CN101814129B - Automatically focused remote iris image acquisition device, method and recognition system - Google Patents
Automatically focused remote iris image acquisition device, method and recognition system Download PDFInfo
- Publication number
- CN101814129B CN101814129B CN 200910077372 CN200910077372A CN101814129B CN 101814129 B CN101814129 B CN 101814129B CN 200910077372 CN200910077372 CN 200910077372 CN 200910077372 A CN200910077372 A CN 200910077372A CN 101814129 B CN101814129 B CN 101814129B
- Authority
- CN
- China
- Prior art keywords
- focusing
- image
- iris
- sharpness
- value
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 50
- 230000007704 transition Effects 0.000 claims abstract description 25
- 238000012545 processing Methods 0.000 claims abstract description 20
- 230000008859 change Effects 0.000 claims abstract description 5
- 230000007246 mechanism Effects 0.000 claims description 26
- 238000004364 calculation method Methods 0.000 claims description 16
- 230000008569 process Effects 0.000 claims description 16
- 238000006243 chemical reaction Methods 0.000 claims description 11
- 238000001514 detection method Methods 0.000 claims description 7
- 238000007792 addition Methods 0.000 claims description 6
- 239000000284 extract Substances 0.000 claims description 4
- 238000000605 extraction Methods 0.000 claims description 4
- 230000000007 visual effect Effects 0.000 claims description 3
- 238000013459 approach Methods 0.000 claims description 2
- 230000000903 blocking effect Effects 0.000 claims description 2
- 238000004422 calculation algorithm Methods 0.000 claims description 2
- 238000001914 filtration Methods 0.000 claims description 2
- 238000010801 machine learning Methods 0.000 claims description 2
- 238000005070 sampling Methods 0.000 claims description 2
- 230000003068 static effect Effects 0.000 claims description 2
- 238000012549 training Methods 0.000 claims description 2
- 235000002911 Salvia sclarea Nutrition 0.000 abstract 1
- 244000182022 Salvia sclarea Species 0.000 abstract 1
- 238000010586 diagram Methods 0.000 description 11
- 238000013461 design Methods 0.000 description 4
- 238000002474 experimental method Methods 0.000 description 2
- NIXOWILDQLNWCW-UHFFFAOYSA-N acrylic acid group Chemical group C(C=C)(=O)O NIXOWILDQLNWCW-UHFFFAOYSA-N 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 239000011521 glass Substances 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000011017 operating method Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000003909 pattern recognition Methods 0.000 description 1
- 238000010187 selection method Methods 0.000 description 1
- 230000006641 stabilisation Effects 0.000 description 1
- 238000011105 stabilization Methods 0.000 description 1
Images
Abstract
The invention relates to an automatically focused remote iris image acquisition device, an automatically focused remote iris image acquisition method and a recognition system. The device comprises an iris camera, a long-focus lens, a focusing drive unit, a lens converter, a focusing rotating machine, a motor drive unit and a computational processing unit; and the automatic focusing method based on the device comprises the following steps of: initializing focusing equipment; continuously acquiring iris images by the camera; computing the definition and definition change of the acquired images by the computational processing unit; determining the focusing speed and direction according to the current state and state transition condition of the system; and detecting clear eye images from the image sequence for iris recognition. Based on the focusing device and the focusing method, the range of depth of field and speed of acquiring remote iris images can be increased greatly. Furthermore, the remote iris image recognition system using the method and the device is more convenient and easy to use.
Description
Technical field
The invention belongs to image processing, Electromechanical Control, computer vision and mode identification technology, relate to the iris recognition that is applied to the living things feature recognition field, remote iris image acquisition device and the method for focusing automatically.
Background technology
In present biometrics identification technology, iris recognition technology has that accuracy is high, uniqueness good, antifalsification is strong, be easy to the advantage such as image processing, has market application foreground widely.The step of iris recognition generally comprises the steps such as iris image acquisition, image pre-service, Iris Location, live body detection, feature extraction and characteristic matching.Wherein, iris image acquisition is a very important link, does well if iris obtains, and can make image acquisition speed slow, and greatly reduces discrimination.
Iris image acquisition is very difficult, and its main cause is because the diameter of iris is very little, and the resolution that iris recognition requires is high, and this makes the image of optical system larger than very, so the optics depth of field is very little, often only has several centimetres of left and right.In this field depth, be difficult to allow the user aim at easily, this has had a strong impact on ease for use and the recognition speed of iris recognition.
All adopt tight shot in initial iris image acquiring patented product.(Chinese patent CN1282048, CN1584917 etc.) tight shot depth of field is very little, can only be satisfied with some to the less demanding occasion of ease for use.In order to allow the people can rapid alignment, adopted sound and light signal to feed back to the people in some patented product, allow the people aim at according to the instruction of machine, but this method still can not be from improving the ease for use of iris collection device in essence.
