, and marks the iris registration image without the exclusive characteristics of the iris as the exclusive characteristics without the iris if the exclusive characteristic analysis module fails to extract the exclusive characteristics of the iris from the at least one characteristic region after comprehensive analysis;
the data output module outputs the analysis result of the exclusive characteristic analysis module, and for the iris registration image marked as the exclusive characteristic without the iris by the exclusive characteristic analysis module, the data output module directly stores the analysis result of the exclusive characteristic analysis module into a database; for the iris registration image extracted by the exclusive characteristic analysis module to have the exclusive characteristic of the iris, the data output module transmits the analysis result of the exclusive characteristic analysis module into the exclusive characteristic coding module of the iris.
In another preferred embodiment, as shown in fig. 4, the iris-specific feature extracting module extracts iris-specific features from the expanded iris region image of the iris registration image, and step d further includes:
the data pulling module pulls the iris registration image which is not processed by the iris exclusive characteristic extraction module from the database;
the first preprocessing module carries out preprocessing such as noise reduction and enhancement on the iris registration image pulled by the data pulling module;
the iris expansion module completes the conversion from the space coordinate to the polar coordinate, and expands the iris annular region image in the preprocessed iris registration image into a rectangular image;
the first segmentation module performs multi-scale segmentation on the unfolded rectangular image and extracts at least one stable characteristic region;
the exclusive characteristic analysis module comprehensively analyzes the gray distribution and morphological characteristics of the at least one characteristic region, decides at most 2 characteristic regions from the at least one characteristic region as the exclusive characteristics of the iris registration image, and marks the iris registration image without the exclusive characteristics of the iris as the exclusive characteristics without the iris if the exclusive characteristic analysis module fails to extract the exclusive characteristics of the iris from the at least one characteristic region after comprehensive analysis;
the data output module outputs the analysis result of the exclusive characteristic analysis module, and for the iris registration image marked as the exclusive characteristic without the iris by the exclusive characteristic analysis module, the data output module directly stores the analysis result of the exclusive characteristic analysis module into a database; for the iris registration image extracted by the exclusive characteristic analysis module to have the exclusive characteristic of the iris, the data output module transmits the analysis result of the exclusive characteristic analysis module into the exclusive characteristic coding module of the iris.
Preferably, the first segmentation module and the iris expansion module can also exchange orders, that is, the first segmentation module performs multi-scale segmentation on the preprocessed iris registration image and extracts at least one stable characteristic region, then the iris expansion module completes the conversion from a space coordinate to a polar coordinate, and maps the at least one stable characteristic region on the iris registration image into a rectangular image expanded by the iris annular region image.
In addition, in other embodiments, a certain number of feature regions may be determined from the at least one feature region according to actual requirements as the iris-specific features of the iris registration image.
The exclusive iris feature is extracted from the original image of the iris registration image, the original appearance of the exclusive iris feature can be reserved, but the original image comprises the iris, the pupil and a large area of skin, so that interference can be formed on the extraction of the exclusive iris feature, the iris is an annular area, and the calculation mode for extracting the feature in the annular area is more complex. The special iris features are extracted from the expanded iris area image of the iris registration image, compared with an annular image, a rectangular image is simpler to process, the expanded iris area image is only expanded aiming at the iris area, interference information of the expanded iris area image is relatively less, the expanded iris area image is expanded according to inner circle and outer circle coordinates of iris segmentation, and the special iris features can be distorted if segmentation errors are obvious. In other embodiments, the extraction mode of the exclusive features of the iris can be selected according to actual conditions.
Further, step e further comprises:
extracting the exclusive characteristics of the iris output by the upper module, wherein the exclusive characteristics of the iris registration image output by the exclusive characteristic extraction module of the iris are extracted;
extracting characteristic points of the exclusive characteristic outline of the iris, preferably including angular points, tangent points, inflection points and the like;
calculating 7 invariant moments of the exclusive characteristic profile of the iris;
and coding the exclusive characteristics of the iris according to the position information of the exclusive characteristics of the iris, the characteristic points and the invariant moment of the exclusive characteristic outline of the iris to form exclusive characteristic codes of the iris, and storing the exclusive characteristic codes of the iris into a database. Iris registration of the user is completed.
