CN109700469B

CN109700469B - Iris anti-counterfeiting artifact living body detection method based on RGB-IR imaging

Info

Publication number: CN109700469B
Application number: CN201811304742.8A
Authority: CN
Inventors: 陈平; 倪蔚民
Original assignee: Suzhou Siyuan Kean Information Technology Co ltd
Current assignee: Suzhou Siyuan Kean Information Technology Co ltd
Priority date: 2014-10-15
Filing date: 2015-10-14
Publication date: 2021-08-06
Anticipated expiration: 2035-10-14
Also published as: CN105243368A; CN104252622A; CN105574483B; CN105574483A; CN105395203B; WO2016058528A1; CN109700470A; CN105243368B; CN109583316B; WO2016058527A1; CN109583316A; CN109700469A; CN105395203A

Abstract

The invention discloses an iris anti-counterfeit living body detection system based on RGB-IR imaging, wherein the system adopts an eyeball physiological motion activity characteristic real-time detection method to realize anti-counterfeit living body detection.

Description

Iris anti-counterfeiting artifact living body detection method based on RGB-IR imaging

The application is as follows: 10/14/2015, application numbers are: 2015106613691 patent application.

Technical Field

The invention relates to the field of biological identification photoelectricity, in particular to a front-mounted and iris identification integrated photoelectric imaging system and method for a high-safety mobile terminal.

Background

The mobile terminal comprises a smart phone, a tablet, wearable equipment and the like, and the mobile terminal equipment is necessarily the most widely applicable equipment in the future in the current information technology mobile development trend.

At present, the applications of mobile terminals in practical applications in mobile secure payment, account secure login and online banking are extremely wide, for example, applications in aspects of balance treasures, WeChat, bank account management and the like, although great convenience is brought to life in the using process of the mobile terminals, a novel economic crime performed through the characteristics of weak security performance of the mobile terminals and the like gradually rises.

In the mobile terminal, the conventional means for identity confirmation in the prior art is password input, but the security performance of the means for identity confirmation is very low, and the password can be revealed only by implanting a simple virus program into the mobile terminal, so that corresponding loss is caused. In order to solve the problem, the mobile terminal security identity authentication is carried out internationally in a biological identification mode; for example, the fingerprint identification technology developed by the AuthenTec company, which is proposed by the apple company, is applied to a mobile phone terminal, so that the identity confirmation security of the mobile terminal is greatly improved; however, in the process of fingerprint identification, because fingerprints are static and unique, but fingerprint information is very easy to acquire, even copied, and the like, as the application of fingerprint technology on mobile terminals is more and more extensive, the security of the fingerprint technology is in a corresponding trend, so that iris identification which is more advantageous in security is a very effective method for solving the security authentication process of the mobile terminals, and an iris identification system is the highest accuracy in the existing biometric identification.

At present, in all iris recognition system technologies and products in mobile terminals, integration of a front photoelectric imaging system and an iris recognition photoelectric imaging system for a face self-photographing function is not realized. However, if the front photoelectric imaging system with the face self-photographing function and the iris recognition photoelectric imaging system are integrated and separately realized, the cost is greatly increased, and the more important volume of the mobile terminal cannot provide an installation space for accommodating 2 sets of separate and independent optical imaging systems.

In addition, although iris recognition is more advantageous than fingerprint face recognition in terms of security of anti-counterfeit objects, if the iris recognition is applied to important occasions such as mobile phone mobile payment with large amount, the security technology of anti-counterfeit object living body detection still needs to be further upgraded, and the threat of potential safety hazards is eliminated. After all, biometric identification is intended to be security by itself, and security by itself is the most basic and important.

Furthermore, the high-security mobile terminal front-end and iris recognition integrated photoelectric imaging system needs to solve the following serious problems:

1. the front-mounted and iris recognition integrated photoelectric imaging system in the application of the mobile terminal meets the requirements of integration of the front-mounted photoelectric imaging system with the face self-photographing function and the iris recognition photoelectric imaging system, and the volume of the front-mounted photoelectric imaging system is controlled within 8.5mm by 6 mm.