Camera lens with autozoom or focusing automatically in some newer patented products comes extended depth of field.(Chinese patent: CN1892401, CN2672768, CN1894719 etc.) these are all very useful methods, but present focusing product still has several problems: first, be all the in-plant focus apparatus that adopts, automatically focus in fixing distance and limited scope; The second, in focusing system, after the focal position changes, can not follow the tracks of fast new focal position, adaptivity is poor.Trace it to its cause, they are most of not specially for iris image design focusing mechanism, and adopt the apparatus and method on digital camera or monitoring camera; And the shooting of iris is similar to micro-imaging, requires that focusing range is large and precision is high, thus these equipment inapplicable be very natural.Must make the automatic focusing mechanism that is applicable to distant range iris identification.
Summary of the invention
The technical matters that (one) will solve
The objective of the invention is to design a kind of remote iris image acquisition device, method and recognition system of automatic focusing, to increase the field depth of distant range iris image acquisition, improve the ease for use of iris recognition.
(2) technical scheme
For achieving the above object, the invention provides a kind of remote iris image acquisition device of automatic focusing.This device comprises following structure: iris camera, long focal length lens, focusing gear unit, lens converter, focusing electric rotating machine, electric-motor drive unit, calculation processing unit, wherein: iris camera, long focal length lens and lens converter are positioned on same axis, the focusing gear unit with focus that electric rotating machine is coaxial to be connected; The iris camera connects by lens converter with long focal length lens; The focusing ring that the focusing gear unit is connected with long focal length lens connects; Focusing electric rotating machine and electric-motor drive unit electrical connection; Calculation processing unit is connected with the electric-motor drive unit signal, controls the rotation of focusing electric rotating machine by electric-motor drive unit; Calculation processing unit is connected with the iris camera, processes the image from the iris camera.
For achieving the above object, the invention provides a kind of distant range iris image acquiring method of automatic focusing, comprise the steps:
Step P1: initialization apparatus, the rotation focusing ring makes the focusing ring focusing position to zero point, i.e. limit switch 11 places;
Step P2: the iris camera constantly gathers image, forms image sequence;
Step P3: the changing value of the sharpness of every two field picture and sharpness in the calculation processing unit sequence of computed images, difference execution in step P4 and step P5 in two threads;
Step P4: according to sharpness and the sharpness changing value of image, direction and the speed of the rotation of conversion focusing ring are returned to step P2; Described conversion is according to sharpness and sharpness changing value, and the automatic current system state of the remote iris image acquisition device of focusing, is decided by the system state switch condition of the remote iris image acquisition device of automatic focusing;
Step P5: change image over to and select thread, the sharpness of image and sharpness changing value in sequence of computed images, and judge whether to pick out by sharpness and the sharpness changing value of image the image that sharpness meets the demands, if the clear picture of selecting, execution in step P6, if that selects is not fogging clear, get back to step P2;
Step P6: carry out human eye detection and extract eye image from the picture rich in detail of picking out;
Step P7: eye image is carried out the quality judgement, judge whether to be used for iris recognition, if the judgement eye image can be used in iris recognition, iris recognition finishes, if the judgement eye image can not be used for iris recognition, continue to gather image, continue execution in step P5.
For achieving the above object, the invention provides a kind of full automatic distant range iris recognition system based on above apparatus and method, comprise following structure: far distance automatic focusing mechanism, remote infrared light supply, driving governor, light source controller, image pick-up card, computing machine, panel and rack, wherein: the focusing electric rotating machine of far distance automatic focusing mechanism is electrically connected on computing machine by driving governor; Iris camera on far distance automatic focusing mechanism is connected on computing machine by image pick-up card; The high-definition picture of iris camera collection is by on passing to; By computing machine calculate camera collection to the sharpness of realtime graphic, and according to the far distance automatic focusing mechanism of variation automatic rotating of sharpness, the focal position of search and tracking object; Remote infrared light supply is connected on computing machine by controller; Above equipment all is installed in rack, and far distance automatic focusing mechanism and remote infrared light supply are all installed towards the front, rack front plate placement.
(3) beneficial effect
The present invention combines the technology such as image processing, Electromechanical Control, computer vision and pattern-recognition, is applied to the iris recognition direction in living things feature recognition field.The present invention is carrying out iris image acquiring at a distance by the method for automatic focusing, improves iris image acquiring scope and picking rate, has improved the ease for use of iris recognition.
The scheme that the present invention adopts, the first, the redesign focusing mechanism makes it take iris image by focusing in larger scope, farther distance.The second, design is the focusing strategy better, can satisfy large-scale focus search, can satisfy high-precision focusing again, and follows the tracks of fast can work as focal variation the time.
Focusing mechanism has adopted exquisite mechanical hook-up, goes for multiple long focal length lens; Drive focusing ring with stepper motor, focusing range is large, slow-roll stabilization and fast; Adopt the strategy of the focusing of state-based conversion, can control focusing mechanism fast search and tracking focus position; Adopted the sharpness of image processing techniques judgement image and detected human eye, having obtained iris image.