In this embodiment, when the user performs iris registration, only the information of the traditional iris features is collected and extracted, and after the user registration is completed, the system extracts the exclusive iris features offline, so that the registration time of the user is saved, and the user experience is prevented from being reduced.
As shown in fig. 5, step j further comprises:
the iris exclusive characteristic judging module judges whether an iris registration image matched with the iris identification image by the iris traditional characteristic code has the iris exclusive characteristic or not, if the iris exclusive characteristic does not exist, the iris exclusive characteristic comparing module directly calls the output module of the iris exclusive characteristic comparing module to directly transmit the output result of the iris traditional characteristic code comparing module to the identity output module; if the special iris feature exists, the special iris feature comparison module triggers the second preprocessing module to perform subsequent operation.
The second preprocessing module performs local processing on the iris recognition image at a corresponding position of the iris recognition image according to matching information of the iris recognition image and the iris registration image and position information of the iris exclusive feature in the iris registration image, wherein the local processing includes filtering, enhancing and the like in the embodiment.
The second segmentation module performs segmentation on the corresponding position of the iris identification image according to the scale of the exclusive features of the iris in the iris registration image, and extracts stable exclusive features of the iris;
and coding the exclusive iris features of the iris recognition image according to the coding mode of the exclusive iris features in the iris registration image, wherein the exclusive iris features output by the upper module are extracted from the iris recognition image extracted by the second segmentation module to form the exclusive iris feature code.
The exclusive characteristic matching module of iris matches the exclusive characteristic of iris discernment image and the exclusive characteristic of iris registration image, in this embodiment, mainly matches the characteristic point and 7 invariant moments of the exclusive characteristic profile of iris, and the matching result is exported for identity output module through output module.
In the application, when a user performs iris registration, an iris registration image is acquired, when the user performs iris recognition, an iris recognition image is acquired, iris registration is to analyze the iris registration image and extract traditional iris characteristics of the iris registration image, and if the iris registration image has special iris characteristics, the special iris characteristics of the iris registration image are also required to be extracted; the iris identification is to analyze the iris identification image and extract the traditional iris characteristics of the iris identification image, then compare the traditional iris characteristics of the iris identification image with the traditional iris characteristics of the iris registration image, if the iris identification image has the exclusive iris characteristics, the exclusive iris characteristics of the iris identification image also need to be extracted. In the iris registration process and the iris recognition process, the method for analyzing and extracting the traditional iris features is the same, but the method for analyzing and extracting the special iris features is completely different.
When the iris recognition is carried out, the iris traditional characteristics of the iris recognition image need to be rotated by a certain angle to be registered with the iris traditional characteristics of the iris registration image, if a certain characteristic of the iris is located at an upper position in the iris registration image and is located at a left position in the iris recognition image, when the iris recognition is carried out, the iris traditional characteristics are rotated by a certain angle to be located at the same position (the upper position or the left position) so as to compare the certain characteristic. After the iris traditional features are compared, the iris registration image and the iris identification image are subjected to angle registration, so that if a certain position on the iris registration image has the special iris features, the special iris features should exist at the position on the iris identification image under the condition that the iris traditional features are compared successfully. Therefore, when iris recognition is carried out, the exclusive iris feature in the iris recognition image is directly extracted according to the position and size information of the exclusive iris feature in the iris registration image, and the exclusive iris feature, namely the mode of extracting the exclusive iris feature from the iris registration image, is decided without searching a feature area.
The recognition of the exclusive characteristic of this application iris combines together with the recognition of the traditional characteristic of iris, on the basis that the traditional characteristic of iris accomplishes the matching, does benefit to its registration of accomplishing through the angular rotation, and the position and the size information of the exclusive characteristic of iris in the registered image directly carry out the extraction of the exclusive characteristic of iris in the relevant position of iris identification image, avoid screening once more, decision-making characteristic region causes the recognition efficiency to reduce.