2. A preposed and iris recognition integrated photoelectric imaging system in the application of a mobile terminal needs a whole set of high-safety anti-counterfeit living body detection method to ensure the safety of biological recognition.

3. In the application of a mobile terminal, a preposed and iris recognition integrated photoelectric imaging system needs to guide the theoretical derivation of the conversion relation designed by the photoelectric imaging system.

4. In the application of the mobile terminal, the front-mounted and iris recognition integrated photoelectric imaging system needs to greatly reduce the cost to be within 10 dollars before being applied in a large scale.

Solving the above problems is the biggest challenge facing today.

Disclosure of Invention

The invention provides a front-mounted and iris recognition integrated photoelectric imaging system for a high-safety mobile terminal.

In order to solve the technical problem, the invention provides an iris anti-counterfeiting object in-vivo detection method based on RGB-IR imaging, which is characterized by being realized by one or more of the following modes: real-time detection of optical activity characteristics of biological tissues generated by RGB-IR imaging wavelength radiation; a real-time detection method for pupil iris diameter change rate biological tissue activity characteristics generated by RGB-IR imaging wavelength radiation; a real-time detection method of the optical reflection position of the cornea generated by RGB-IR imaging wavelength radiation; and iv, detecting the activity characteristic of the physiological movement of the eyeball in real time based on RGB-IR imaging.

Summarizing the above description, the present invention realizes a high-security mobile terminal front and iris recognition integrated photoelectric imaging system and method thereof:

1. the front-mounted and iris recognition integrated photoelectric imaging system realizes integration of the front-mounted photoelectric imaging system and the iris recognition photoelectric imaging system which meet the function of face self-photographing, and the volume of the front-mounted and iris recognition integrated photoelectric imaging system is controlled within 8.5mm by 6 mm.

2. The preposed and iris recognition integrated photoelectric imaging system realizes a whole set of high-safety anti-counterfeit living body detection method and ensures the safety of biological recognition.

3. The preposed and iris recognition integrated photoelectric imaging system realizes theoretical derivation of a conversion relation for guiding the design of the photoelectric imaging system.

4. The front-mounted and iris recognition integrated photoelectric imaging system can greatly reduce the cost, and the cost can be reduced to more than 10 dollars and can be applied in a large scale.

Drawings

The following describes embodiments of the present invention in further detail with reference to the accompanying drawings.

FIG. 1 is a general block diagram of a front-end and iris recognition integrated optoelectronic imaging system of the present invention;

FIG. 2 is a schematic diagram of each imaging pixel cell of the imaging array of image sensor 105 of FIG. 1 independently receiving RGB-IR wavelength channels.

Fig. 3 is a schematic diagram of a reset integration and readout circuit for reset integration and readout of charge (electron) voltages for image sensor 105 of fig. 2.

Fig. 4 is a schematic diagram of a 2 x 2 cross-pixel cell 4 arrangement format of an imaging array of the RGB-IR wavelength channels of the image sensor 105 of fig. 2;

fig. 5 is a diagram illustrating interpolation of RAW data for 4 adjacent pixels in the imaging array of image sensor 105 of fig. 2 between pixels of the same wavelength channel.

FIG. 6 is a diagram illustrating a contrast region for defining an iris image according to the present invention.

Fig. 7 is a diagram illustrating the pupil and iris diameter of an iris image defined according to the present invention.

FIG. 8 is a schematic diagram of the optical reflection points of the present invention defining different positions of the cornea of an iris image.

Fig. 9 is a schematic diagram illustrating the degree of the eyelid physiological motion activity characteristic generated by the eyeball physiological motion according to the present invention.

FIG. 10 is a schematic diagram illustrating the degree of the activity of the off-axis strabismus physiological movement generated by the physiological movement of the eyeball according to the present invention.