Description of drawings
Fig. 1 is the structural drawing of automatic focusing mechanism;
Fig. 2 is the schematic diagram of focusing gearing;
Fig. 3 is the method flow diagram of automatically focusing;
Fig. 4 is the focusing strategy schematic diagram that utilizes focusing ring 21;
Fig. 5 is that sharpness is with the schematic diagram of variable in distance;
The schematic diagram that when Fig. 6 is movement of objects, the focal position changes;
Fig. 7 is the sharpness computation flow process;
Fig. 8 is the schematic diagram of example device;
Fig. 9 is the operating procedure process flow diagram of instance system;
The critical piece description of symbols
Focusing ring 21 focusing gear units 3
Electric-motor drive unit 6 calculation processing units 7
The remote infrared light supply D2 of far distance automatic focusing mechanism D1
Driving controller of stepping motor D3 light source controller D4
Image pick-up card D5 main frame D6
Panel D7 rack D8
Embodiment
Describe each related detailed problem in technical solution of the present invention in detail below in conjunction with accompanying drawing.Be to be noted that described embodiment only is intended to be convenient to the understanding of the present invention, and it is not played any restriction effect.
The remote iris image acquisition device of (one) automatically focusing
Structural design of the present invention such as Fig. 1 are comprised of following and part: iris camera 1, long focal length lens 2, focusing gear unit 3, lens converter 4, focusing electric rotating machine 5, electric-motor drive unit 6, calculation processing unit 7.
Iris camera 1 is connected with camera lens by lens converter 4 connections; Focusing gear unit 3 connects with focusing ring 21 and focusing electric rotating machine 5 on camera lens is connected; Focusing electric rotating machine 5 and electric-motor drive unit 6 electrical connections; Calculation processing unit 7 is connected signal and is connected with electric-motor drive unit, can control by electric-motor drive unit 6 rotation of focusing electric rotating machines 5; Calculation processing unit 7 is connected connection with the iris camera, process the image from camera 1.
The focal length value of long focal length lens 2 is that 100mm is to the arbitrary value between 500mm.Long focal length lens 2 can be connected connection by lens converter 4 with the iris camera.
Wherein, focusing gear unit 3 structural drawing such as Fig. 2.One cavity ring 31 and a jig 32 are arranged on described focusing gear unit 3, have the elasticity three-jaw jig 32, one jigs 32 are installed on a cavity ring 31 and between focusing ring 21, be used for blocking in a described cavity ring 31 and camera lens 2 and the focusing rings 21 of different sizes; Described focusing gear unit 3 also has a belt 33 and a limit switch 34, the be locked rotating shaft of cavity ring 31 and the focusing electric rotating machine 5 that is locked of a described belt 33, like this when 5 rotation of focusing electric rotating machine, also can drive cavity ring 31 rotations, focus thereby drive 21 pairs of camera lenses 2 of focusing ring; A described limit switch 34 will be run into when focusing ring 21 rotates to a position, and limit switch 34 sends a signal can make stepper motor 5 stop operating, and so just makes focusing gear unit 3 that an initial position has been arranged.
The distant range iris image acquiring method of (two) automatically focusing
Based on the remote iris image acquisition device of above automatic focusing, Atomatic focusing method process flow diagram such as Fig. 3 that we adopt: method has two threads, and the first thread is used for focusing, and the second thread is used for image and selects journey.The focusing of the first thread is made of following step:
Step P1: initialization apparatus, the rotation focusing ring makes focusing ring 21 focusing positions to zero point, i.e. limit switch 11 places;
Step P2: iris camera 1 constantly gathers image, forms image sequence;
Step P3: calculation processing unit 7 calculates the sharpness of every two field picture in the image sequence that collects by iris camera 1 and the changing value of sharpness, respectively execution in step P4 and step P5;
Step P4: according to sharpness and the sharpness changing value of image, direction and the speed of the rotation of conversion focusing ring are returned to step P2; Described conversion is according to sharpness and sharpness changing value, and whose current state, is decided by defined state transition condition.Automatically the state of the remote iris image acquisition device of focusing is divided into four kinds, stationary state, extensive search state, search condition and focus condition among a small circle.
Step P5: change image over to and select thread, image with sharpness and sharpness changing value is selected, constantly judge whether to pick out the image that sharpness meets the demands from image sequence, if the clear picture of selecting, execution in step P6, if that selects is not fogging clear, get back to step P2;
Step P6: carry out human eye detection and extract eye image from the picture rich in detail of picking out;
Step P7: eye image is carried out the quality judgement, judge whether to be used for iris recognition, if the judgement eye image can be used in iris recognition, iris recognition finishes, if the judgement eye image can not be used for iris recognition, continue to gather image, continue execution in step P5.
(1) computed image sharpness
In the said method flow process, the method for computed image sharpness is as follows:
Step P31: the image fritter of randomly drawing 100 20*20 on the image that gathers;
Step P32: to the edge energy of each image fritter computed image, the edge calculation energy approach is: asks at image fritter X, and the quadratic sum of the Grad of Y both direction, and the edge energy addition of having a few of image fritter, obtain edge energy;
Step P33: the edge energy value addition of 100 image fritters, and normalize to a numerical value between 0～1, described numerical value is the changing value of image definition.
