CN111241888A - Iris living body detection method based on infrared lamp flicker - Google Patents
Iris living body detection method based on infrared lamp flicker Download PDFInfo
- Publication number
- CN111241888A CN111241888A CN201811446375.5A CN201811446375A CN111241888A CN 111241888 A CN111241888 A CN 111241888A CN 201811446375 A CN201811446375 A CN 201811446375A CN 111241888 A CN111241888 A CN 111241888A
- Authority
- CN
- China
- Prior art keywords
- activity
- flashing
- iris
- infrared lamp
- living body
- 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
- 238000001514 detection method Methods 0.000 title claims abstract description 36
- 230000000694 effects Effects 0.000 claims abstract description 83
- 239000013598 vector Substances 0.000 claims abstract description 76
- 238000013145 classification model Methods 0.000 claims abstract description 13
- 238000010801 machine learning Methods 0.000 claims abstract description 6
- 210000001747 pupil Anatomy 0.000 claims description 9
- 210000000554 iris Anatomy 0.000 description 65
- 238000000034 method Methods 0.000 description 9
- 238000004458 analytical method Methods 0.000 description 2
- 210000005252 bulbus oculi Anatomy 0.000 description 2
- 238000001727 in vivo Methods 0.000 description 2
- 230000004478 pupil constriction Effects 0.000 description 2
- 230000002159 abnormal effect Effects 0.000 description 1
- 238000001574 biopsy Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000010183 spectrum analysis Methods 0.000 description 1
- 230000000638 stimulation Effects 0.000 description 1
- 230000001960 triggered effect Effects 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
- G06V40/197—Matching; Classification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
Abstract
The invention provides an iris living body detection method based on infrared lamp flickering, which comprises the following steps: s1, establishing a flicker rule base; s2, collecting a first activity feature vector of a positive sample and a second activity feature vector of a negative sample according to a scintillation rule in a scintillation rule base, wherein the first activity feature vector and the second activity feature vector are used as training samples of the scintillation rule, and the training samples are trained through a machine learning algorithm to obtain a classification model; s3, selecting a flashing rule in a flashing rule base, and collecting a third activity characteristic vector of the current iris recognition object when the infrared lamp flashes according to the flashing rule; the S4 classification model predicts whether the iris recognition object is a living body according to the third activity feature vector.
Description
Technical Field
The invention relates to the technical field of in-vivo detection, in particular to an iris in-vivo detection method based on infrared lamp flickering.
Background
Because image data is processed in iris recognition, iris fraud is caused by means of pictures, contact lenses engraved with iris textures, 3D eyeball models and the like, so that iris anti-counterfeiting is carried forward, but the existing iris anti-counterfeiting method has the following defects:
1. the method of the frequency spectrum analysis, the iris image printed with high definition or the iris image with motion blur can not be judged accurately; 2. a method of reflectance information analysis that will fail for a printed iris image worn with contact lenses; 3. the texture analysis method is characterized in that when an iris image is near a fuzzy and clear boundary region, the method cannot accurately judge whether the iris image is a living body; 4. the method for detecting the pupil constriction needs strong light to stimulate the pupil, so that the pupil constriction is caused to achieve the aim of living body detection, the user experience is poor, and the user can hardly accept the method. Therefore, a safer and more reliable iris biopsy method is urgently needed to be provided.
Disclosure of Invention
In order to solve the technical problem, the invention provides an iris living body detection method based on infrared lamp flickering, which performs living body detection through an infrared lamp and a camera of iris recognition equipment, and comprises the following steps:
s1, establishing a flicker rule base;
s2, collecting a first activity feature vector of a positive sample and a second activity feature vector of a negative sample according to a scintillation rule in a scintillation rule base, wherein the first activity feature vector and the second activity feature vector are used as training samples of the scintillation rule, and the training samples are trained through a machine learning algorithm to obtain a classification model;
s3, selecting a flashing rule in a flashing rule base, and collecting a third activity characteristic vector of the current iris recognition object when the infrared lamp flashes according to the flashing rule;
the S4 classification model predicts whether the iris recognition object is a living body according to the third activity feature vector.