Detailed Description

The embodiment 1 provides a front-mounted mobile terminal and face/iris recognition integrated photoelectric imaging system and method. The method comprises a preposed photoelectric imaging method, an iris recognition photoelectric imaging method, a method for carrying out interpolation reconstruction between original RAW data pixels with the same wavelength channel used in the preposed photoelectric imaging method or the iris recognition photoelectric imaging method, and an iris anti-counterfeiting living body detection method.

As shown in FIG. 1, the system is provided with an optical filter (101 or 104) (for filtering imaging wavelength), an optical imaging lens (102) (for physically refracting and focusing imaging wavelength), a fixed mounting seat (103) (for fixedly mounting the optical imaging lens), an image sensor (105) (for photoelectrically converting output imaging image), an illuminating light source (106) including an RGB-LED illuminating light source (106RGB) and an IR-LED illuminating light source (106IR), the RGB-LED illuminating light source (106RGB) is used for generating RGB imaging wavelength radiation for the front photoelectric imaging system, the IR-LED illuminating light source (106IR) is used for generating IR imaging wavelength radiation for the iris recognition photoelectric imaging system), and an imaging system fixed mounting substrate (107) (for providing a front and iris recognition photoelectric imaging system fixed mounting carrier), a mobile terminal main board (110) (for implementing mobile terminal function circuit carrier) is further arranged on the imaging system fixed mounting substrate (107), an LED current driver 108 (for driving and controlling the LED illumination source radiation intensity, radiation angle position, and radiation time) and a processor chip 109 (for driving and controlling the LED current driver and the image sensor) are integrated on the mobile terminal main board 110.

In embodiment 1 of the present invention, the front-end and iris recognition integrated photoelectric imaging system includes an optical path for the front-end photoelectric imaging system and an optical path for the iris recognition photoelectric imaging system; the optical path of the front photoelectric imaging system comprises the following steps:

the RGB-LED illumination source 106RGB radiates RGB imaging wavelengths, the optical filter (101 or 104) filters the RGB imaging wavelengths, the optical imaging lens 102 physically refractively focuses the RGB imaging wavelengths, and the imaging array of the image sensor 105 independently receives the RGB wavelength channels.

The optical path of the iris recognition photoelectric imaging system comprises the following steps:

the IR-LED illumination source 106IR radiates IR imaging wavelengths, the optical filter (101 or 104) filters the IR imaging wavelengths, the optical imaging lens 102 physically refractively focuses the IR imaging wavelengths, and the imaging array of the image sensor 105 independently receives the IR wavelength channels.

In embodiment 1 of the present invention, the imaging array of the image sensor 105 is configured as an RGB-IR wavelength channel with independent reception function; the LED illumination sources (LED illumination source 106RGB and LED illumination source 106IR-LED) are configured to have radiation wavelength ranges that match the RGB-IR imaging wavelength channels of the image sensor 105; the optical filter (101 or 104) is configured to have a filtering wavelength range that is mutually matched with the image sensor 105RGB-IR imaging wavelength channel; the optical imaging lens 102 is configured to have a focusing wavelength range that is mutually matched with the RGB-IR imaging wavelength channel of the image sensor 105; the processor chip 109 is configured for driving the image sensor 105 settings, i.e. controlling the image pixel value data output by the RGB-IR wavelength channel imaging array of the image sensor 105, and driving the control LED current driver 108; the LED current driver 108 is configured to drive and control the LED illumination sources (106RGB and 106IR-LED) radiation intensity, radiation angular position, radiation time.

The optical imaging lens 102 is configured as a fixed focal length lens, and may be any one of a liquid-driven lens, a liquid crystal-driven lens, a VCM voice coil-driven lens, a MEMS-driven lens, an EDOF wavefront phase modulation lens, or a wafer-level array microlens, for example.