Fig. 5 is that sharpness is along with the schematic diagram of the variable in distance of focusing.Fixedly the time, the point of sharpness maximum is at the focus place when object, and along with the distance of leaving the focal position, sharpness descends gradually, and the sharpness peak value is very sharp-pointed, can be used for judge whether sequence has the focus existence.Wherein, X-axis is distance, and Y-axis is sharpness, and Q is change curve, and P is the focal position, and D is for can identify field depth.
Fig. 6 is object when changing (in the situation that the people moves), the variation schematic diagram of the focal position of the long focal length lens 2 of iris camera 1.When the people moved to B from position A, the definition values of iris camera 1 different focus distance also was changed to curve QB from curve QA, and the focal position becomes PB from PA.So this device can realize when the people moves, device can trace into new focal position PB from original focal position PA fast.
(2) the focusing strategy of state-based conversion
In the said method flow process, according to sharpness and the sharpness changing value of image, the current system state of the remote iris image acquisition device of the direction of conversion focusing ring 21 rotations, speed and automatic focusing is decided by defined system state switch condition.System state is defined as stationary state S1, extensive search state S2, search condition S3 and focus condition S4 among a small circle.Switch condition between them is: C11, C12, C23, C34, C44, C43, C32, C21.System state and state transition condition are as shown in the focusing strategy schematic diagram of Fig. 4, wherein:
Stationary state S1: focusing ring 21 is in leads a position, transfixion.If satisfy state transition condition C11 (being that sharpness is lower than a threshold value T1), thinking does not have object to enter the visual field of iris camera, does not rotate focusing ring 21, and this moment, focusing ring 21 still remained static; If satisfy state transition condition C12 (being that sharpness is higher than threshold value T1), thinking has object to come into view, and system enters the extensive search state.
Extensive search state S2 is: rotation focusing ring 21 from 0 in the hunting zone of maximum step number quick rotation: the certain step number of every rotation gathers a two field picture, and calculates the value of sharpness, and records current location.If satisfy state transition condition C23 (namely when definition values greater than a threshold value T2, or definition values is when obvious peak value occurring), think to enter and to search among a small circle S3 near the focal position; If satisfy state transition condition C21 (if namely not finding the focal position), get back to null position S1.
Search condition S3 is among a small circle: near the focal position, rotate so that small step is long near the scope of focusing ring 21 focus: often rotate to an angle, gather a two field picture, and calculate the value of sharpness, and record current step-length position.If satisfy state transition condition C34 (namely when definition values greater than a threshold value T3, or definition values is when obvious peak value occurring), think to arrive the focal position, record this peak value, enter focus mode; If satisfy state transition condition C32 (if namely not finding the focal position), proceed extensive search S2.
Focus mode S4: focusing ring 21 stops at the focal position, calculates at set intervals sharpness changing value successively.If satisfy state transition condition C43 (namely work as sharpness and alter a great deal, or sharpness is less than a half of initial focus peak value), think that variation has occured focus, the new focus of search S3 among a small circle; Otherwise, if satisfy state transition condition C44 (being that sharpness changes a half that is not less than initial focus peak value), suspend a period of time, continue to calculate the sharpness changing value, state is constant.
Described threshold value T1, T2 and T3 obtain by experiment.When experiment, with the some pairs of camera collection (more than 500 width) image; Manual image is divided into four classes: there is no object, object is arranged but image blurring, image is more clear, image is very clear; Then calculate the value of each width image definition; The definition values of four class images is respectively normal distribution, utilizes bayes method to determine cut off value T1, T2 and T3 between three classes to be each above-mentioned threshold value.
(3) image selection method
In above-mentioned method flow, select clearly that image is used for carrying out iris recognition in focusing, wherein the concrete grammar of step P5 is:
Step P51: every collection 10 two field pictures, choose 1 two field picture of sharpness maximum from 10 two field pictures;
Step P52: if the sharpness of image is not processed less than a threshold value T3;
Step P53: if the sharpness of image more than or equal to threshold value T3, continues to process to this image.
Wherein, the concrete grammar of step P6 eye detection is:
At first, utilize by machine learning algorithm (Adaboostting) training generation based on Ha Er small echo (Haar) feature cascade sorter.In system's operational process of the remote iris image acquisition device of automatically focusing, the down-sampled high-definition picture that collects, and utilize feature cascade sorter to carry out eye detection on different scale.If eyes do not detected, do not process; If eyes detected, go out the image of the eyes part under original resolution at the high-definition picture up-sampling that collects, proceed to process.
(4) iris image quality judgement
Above-mentioned image clearly not necessarily can carry out iris recognition, so the present invention need to carry out the picture quality judgement again to eye image, the method for step P7 is as described in Figure 7:
Step P71: the high band component E1 that calculates original eye image; High fdrequency component E1 obtains by the following method: generate the operator of a 8*8,4*4 the point value at operator center is-3, and other value is+1; Operator with this 8*8 carries out filtering on image; Then that filtered all picture point additions are obtained and be defined as E1;
Step P72: down-sampled eye image is that eye image is dwindled one times;
Step P73: calculate the high band component E2 of down-sampled eye image, obtain the computing method of E2 with step P71;
Step P74: ask for the ENERGY E 1 of high band and than the ratio R of the ENERGY E 2 of low-frequency range;
Step P75: in conjunction with high band ENERGY E 1, the definition values C that obtains iris image quality is: C=α E1+ β R, and α, β are weighted operators, are drawn by experience;
If the definition values C that obtains does not process iris image less than a threshold value T4; Otherwise iris image is carried out pre-service, Iris Location, feature extraction and aspect ratio pair.