The improvement of the iris living body detection method based on infrared lamp flickering is that the flickering rule base comprises at least one group of flickering rules, and the flickering rules are rules of infrared lamp flickering.
The iris living body detection method based on infrared lamp flickering is further improved in that the positive sample refers to a living body, and the negative sample refers to a prosthesis.
The iris living body detection method based on infrared lamp flashing is further improved in that step S2 further comprises:
s21 selecting a group of flashing rules in the flashing rule base;
s22 the infrared lamp flickers according to the flickering rule, and the camera records the video of the positive sample or the negative sample and intercepts n frames of images;
s23, calculating and analyzing the activity characteristic vectors of the n frames of images, forming a first activity characteristic vector of a positive sample or a second activity characteristic vector of a negative sample, and storing the first activity characteristic vector or the second activity characteristic vector in a database;
and S24, repeating the steps until the first activity characteristic vectors and the second activity characteristic vectors corresponding to all the scintillation rules in the scintillation database are collected.
The iris living body detection method based on infrared lamp flickering is further improved in that the step S23 further comprises the following steps:
calculating the activity characteristics of each frame of image in the n frames of images to obtain n activity characteristics, wherein the n activity characteristics form the activity characteristic vectors of the n frames of images;
and normalizing the activity characteristic vector to obtain a first activity characteristic vector of the positive sample or a second activity characteristic vector of the negative sample.
The iris living body detection method based on infrared lamp flicker is further improved in that the activity characteristic refers to a product obtained by multiplying the average gray value of each frame of image, the number of the reflective points of the pupil and iris area in each frame of image and the average gray value of the reflective points.
The iris living body detection method based on infrared lamp flickering is further improved in that the activity characteristic vectors are normalized by adopting L2-norm.
The iris living body detection method based on infrared lamp flickering is further improved in that the number of the infrared lamps is at least two.
The iris living body detection method based on infrared lamp flickering is further improved in that the flickering rule comprises a flickering state and a flickering time length.
The iris living body detection method based on infrared lamp flickering can realize living body detection based on the original infrared lamp and camera of the iris recognition equipment, does not need to add hardware equipment, has low cost and good universality on the iris recognition equipment, does not need active cooperation of a user, has no abnormal stimulation on the user, can effectively resist photo attack, video attack and 3D eyeball model attack, has high living body detection accuracy, and achieves better technical effect on the premise of not improving the hardware cost.
Drawings
FIG. 1 is a schematic flow chart of an iris living body detection method based on infrared lamp flickering.
Fig. 2 is a schematic flow chart of the first activity feature vector acquisition.
Detailed Description
The present invention is described in further detail below with reference to the attached drawings.
When the infrared lamp flickers, the living iris changes along with the different flickering states of the infrared lamp, and the change of the false iris along with the different flickering states of the infrared lamp is different from the change of the living iris, so that the iris living body can be detected by using the flickering of the infrared lamp.
The invention provides an iris living body detection method based on infrared lamp flickering, which is characterized in that living body detection is carried out through an infrared lamp and a camera of iris recognition equipment, when a user carries out iris recognition on the iris recognition equipment, the iris recognition equipment firstly carries out living body detection on the user, and as shown in figure 1, the iris living body detection method comprises the following steps:
s1, establishing a flicker rule base;
the flashing rule base comprises at least one group of flashing rules, the flashing rules refer to flashing rules of the infrared lamp, in the embodiment, the flashing rule base comprises 5 groups of flashing rules, and the number of the flashing rules can be set according to requirements in other embodiments. The flashing rule comprises a flashing state and flashing time, wherein the flashing state refers to the on and off of the infrared lamp, the flashing time refers to the on and off time of the infrared lamp, the on and off state switching of the infrared lamp and the duration time combination of different states form the flashing rule. When the infrared lamp executes a group of flashing rules, the infrared lamp flashes according to the flashing state and the flashing time length specified by the group of flashing rules. In the present embodiment, the number of the infrared lamps is set to 2, so that the flashing rule includes the flashing state and the flashing time length of each infrared lamp, and in other embodiments, the number of the infrared lamps can be set according to actual requirements.