The imaging wavelength of the invention comprises an RGB imaging wavelength of 400-700nm and an IR imaging wavelength of 800-900 nm; the imaging wavelengths in this embodiment include an RGB imaging wavelength of 400-650nm and an IR imaging wavelength of 750-850 nm. Example 1 of the present invention by way of example, the IR imaging wavelength range, essentially the imaging wavelength range is a bandwidth characteristic, which may also be equivalently understood as being described by the imaging wavelength center (wavelengthcenter) and the full width at half maximum (FWHM), as expressed by the 800-900nm range, the center wavelength 850nm ± 30nm full width at half maximum. Further, as an example of the variation of the imaging wavelength range, a narrow band may be used as a half-peak bandwidth of 850nm ± 15nm of the center wavelength.

The front photoelectric imaging system adopts RGB imaging wavelength, and the focusing working object distance WD is at least 30-100 cm; the iris recognition photoelectric imaging system adopts IR imaging wavelength, and the focusing working object distance WD is at least 10-30 cm.

The iris recognition photoelectric imaging system has the following optical imaging requirements:

the imaging wavelength WI of the iris recognition photoelectric imaging system meets the following requirements: WI is more than or equal to 800nm and less than or equal to 900nm or more than or equal to 750nm and less than or equal to 850 nm:

the focusing working object distance WD of the iris recognition photoelectric imaging system satisfies the following conditions: WD is more than or equal to 10cm and less than or equal to 30 cm;

the pixel spatial resolution PSR (pixel spatial resolution) of the iris recognition photoelectric imaging system should satisfy: PSR is more than or equal to 13 pixel/mm;

the optical magnification OM (optical magnification) of the iris recognition photoelectric imaging system should satisfy: OM ═ PS ═ PSR;

wherein PS described above is the physical dimension of each imaging pixel unit of the image sensor 105; PSR is the pixel space resolution of the iris recognition photoelectric imaging system;

the optical spatial resolution of image of plane of the iris identification photoelectric imaging system should satisfy: at a modulation transfer function of 60% (MTF 0.6), 1/(4 PS) ≦ OSRI ≦ 1/(2 PS) lp/mm (line pair per mm).

The front-mounted photoelectric imaging system has the following optical imaging requirements:

the imaging wavelength WI of the front photoelectric imaging system meets the following conditions: WI is more than or equal to 400nm and less than or equal to 700nm or WI is more than or equal to 400nm and less than or equal to 650 nm;

the focusing working object distance WD of the front photoelectric imaging system satisfies the following conditions: WD is more than or equal to 30cm and less than or equal to 100 cm;

the pixel spatial resolution psr (pixel spatial resolution) of the front-facing electro-optical imaging system should satisfy: PSR is less than or equal to 4 pixel/mm;

the optical magnification om (optical magnification) of the front photoelectric imaging system should satisfy: OM ═ PS ═ PSR;

wherein PS described above is the physical dimension of each imaging pixel unit of the image sensor 105; PSR is the pixel spatial resolution of the front photoelectric imaging system;

the optical spatial resolution of image of plane of the front photoelectric imaging system should satisfy: at a modulation transfer function of 60% (MTF 0.6), 1/(4 PS) ≦ OSRI ≦ 1/(2 PS) lp/mm (line pair per mm).

In this embodiment, the imaging array of image sensor 105 independently receives each imaging pixel cell structure of the RGB-IR wavelength channels as shown in FIG. 2.

The imaging array of image sensor 105 independently receives each imaging pixel cell of the RGB-IR wavelength channels, including the following: a micro lens 201(micro lens) for converging the photon 200; a separate RGB-IR wavelength channel filter layer 202(RGB-IR filter) for filtering photons 200; a semiconductor photodiode 203(photo diode) for capturing photons 200 of an incident wavelength to perform photoelectric quantum conversion; a reset integration and readout circuit 204 for resetting the integration and readout charge (electron) voltage; an analog-to-digital converter ADC205 for converting the voltage value into a quantized value. A micro lens 201(micro lenses), an independent RGB-IR wavelength channel filter layer 202(RGB-IR filter), a semiconductor photodiode 203(photo diode), a reset integrating and reading circuit 204, and an analog-digital converter ADC205 are sequentially disposed from top to bottom; the incident photons 200 pass through the microlens 201, the independent RGB-IR wavelength channel filter layer 202, and the semiconductor photodiode 203 in that order.