Application Example: full automatic distant range iris recognition system
According to above-mentioned focusing mechanism and focusing method, form a cover iris capturing instance system, as shown in Figure 8, wherein, described full automatic distant range iris recognition system is comprised of following part: far distance automatic focusing mechanism D1, remote infrared light supply D2, driving governor D3, light source controller D4, image pick-up card D5, computing machine D6, panel D7 and rack D8, and its connected mode is: the stepper motor of far distance automatic focusing mechanism D1 is electrically connected on computing machine D6 by driving governor D3; Iris camera 1 on far distance automatic focusing mechanism D1 is connected on computing machine D6 by image pick-up card D5, is connected on computing machine D6 by the USB line apart from the wide visual angle camera 7 of automatic focusing mechanism D1; Remote infrared light supply D2 is connected on computing machine D6 by controller D4; Above equipment all is installed to D8 in rack, and far distance automatic focusing mechanism D1 and remote infrared light supply D2 install towards the front.For attractive in appearance, rack D8 front plate placement D7, panel D7 can be the acrylic board with the cavity, also can be the glass of transmitted infrared light, reflect visible light.
Be the example that full automatic distant range iris recognition system shown in Figure 8 is shown as Fig. 9, this recognition system operation has following steps:
Step F 1: make the face area of iris camera 1 aligning people in far distance automatic focusing mechanism D1, and start iris camera 1 collection high-definition picture;
Step F 2: calculated the sharpness of the realtime graphic that collects by camera 1 by computing machine, and focusing mechanism D1 far distance automatic according to the variation automatic rotating of sharpness, the focal position of search and tracking object;
Step F 3: iris camera 1 is selected the most clearly image in image sequence, detects eye image, extracts the image of people's eye portion;
Step F 4: the judgement of eye image quality, select image the most clearly in the eye image sequence, be used for iris recognition.
In this example, because full automatic distant range iris recognition system of the present invention has adopted apart from automatic focusing mechanism, effective identification range of described full automatic distant range iris recognition system improves greatly.The people goes to the scope of 2.5～3.2 meters of rack precontracts, freezes, and this recognition system can be adjusted camera D1 position according to people's height automatically, carries out simultaneously the iris that the people is taken in focusing automatically.
The user can feel comfortable nature when using this full automatic distant range iris recognition system of the present invention, and do not worry the position not to and can not complete identification.On the other hand, for the user of the full automatic distant range iris recognition system of first use, can not complete because it there is no experience yet.
The above; only be the embodiment in the present invention; but protection scope of the present invention is not limited to this; anyly be familiar with the people of this technology in the disclosed technical scope of the present invention; can understand conversion or the replacement expected; all should be encompassed in of the present invention comprise scope within, therefore, protection scope of the present invention should be as the criterion with the protection domain of claims.
Claims (7)
1. a remote iris image acquisition device of automatically focusing, is characterized in that, comprising: iris camera, long focal length lens, focusing gear unit, lens converter, focusing electric rotating machine, electric-motor drive unit, calculation processing unit, wherein:
Iris camera, long focal length lens and lens converter are positioned on same axis, the focusing gear unit with focus that electric rotating machine is coaxial to be connected; The iris camera connects by lens converter with long focal length lens; The focusing ring that the focusing gear unit is connected with long focal length lens connects; Focusing electric rotating machine and electric-motor drive unit electrical connection; Calculation processing unit is connected with the electric-motor drive unit signal, controls the rotation of focusing electric rotating machine by electric-motor drive unit; Calculation processing unit is connected with the iris camera, processing is from the image of iris camera, described focusing gear unit contains cavity ring, jig, belt and limit switch, jig is being installed on cavity ring and between focusing ring, jig has the elasticity three-jaw, is used for blocking camera lens and the focusing ring of different sizes in described cavity ring; The be locked rotating shaft of cavity ring and the focusing electric rotating machine that is locked of described belt be used for to drive the cavity ring rotation, thereby drive focusing ring, camera lens is focused; Be connected with limit switch when focusing ring rotates to a position, limit switch sends a signal stops operating the focusing electric rotating machine, makes the focusing gear unit have an initial position.
2. the remote iris image acquisition device of automatically focusing as claimed in claim 1 is characterized in that the focal length value of long focal length lens is that 100mm is to the arbitrary value between 500mm.