The invention carries out double matching of the traditional iris characteristics and the exclusive iris characteristics on the user carrying the exclusive iris characteristics, thereby improving the accuracy of iris identification. Meanwhile, the quality of user experience is fully considered, the exclusive iris feature of the iris registration image is extracted in a background offline mode, and therefore when a user registers, only the traditional iris feature is extracted, and the registration time is the same as that of a traditional iris recognition system. In addition, during iris recognition, when the iris registration image with the iris exclusive characteristics is subjected to the iris exclusive characteristic extraction, the iris exclusive characteristics are locally extracted from the iris recognition image according to the matching information when the iris recognition image matched with the iris registration image is subjected to the iris exclusive characteristic extraction, the whole image is analyzed when the iris registration image is not adopted, then at least one characteristic region is extracted, and then the mode of the iris exclusive characteristics is decided, so that the time is greatly saved, and the user experience is not influenced.
While the present invention has been described in detail and with reference to the accompanying drawings and examples, it will be apparent to one skilled in the art that various changes and modifications can be made therein. Therefore, certain details of the embodiments are not to be interpreted as limiting, and the scope of the invention is to be determined by the appended claims.
Claims (10)
1. The utility model provides a system for reinforcing iris discernment precision, includes traditional characteristic module of iris, the exclusive characteristic module of iris and iris discernment module, its characterized in that, traditional characteristic module of iris includes:
the data acquisition module is used for acquiring an iris registration image or an iris identification image;
the quality evaluation module is used for evaluating whether the iris registration image or the iris identification image meets the quality requirement;
the iris segmentation module is used for segmenting the inner circle and the outer circle of the iris from the iris registration image or the iris identification image;
the iris coding module is used for coding the segmented iris registration image or iris identification image to form an iris traditional characteristic code;
the iris-specific feature module includes:
the iris exclusive characteristic extraction module is used for processing and analyzing the iris registration image acquired by the iris traditional characteristic module and extracting the iris exclusive characteristic of the iris registration image;
the special iris feature coding module is used for coding the special iris features of the iris registration image and the iris identification image to form special iris feature codes;
the iris recognition module includes:
the iris traditional characteristic comparison module is used for comparing and identifying the iris identification image and the iris traditional characteristic code of the iris registration image;
the iris exclusive characteristic comparison module is used for judging whether an iris registration image successfully matched with the iris identification image by the traditional iris characteristic code has the iris exclusive characteristic or not, and if the iris registration image has the iris exclusive characteristic, comparing and identifying the iris identification image and the iris exclusive characteristic code of the iris registration image;
and the identity output result is used for outputting the comparison result.
2. The system for enhancing iris recognition accuracy as claimed in claim 1, wherein the quality evaluation module comprises:
the primary quality evaluation module is used for evaluating the quality of the whole iris registration image or iris identification image acquired by the data acquisition module;
and the final quality evaluation module is used for evaluating the quality of the iris area image in the iris registration image or the iris identification image.
3. A method for enhancing iris recognition accuracy, comprising iris registration and iris recognition, wherein the iris registration comprises the following steps:
a. the data acquisition module acquires an iris registration image;
b. the iris segmentation module segments the inner circle and the outer circle of the iris in the iris registration image;
c. the iris coding module codes the segmented iris registration image to form an iris traditional characteristic code;
d. the special iris feature extraction module analyzes the iris registration image and extracts the special iris features of the iris registration image with the special iris features;
e. the special iris feature coding module codes the special iris features extracted from the iris registration image to form special iris feature codes;
the iris recognition comprises the following steps:
f. the data acquisition module acquires an iris recognition image;
g. the iris segmentation module segments the inner circle and the outer circle of the iris in the iris recognition image;
h. the iris coding module codes the segmented iris recognition image to form an iris traditional characteristic code;
i. the iris traditional characteristic comparison module compares and identifies the iris identification image with the iris traditional characteristic codes of the iris registration image;
j. the iris exclusive characteristic comparison module judges whether an iris registration image successfully matched with the iris identification image by traditional characteristic codes has iris exclusive characteristics or not, and if the iris registration image has the iris exclusive characteristics, the iris identification image and the iris exclusive characteristic codes of the iris registration image are compared and identified;
k. and the identity output module outputs a comparison result.
4. A method for enhancing iris recognition accuracy as claimed in claim 3, wherein: after the data acquisition module acquires the iris registration image or the iris identification image, the primary quality evaluation module is required to perform primary quality evaluation on the acquired iris registration image or iris identification image.