S2, collecting a first activity feature vector of a positive sample and a second activity feature vector of a negative sample according to a scintillation rule in a scintillation rule base, wherein the first activity feature vector and the second activity feature vector are used as training samples of the scintillation rule, and the training samples are trained through a machine learning algorithm to obtain a classification model; the positive sample refers to a living iris, and the negative sample refers to a prosthetic iris, such as a photograph, iris, and the like.
In connection with fig. 2, in particular,
s21 selecting a group of flashing rules in the flashing rule base;
s22 the infrared lamp flickers according to the flicking rule, the camera records the video of the positive sample or the negative sample while the infrared lamp flickers, namely the camera records the video of the living body iris or the false body iris when the infrared lamp flickers according to the flicking rule, and intercepts the n frames of images of the recorded video;
s23, calculating the active characteristics of each frame of image in n frames of images, preferably, the active characteristics refer to the product of the average gray value of the whole image of each frame of image, the number of the reflective points of the pupil and the iris area in each frame of image and the average gray value of the reflective points, namely, the average gray value of the whole image in each frame of image in n frames of image, the number of the reflective points of the pupil and the iris area in each frame of image and the average gray value of the reflective points of the pupil and the iris area in each frame of image are calculated and multiplied to obtain the active characteristics of each frame of image, wherein n active characteristics can be obtained by n frames of images, and form the active characteristic vector of the n frames of images;
normalizing the activity characteristic vector by using L2-norm, or normalizing the activity characteristic vector by using other suitable methods in other embodiments to obtain a normalized activity characteristic vector, forming a first activity characteristic vector of a positive sample or a second activity characteristic vector of a negative sample, and storing the first activity characteristic vector or the second activity characteristic vector in a database;
s24, repeating the steps until the first activity characteristic vector and the second activity characteristic vector corresponding to all the scintillation rules in the scintillation rule base are collected, wherein the first activity characteristic vector and the second activity characteristic vector are used as training samples, then inputting the training samples into an svm algorithm for training, and obtaining a classification model through a machine learning mode.
S3, randomly selecting a group of flashing rules in the flashing rule base, collecting a third activity characteristic vector of the current user iris identification object when the infrared lamp flashes according to the flashing rules, wherein the collection and calculation of the third activity characteristic vector are similar to the second activity characteristic vector and the first activity characteristic vector, namely:
s31 selecting a group of flashing rules in the flashing rule base;
s32 the infrared lamp flickers according to the flickering rule, the camera records the video of the current iris recognition object while the infrared lamp flickers, and captures n frames of images of the recorded video;
s23, calculating the active characteristics of each frame of image in n frames of images, preferably, the active characteristics refer to the product of the average gray value of the whole image of each frame of image, the number of the reflective points of the pupil and the iris area in each frame of image and the average gray value of the reflective points, namely, the average gray value of the whole image in each frame of image in n frames of image, the number of the reflective points of the pupil and the iris area in each frame of image and the average gray value of the reflective points of the pupil and the iris area in each frame of image are calculated and multiplied to obtain the active characteristics of each frame of image, wherein n active characteristics can be obtained by n frames of images, and form the active characteristic vector of the n frames of images;
normalizing the activity characteristic vector by using L2-norm, or normalizing the activity characteristic vector by using other suitable methods in other embodiments to obtain a normalized activity characteristic vector, forming a third activity characteristic vector of the current iris recognition object, and storing the third activity characteristic vector in a database;
the S4 classification model predicts whether the iris recognition object is a living body according to the third activity characteristic vector;
specifically, the third activity feature vector obtained in S3 is input to the svm algorithm, and the classification model obtained in S2 is used to predict the third activity feature vector, so as to obtain a prediction result.