3. a distant range iris image acquiring method of automatically focusing, is characterized in that, the step of obtaining the distant range iris image with the remote iris image acquisition device of automatically focusing is as follows:
Step P1: initialization apparatus, the rotation focusing ring makes the focusing ring focusing position to zero point, i.e. the limit switch place;
Step P2: the iris camera constantly gathers image, forms image sequence;
Step P3: the sharpness of every two field picture in the calculation processing unit sequence of computed images, then calculate the changing value of sharpness, execution in step P4 and step P5 respectively in two threads;
Step P4: according to sharpness and the sharpness changing value of image, direction and the speed of the rotation of conversion focusing ring are returned to step P2; Described conversion is according to sharpness and sharpness changing value, and the automatic current system state of the remote iris image acquisition device of focusing, is decided by the system state switch condition of the remote iris image acquisition device of automatic focusing;
Step P5: change image over to and select thread, the sharpness of image in sequence of computed images, calculate again the sharpness changing value, and judge whether to pick out by sharpness and the sharpness changing value of image the image that sharpness meets the demands, if the clear picture of selecting, execution in step P6, if select not fogging clear, get back to step P2;
Step P6: carry out human eye detection and extract eye image from the picture rich in detail of picking out;
Step P7: eye image is carried out the quality judgement, judge whether to be used for iris recognition, if the judgement eye image can be used in iris recognition, iris recognition finishes, if the judgement eye image can not be used for iris recognition, continue to gather image, continue execution in step P5;
The step that described image definition is calculated is as follows:
Step P31: the image fritter of randomly drawing 100 20*20 on the image that gathers;
Step P32: to the edge energy of each image fritter computed image, the edge calculation energy approach is: asks at image fritter X, and the quadratic sum of the Grad of Y both direction, and the edge energy addition of having a few of image fritter, obtain edge energy;
Step P33: the edge energy value addition of 100 image fritters, and normalize to a numerical value between 0～1, described numerical value is the changing value of image definition;
The described step that the eye image quality is judged is as follows:
Step P71: the high band component E1 that calculates original eye image; High fdrequency component E1 obtains by the following method: generate the operator of a 8*8,4*4 the point value at operator center is-3, and other value is+1; Operator with this 8*8 carries out filtering on image; Then that filtered all picture point additions are obtained and be defined as E1;
Step P72: down-sampled eye image is that eye image is dwindled one times;
Step P73: the high band component E2 that calculates down-sampled image; Obtain the method for high band component E2 with step P71;
Step P74: ask for the ENERGY E 1 of high band and than the ratio R of the ENERGY E 2 of low-frequency range,
Step P75: in conjunction with high band ENERGY E 1, the definition values C that obtains iris image quality is: C=α E1+ β R, and α, β are that weighted operator is drawn by experience;
If the definition values C that obtains does not process iris image less than a threshold value T4; Otherwise iris image is carried out pre-service, Iris Location, feature extraction and aspect ratio pair.
4. the distant range iris image acquiring method of automatically focusing as claimed in claim 3 is characterized in that, the switch condition of the system state of the remote iris image acquisition device of described automatic focusing and focusing strategy are as described below:
Sharpness and sharpness changing value according to image, the current system state of the remote iris image acquisition device of the direction of conversion focusing ring rotation and speed and automatic focusing, decided by the system state switch condition, system state is defined by system software, the system state of remote iris image acquisition device of focusing automatically is divided into stationary state, extensive search state, search condition and focus condition among a small circle, the switch condition between them is:
C11，C12，C23，C34，C44，C43，C32，C21；
The switch condition of described system state and focusing strategy are as described below:
Stationary state: focusing ring is in null position, and transfixion is if satisfy state transition condition C11, described state transition condition C11 is that sharpness is lower than a threshold value T1, thinking does not have object to enter the visual field of iris camera, does not rotate focusing ring, and this moment, focusing ring still remained static; If satisfy state transition condition C12, described state transition condition C12 be sharpness higher than threshold value T1, thinking has object to come into view, system enters the extensive search state;
The extensive search state is: the rotation focusing ring from 0 in the hunting zone of maximum step number quick rotation: the certain step number of every rotation gathers a two field picture, and calculates the value of sharpness, and records current location; If satisfy state transition condition C23, described state transition condition C23 be when definition values greater than a threshold value T2, or definition values is when obvious peak value occurring, thinks to enter search among a small circle near the focal position; If satisfy state transition condition C21 for not finding the focal position, get back to null position;
Search condition is among a small circle: near the focal position, rotate so that small step is long near the scope of focusing ring focus: often rotate to an angle, gather a two field picture, and calculate the value of sharpness, and record current step-length position; If satisfy state transition condition C34, greater than a threshold value T3, or definition values thinks to arrive the focal position when obvious peak value occurring, records this peak value, enters focus mode when described state transition condition C34 definition values; If satisfy state transition condition C32 for not finding the focal position, proceed extensive search;
Focus mode: focusing ring stops at the focal position, calculate at set intervals sharpness changing value successively, if satisfy state transition condition C43, described state transition condition C43 is when the half of sharpness less than initial focus peak value, think that variation has occured focus, the new focus of search among a small circle; Otherwise if satisfy state transition condition C44, described state transition condition C44 is that sharpness changes a half that is not less than initial focus peak value, suspends a period of time, continues to calculate the sharpness changing value, and state is constant.