5. A method for enhancing iris recognition accuracy as claimed in claim 3, wherein: after the iris segmentation module completes segmentation of the inner and outer circles of the iris in the iris registration image or the iris identification image, the final quality evaluation module is required to evaluate the image quality of the iris region in the iris registration image or the iris identification image after the segmentation is completed.
6. The method for enhancing iris identification accuracy as claimed in claim 3, wherein the step d further comprises:
the data pulling module pulls the iris registration image;
the first preprocessing module carries out preprocessing such as noise reduction and enhancement on the iris registration image;
the first segmentation module performs multi-scale segmentation on the iris registration image and extracts a stable characteristic region;
the exclusive characteristic analysis module comprehensively analyzes the gray distribution and morphological characteristics of the iris registration image characteristic region and determines whether the iris exclusive characteristics exist or not;
and the data output module outputs the analysis result of the exclusive characteristic analysis module.
7. The method for enhancing iris identification accuracy as claimed in claim 3, wherein the step d further comprises:
the data pulling module pulls the iris registration image;
the first preprocessing module carries out preprocessing such as noise reduction and enhancement on the iris registration image;
the iris expansion module completes the conversion from the space coordinate to the polar coordinate and expands the iris annular region image in the iris registration image to form a rectangular image;
the first segmentation module performs multi-scale segmentation on the unfolded iris rectangular image and extracts a stable characteristic region;
the exclusive characteristic analysis module comprehensively analyzes the gray distribution and morphological characteristics of the iris registration image characteristic region and determines whether the iris exclusive characteristics exist or not;
and the data output module outputs the analysis result of the exclusive characteristic analysis module.
8. The method for enhancing iris identification accuracy as claimed in claim 3, wherein the step e further comprises:
extracting the exclusive iris feature of the iris registration image;
extracting characteristic points of the exclusive characteristic profile of the iris;
calculating the invariant moment of the exclusive characteristic profile of the iris;
and coding the exclusive characteristics of the iris according to the characteristic points and the invariant moment of the exclusive characteristic outline of the iris to form exclusive characteristic codes of the iris.
9. The method for enhancing iris recognition accuracy as claimed in claim 3, wherein the step j further comprises:
the iris exclusive characteristic judgment module judges whether an iris registration image matched with the iris identification image by the iris traditional characteristic code has the iris exclusive characteristic or not, and if the iris exclusive characteristic does not exist, the comparison result of the iris traditional characteristic code is transmitted to the identity output module;
if the iris registration image has the exclusive iris feature, the second preprocessing module carries out local processing on the corresponding position of the iris identification image according to the position information of the exclusive iris feature in the iris registration image;
the second segmentation module performs segmentation on the corresponding position of the iris identification image according to the scale of the exclusive features of the iris in the iris registration image, and extracts stable exclusive features of the iris;
the special iris feature coding module codes the special iris features of the extracted iris identification image to form special iris feature codes;
the special iris feature matching module matches the special iris feature code of the iris identification image with the special iris feature code of the iris registration image.
10. The method for enhancing iris recognition accuracy as claimed in claim 7, wherein: the exclusive characteristic analysis module decides at most 2 characteristic regions from the characteristic regions extracted by the first segmentation module as the exclusive characteristics of the iris registration image.