Further, for each flashing rule, the same n frames of images need to be intercepted from the video of the positive sample, the negative sample and the iris identification object collected under the flashing rule, for example, the flashing rule corresponds to one flashing cycle, the n frames of images in the same time period in the flashing cycle of the infrared lamp need to be intercepted, in other embodiments, the situation that the infrared lamp flashes for several flashing cycles may exist, at this time, the n frames of images in a certain time period in which the flashing state of the infrared lamp is the same can be intercepted, and the accuracy of data is ensured, and the n frames of images are continuous n frames of images. Preferably, in this embodiment, when the infrared lamp flashes according to the flashing rule, the camera is synchronously triggered to start recording the video, that is, in the same flashing rule, the states of the infrared lamps corresponding to the start of recording the video are the same, so in order to save time and ensure the accuracy of capturing n frames of images, in this embodiment, the first n frames of images of the video are captured uniformly.
Further, in this embodiment, the number of the cameras is one, in other embodiments, the number of the cameras may be multiple, when the number of the cameras is multiple, a first activity feature vector and a second activity feature vector corresponding to each camera need to be collected in advance, and a classification model of each camera is obtained, when iris recognition is performed on an iris recognition object, each camera collects a third activity feature vector of the iris recognition object, predicts the third activity feature vector of each camera through a corresponding classification model thereof, obtains a prediction result of whether the iris recognition object is a living body, and comprehensively analyzes the prediction result of each camera to obtain a final result.
The iris living body detection method based on infrared lamp flashing establishes flashing rules, carries out living body detection according to different characteristics of living body irises and false body irises with different activity characteristics under different flashing rules, establishes a classification model by utilizing a machine learning method to carry out living body detection on iris recognition objects, is simple, convenient and quick, and does not need to add other parts on iris recognition equipment.
It is to be understood that the described embodiments are merely exemplary of the invention, and not restrictive of the full scope of the invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Claims (9)
1. An iris living body detection method based on infrared lamp flickering is characterized by comprising the following steps of:
s1, establishing a flicker rule base;
s2, collecting a first activity feature vector of a positive sample and a second activity feature vector of a negative sample according to a scintillation rule in a scintillation rule base, wherein the first activity feature vector and the second activity feature vector are used as training samples of the scintillation rule, and the training samples are trained through a machine learning algorithm to obtain a classification model;
s3, selecting a flashing rule in a flashing rule base, and collecting a third activity characteristic vector of the current iris recognition object when the infrared lamp flashes according to the flashing rule;
the S4 classification model predicts whether the iris recognition object is a living body according to the third activity feature vector.
2. The iris living body detection method based on infrared lamp flickering as claimed in claim 1, wherein: the flashing rule base comprises at least one group of flashing rules, and the flashing rules are rules of flashing of infrared lamps.
3. The iris living body detection method based on infrared lamp flickering as claimed in claim 1, wherein: the positive sample refers to a living body, and the negative sample refers to a prosthesis.
4. The iris liveness detection method based on infrared lamp flash as claimed in claim 1, wherein the step S2 further comprises:
s21 selecting a group of flashing rules in the flashing rule base;
s22 the infrared lamp flickers according to the flickering rule, and the camera records the video of the positive sample or the negative sample and intercepts n frames of images;
s23, calculating and analyzing the activity characteristic vectors of the n frames of images, forming a first activity characteristic vector of a positive sample or a second activity characteristic vector of a negative sample, and storing the first activity characteristic vector or the second activity characteristic vector in a database;
and S24, repeating the steps until the first activity characteristic vectors and the second activity characteristic vectors corresponding to all the scintillation rules in the scintillation database are collected.
5. The iris liveness detection method based on infrared lamp flash as claimed in claim 4, wherein the step S23 further comprises:
calculating the activity characteristics of each frame of image in the n frames of images to obtain n activity characteristics, wherein the n activity characteristics form the activity characteristic vectors of the n frames of images;
and normalizing the activity characteristic vector to obtain a first activity characteristic vector of the positive sample or a second activity characteristic vector of the negative sample.
6. The iris living body detection method based on infrared lamp flickering as claimed in claim 5, wherein: the active characteristic refers to the product obtained by multiplying the average gray value of each frame of image, the number of the reflective points of the pupil and iris area in each frame of image and the average gray value of the reflective points.
7. The iris living body detection method based on infrared lamp flickering as claimed in claim 5, wherein: the activity feature vectors were normalized using L2-norm.