5. the distant range iris image acquiring method of automatically focusing as claimed in claim 3 is characterized in that, the concrete steps that described image is selected are as follows:
Step P51: every collection 10 two field pictures, choose 1 two field picture of sharpness maximum from 10 two field pictures;
Step P52: if the sharpness of image is not processed less than a threshold value T3;
Step P53: if the sharpness of image more than or equal to threshold value T3, continues to process to this image.
6. the distant range iris image acquiring method of automatically focusing as claimed in claim 5 is characterized in that, the step of described extraction eye image is as follows:
Utilization is generated based on Ha Er wavelet character cascade sorter by the machine learning algorithm training; Down-sampled high-resolution image, and utilize above-mentioned sorter to carry out eye detection; If eyes do not detected, do not process; If eyes detected, go out the image of the eyes part under original resolution at the high-definition picture up-sampling that collects, proceed to process.
7. a right to use requires the full automatic distant range iris recognition system of the distant range iris image acquiring method of 3 or 4 described automatic focusings, comprise far distance automatic focusing mechanism, remote infrared light supply, driving governor, light source controller, image pick-up card, computing machine, panel and rack, wherein:
The focusing electric rotating machine of far distance automatic focusing mechanism is electrically connected on computing machine by driving governor; Iris camera on far distance automatic focusing mechanism is connected on computing machine by image pick-up card; The high-definition picture of iris camera collection passes on computing machine by image pick-up card; By computing machine calculate camera collection to the sharpness of realtime graphic, and according to switch condition and the focusing strategy of the system state of the remote iris image acquisition device of automatic focusing, the far distance automatic focusing mechanism of automatic rotating, the focal position of realizing search and tracking object; Remote infrared light supply is connected on computing machine by light source controller; Above equipment all is installed in rack, and far distance automatic focusing mechanism and remote infrared light supply are all installed towards the front, rack front plate placement.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 200910077372 CN101814129B (en) | 2009-02-19 | 2009-02-19 | Automatically focused remote iris image acquisition device, method and recognition system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 200910077372 CN101814129B (en) | 2009-02-19 | 2009-02-19 | Automatically focused remote iris image acquisition device, method and recognition system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN101814129A CN101814129A (en) | 2010-08-25 |
| CN101814129B true CN101814129B (en) | 2013-05-08 |
Family
ID=42621381
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN 200910077372 Active CN101814129B (en) | 2009-02-19 | 2009-02-19 | Automatically focused remote iris image acquisition device, method and recognition system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101814129B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2022066816A3 (en) * | 2020-09-25 | 2022-04-28 | Sterling Labs Llc | Pose optimization in biometric authentication systems |
Families Citing this family (26)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103096118B (en) * | 2011-10-28 | 2015-10-07 | 浙江大华技术股份有限公司 | Camera zoom detection method and device |
| CN102707425B (en) | 2012-06-21 | 2014-04-16 | 爱威科技股份有限公司 | Image processing method and device |
| CN105025218B (en) * | 2014-04-30 | 2018-08-28 | 瑞昱半导体股份有限公司 | Passive type automatic focusing mechanism and method |
| CN104268517B (en) * | 2014-09-19 | 2018-11-09 | 武汉虹识技术有限公司 | A kind of Atomatic focusing method and system applied to iris authentication system |
| CN104284095B (en) * | 2014-10-28 | 2017-07-14 | 福建福光股份有限公司 | The fast automatic focusing method and system of a kind of long-focus visible ray industrial lens |
| CN104639894A (en) * | 2014-12-11 | 2015-05-20 | 北京中星微电子有限公司 | Image focusing method and device for surveillance camera |
| CN105260725A (en) * | 2015-10-23 | 2016-01-20 | 北京无线电计量测试研究所 | Iris recognition system |
| CN105915870A (en) * | 2015-12-30 | 2016-08-31 | 乐视致新电子科技（天津）有限公司 | Automatic focusing apparatus and method of projector |
| CN105652429B (en) * | 2016-03-22 | 2018-07-17 | 哈尔滨理工大学 | A kind of microscopic cell glass slide scanning auto focusing method based on machine learning |
| CN106231201B (en) * | 2016-08-31 | 2020-06-05 | 成都极米科技股份有限公司 | Automatic focusing method and device |
| CN106713706B (en) * | 2016-11-14 | 2019-07-02 | 浙江吉利控股集团有限公司 | A kind of camera shooting acquisition device for vehicle test |
| CN107219207B (en) * | 2017-07-04 | 2023-10-20 | 福州大学 | Automatic focusing method of CCD biochip fluorescence scanner |