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2018108998495 | 2018-08-09 | ||
| CN201810899849.5A CN108537213A (en) | 2018-08-09 | 2018-08-09 | Enhance the system and method for iris recognition precision |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110826384A true CN110826384A (en) | 2020-02-21 |
Family
ID=63488226
Family Applications (2)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810899849.5A Pending CN108537213A (en) | 2018-08-09 | 2018-08-09 | Enhance the system and method for iris recognition precision |
| CN201910255205.7A Pending CN110826384A (en) | 2018-08-09 | 2019-04-01 | System and method for enhancing iris recognition accuracy |
Family Applications Before (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810899849.5A Pending CN108537213A (en) | 2018-08-09 | 2018-08-09 | Enhance the system and method for iris recognition precision |
Country Status (1)
| Country | Link |
|---|---|
| CN (2) | CN108537213A (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109214364A (en) * | 2018-10-24 | 2019-01-15 | 广州市真地信息技术有限公司 | A kind of iris identification method, system, device and storage medium |
| CN111553384A (en) * | 2020-04-03 | 2020-08-18 | 上海聚虹光电科技有限公司 | Matching method of multispectral image and single-spectral image |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20040037452A1 (en) * | 2000-10-07 | 2004-02-26 | Qritek Co., Ltd., | Iris identification system and method and computer readable storage medium stored therein computer executable instructions to implement iris identification method |
| CN103577813A (en) * | 2013-11-25 | 2014-02-12 | 中国科学院自动化研究所 | Information fusion method for heterogeneous iris recognition |
| CN106326841A (en) * | 2016-08-12 | 2017-01-11 | 合肥虹视信息工程有限公司 | Quick iris recognition algorithm |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN100342390C (en) * | 2004-04-16 | 2007-10-10 | 中国科学院自动化研究所 | Identity identifying method based on iris plaque |
| US8369595B1 (en) * | 2012-08-10 | 2013-02-05 | EyeVerify LLC | Texture features for biometric authentication |
-
2018
- 2018-08-09 CN CN201810899849.5A patent/CN108537213A/en active Pending
-
2019
- 2019-04-01 CN CN201910255205.7A patent/CN110826384A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20040037452A1 (en) * | 2000-10-07 | 2004-02-26 | Qritek Co., Ltd., | Iris identification system and method and computer readable storage medium stored therein computer executable instructions to implement iris identification method |
| CN103577813A (en) * | 2013-11-25 | 2014-02-12 | 中国科学院自动化研究所 | Information fusion method for heterogeneous iris recognition |
| CN106326841A (en) * | 2016-08-12 | 2017-01-11 | 合肥虹视信息工程有限公司 | Quick iris recognition algorithm |
Non-Patent Citations (1)
| Title |
|---|
| 孟浩;徐翠平;: "虹膜识别算法的研究" * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108537213A (en) | 2018-09-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108921100B (en) | Face recognition method and system based on visible light image and infrared image fusion | |
| Gottschlich | Curved-region-based ridge frequency estimation and curved Gabor filters for fingerprint image enhancement | |
| EP3680794B1 (en) | Device and method for user authentication on basis of iris recognition | |
| JP2004206444A (en) | Individual authentication method and iris authentication device | |
| WO2009029638A1 (en) | Iris recognition | |
| CN102254188A (en) | Palmprint recognizing method and device | |
| CN112036383A (en) | Hand vein-based identity recognition method and device | |
| Abidin et al. | Iris segmentation analysis using integro-differential and hough transform in biometric system | |
| CN111178130A (en) | Face recognition method, system and readable storage medium based on deep learning | |
| Lee et al. | An automated video-based system for iris recognition | |
| Jain et al. | Fingerprint image analysis: role of orientation patch and ridge structure dictionaries | |
| CN110929570B (en) | Iris rapid positioning device and positioning method thereof | |
| CN110826384A (en) | System and method for enhancing iris recognition accuracy | |
| CN117593293B (en) | Intelligent processing system and method for nasal bone fracture image | |
| CN109523484B (en) | Fractal feature-based finger vein network repair method | |
| Wang et al. | Hand vein recognition based on improved template matching | |
| JP2005259049A (en) | Face collation device | |
| Latha et al. | A robust person authentication system based on score level fusion of left and right irises and retinal features | |
| CN108288040A (en) | Multi-parameter face identification system based on face contour | |
| Resmi et al. | An empirical study and evaluation on automatic ear detection | |
| Amjed et al. | Noncircular iris segmentation based on weighted adaptive hough transform using smartphone database | |
| CN111611848B (en) | Cadaver iris recognition method and device | |
| Leo et al. | Highly usable and accurate iris segmentation | |
| KAMIL et al. | An Improved Iris Segmentation using Pupil Candidate Bank for Non-Cooperative Iris Recognition | |
| Rossant et al. | A robust iris identification system based on wavelet packet decomposition and local comparisons of the extracted signatures |
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 | ||
| AD01 | Patent right deemed abandoned | ||
| AD01 | Patent right deemed abandoned |
Effective date of abandoning: 20231229 |