8. The iris living body detection method based on infrared lamp flickering as claimed in claim 1, wherein: the number of the infrared lamps is at least two.
9. The iris living body detection method based on infrared lamp flickering as claimed in claim 1, wherein: the flashing rule comprises a flashing state and a flashing time length.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811446375.5A CN111241888A (en) | 2018-11-29 | 2018-11-29 | Iris living body detection method based on infrared lamp flicker |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811446375.5A CN111241888A (en) | 2018-11-29 | 2018-11-29 | Iris living body detection method based on infrared lamp flicker |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111241888A true CN111241888A (en) | 2020-06-05 |
Family
ID=70868395
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811446375.5A Pending CN111241888A (en) | 2018-11-29 | 2018-11-29 | Iris living body detection method based on infrared lamp flicker |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111241888A (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080095411A1 (en) * | 2006-09-29 | 2008-04-24 | Wen-Liang Hwang | Iris recognition method |
| CN101833646A (en) * | 2009-03-11 | 2010-09-15 | 中国科学院自动化研究所 | Living iris detection method |
| CN105354545A (en) * | 2015-10-28 | 2016-02-24 | 广东欧珀移动通信有限公司 | Iris information acquisition method and acquisition system |
-
2018
- 2018-11-29 CN CN201811446375.5A patent/CN111241888A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080095411A1 (en) * | 2006-09-29 | 2008-04-24 | Wen-Liang Hwang | Iris recognition method |
| CN101833646A (en) * | 2009-03-11 | 2010-09-15 | 中国科学院自动化研究所 | Living iris detection method |
| CN105354545A (en) * | 2015-10-28 | 2016-02-24 | 广东欧珀移动通信有限公司 | Iris information acquisition method and acquisition system |
Non-Patent Citations (2)
| Title |
|---|
| 刘建征;杨华易;: "基于闪烁光照的视频回放攻击检测方法" * |
| 程宇奇;葛微;陈彦平;: "虹膜身份识别技术" * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107122744B (en) | Living body detection system and method based on face recognition | |
| CN101271517B (en) | Face region detecting device and method | |
| CN108197546B (en) | Illumination processing method and device in face recognition, computer equipment and storage medium | |
| CN106022209B (en) | A kind of method and device of range estimation and processing based on Face datection | |
| EP2959454B1 (en) | Method, system and software module for foreground extraction | |
| JP6521845B2 (en) | Device and method for measuring periodic fluctuation linked to heart beat | |
| CN108288027A (en) | A kind of detection method of picture quality, device and equipment | |
| CN108389207A (en) | A kind of the tooth disease diagnosing method, diagnostic device and intelligent image harvester | |
| CN106778695A (en) | A kind of many people's examing heartbeat fastly methods based on video | |
| CN105163110A (en) | Camera cleanliness detection method and system and shooting terminal | |
| CN109697716B (en) | Identification method and equipment of cyan eye image and screening system | |
| CN106886216A (en) | Robot automatic tracking method and system based on RGBD Face datections | |
| CN107392153B (en) | Human body fatigue degree judging method | |
| CN105105738A (en) | Information processing method and device and electronic device | |
| CN109508648A (en) | A kind of face snap method and apparatus | |
| CN108742656A (en) | Fatigue state detection method based on face feature point location | |
| CN107431761A (en) | Image processing equipment, image processing method and image processing system | |
| RU2556417C2 (en) | Detecting body movements using digital colour rear projection | |
| CN111353326A (en) | In-vivo detection method based on multispectral face difference image | |
| CN113596344A (en) | Shooting processing method and device, electronic equipment and readable storage medium | |
| CN108961209A (en) | Pedestrian image quality evaluating method, electronic equipment and computer-readable medium | |
| CN110910449A (en) | Method and system for recognizing three-dimensional position of object | |
| CN114140775A (en) | Data recording and physiological state detection method, device, equipment and storage medium | |
| CN112784740B (en) | Gait data acquisition and labeling method and application | |
| JPH08225028A (en) | Awakening condition detecting device for driver |
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 |
Effective date of abandoning: 20231229 |
|
| AD01 | Patent right deemed abandoned |