| CN107395993B (en) * | 2017-09-08 | 2023-06-30 | 北京睿智奥恒视觉科技有限公司 | Full-automatic focusing method and system |
| CN108158555A (en) * | 2017-12-31 | 2018-06-15 | 执鼎医疗科技（杭州）有限公司 | A kind of system and method for OCT equipment being accelerated to be aligned with human eye by iris recognition |
| CN109254382B (en) * | 2018-10-12 | 2021-11-26 | 南通大学 | Thread detection automatic focusing method based on machine vision |
| CN109101959A (en) * | 2018-11-02 | 2018-12-28 | 张彦龙 | A kind of iris image extraction element based on liquid lens and VCM |
| CN109376729B (en) * | 2018-12-28 | 2022-02-22 | 武汉虹识技术有限公司 | Iris image acquisition method and device |
| CN111263059B (en) * | 2020-01-19 | 2021-02-26 | 上海海鸥数码照相机有限公司 | Manual focusing shooting method and device for lens-exchangeable paraxial camera |
| CN113536825A (en) * | 2020-04-13 | 2021-10-22 | 上海聚虹光电科技有限公司 | Focusing algorithm for iris recognition in motion |
| CN113709353B (en) * | 2020-05-20 | 2023-03-24 | 杭州海康威视数字技术股份有限公司 | Image acquisition method and device |
| CN112000226B (en) * | 2020-08-26 | 2023-02-03 | 杭州海康威视数字技术股份有限公司 | Human eye sight estimation method, device and sight estimation system |
| CN112333383A (en) * | 2020-10-27 | 2021-02-05 | 浙江华创视讯科技有限公司 | Automatic focusing method and device |
| CN114765661B (en) * | 2020-12-30 | 2022-12-27 | 杭州海康威视数字技术股份有限公司 | Iris identification method, device and equipment |
| CN113662505B (en) * | 2021-10-22 | 2022-02-08 | 北京茗视光眼科医院管理有限公司 | System for measuring myopia diopter |
| CN114047655A (en) * | 2021-11-11 | 2022-02-15 | 吉林凯高医疗科技有限公司 | Automatic focusing device |
| CN114422660A (en) * | 2021-12-06 | 2022-04-29 | 江苏航天大为科技股份有限公司 | Imaging focusing system of designated monitoring area |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101116609A (en) * | 2007-08-30 | 2008-02-06 | 中国科学技术大学 | Scanning type automatic zooming iris image gathering system and gathering method thereof |
-
2009
- 2009-02-19 CN CN 200910077372 patent/CN101814129B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101116609A (en) * | 2007-08-30 | 2008-02-06 | 中国科学技术大学 | Scanning type automatic zooming iris image gathering system and gathering method thereof |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2022066816A3 (en) * | 2020-09-25 | 2022-04-28 | Sterling Labs Llc | Pose optimization in biometric authentication systems |
Also Published As
| Publication number | Publication date |
|---|---|
| CN101814129A (en) | 2010-08-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101814129B (en) | Automatically focused remote iris image acquisition device, method and recognition system | |
| CN101690165B (en) | Control method based on a voluntary ocular signal, particularly for filming | |
| CN105554385B (en) | A kind of remote multi-modal biological characteristic recognition methods and its system | |
| Coombs et al. | Real-time binocular smooth pursuit | |
| CN105431078B (en) | System and method for the tracking of coaxial eye gaze | |
| CN100373394C (en) | Petoscope based on bionic oculus and method thereof | |
| CN100343867C (en) | Method and apparatus for distinguishing direction of visual lines | |
| CN102045577B (en) | Observer tracking system and three-dimensional stereo display system for three-dimensional stereo display | |
| CN102855471B (en) | Remote iris intelligent imaging device and method | |
| CN106127145B (en) | Pupil diameter and tracking | |
| CN101807110B (en) | Pupil positioning method and system | |
| CN101587542A (en) | Field depth blending strengthening display method and system based on eye movement tracking | |
| CN102855476A (en) | Self-adaptive binocular iris synchronous collection system of single image sensor | |
| CN103067662A (en) | Self-adapting sightline tracking system | |
| CN203027358U (en) | Adaptive sight line tracking system | |
| CN108198199A (en) | Moving body track method, moving body track device and electronic equipment | |
| CN101930606A (en) | Field depth extending method for image edge detection | |
| CN105785561B (en) | A kind of digital microscope and its focus method | |
| GB2495323A (en) | Method of capturing an iris image free from specularities caused by spectacles | |
| CN202196428U (en) | Single image sensor self-adaptive binocular iris synchronous acquisition system | |
| Kyrkou | C 3 Net: end-to-end deep learning for efficient real-time visual active camera control | |
| CN206559469U (en) | A kind of target following equipment based on FPGA | |
| CN104410775A (en) | High-resolution microscopic visual imaging device and control method | |
| Rougeaux et al. | Robust tracking by a humanoid vision system | |
| CN109212547A (en) | A kind of ranging experimental provision and the distance measuring method based on moving-target edge sensitive |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20181113 Address after: Unit 1005, Block MSD-G1, Teda, 57 Second Avenue, Tianjin Economic and Technological Development Zone, 300457 Patentee after: Tianjin Zhongke Rainbow Star Technology Co., Ltd. Address before: 100080 No. 95 East Zhongguancun Road, Beijing, Haidian District Patentee before: Institute of Automation, Chinese Academy of Sciences |