The micro lens 201(micro lens) has the photon converging efficiency or fill factor (fill factor) FF more than or equal to 95 percent; an RGB-IR wavelength channel filter layer 202(RGB-IR filter) for filtering to generate individual RGB-IR wavelength channels; in specific embodiment 1 of the present invention, the B wavelength channel: 400nm-500 nm; g wavelength channel: 500nm-600 nm; r wavelength channel: 600nm-700 nm; IR wavelength channel: 800nm-900 nm; or even further, the B wavelength channel: 400nm-500 nm; g wavelength channel: 500nm-590 nm; r wavelength channel: 590nm-670 nm; IR wavelength channel: 750nm-850 nm. Filter layer 202 has RGB-IR channel wavelength distribution functions FR (λ), FG (λ), FB (λ), FIR (λ); the semiconductor photodiode 203 has the capability of generating a photoelectric quantum conversion by forming electron-hole pairs at the semiconductor PN junction upon receiving photons 200 of an incident wavelength.

The semiconductor photodiode 203 receives photons 200 of incident wavelengths for photoelectric quantum conversion, and the photoelectric quantum conversion constants QR, QG, QB, and QIR of the incident wavelengths of RGB-IR are defined as follows:

(EQ1)

λ is the imaging wavelength, and the preferred RGB imaging wavelength in embodiment 1 of the present invention is 700nm and the IR imaging wavelength is 800-900nm, and as an equivalent understanding, the RGB imaging wavelength may be further selected to be 650nm and the IR imaging wavelength may be further selected to be 850nm and 750-850 nm.

g (λ), r (λ), b (λ), IR (λ) are the photoelectric quantum conversion efficiency sensitivity functions of the RGB-IR wavelength channels of the photodiodes 203 of the image sensor 105, FR (λ), FG (λ), FB (λ), FIR (λ) are the RGB-IR channel wavelength distribution functions of the filter layer 202 of the image sensor 105, f (λ) is the filtering rate wavelength distribution function of the optical filter (101 or 104), S (λ) is the radiance wavelength distribution function of the LED illumination sources (106RGB and 106 IR-LED); l (λ) is a transmittance wavelength distribution function of the optical imaging lens 102.

According to IThe SO measurement unit definition standard, at the imaging wavelength of 400-700nm, the photoelectric quantum conversion constant units of QR, QG, QB are V/lux-sec (volt per lux per second) or ke^-Lux-sec. The specific example 1 of the present invention has a value of, for example, 2.0V/lux-sec; the unit of the photoelectric quantum conversion constant of QIR is V/(mw/cm) at the imaging wavelength of 800-²Sec) (volts per milliwatt per square centimeter per second) or ke^-/(mw/cm²-sec); the specific example 1 of the invention has the structure as 8000V/(mw/cm)²-sec)。

A reset integrating and reading out circuit 204 for resetting the integrating and reading out charge (electron) voltages, charge (electron) voltages V for resetting the integrating photodiode 203, and charge (electron) voltages V for reading out the photodiode 203, respectively (formulas for resetting the charge (electron) voltages V for integrating the photodiode 203, and charge (electron) voltages V for reading out the photodiode 203, respectively, are as follows);

charge (electronic) voltage V ═ Q/C (EQ2)

Wherein: q is the Charge (electrons) of the reset integration of the photodiode 203, C is the equivalent capacitance of the photodiode 203, and further, the photodiode 203 has a full Charge (electron) capacity FCC (full Charge Capacity) of 10ke or more^-(thousand electrons) (Kelectrons); the voltage reset integration and readout circuit 204 has a charge (electron) -voltage conversion gain cg (conversion gain): CG 1/C V/Q unit: μ V/e^-Microvolts per charge (electron); the voltage reset integration and readout circuit 204 has a Global frame mode reset integration and readout (Global Shutter) or a Rolling Shutter mode reset integration and readout (Rolling Shutter).

Fig. 3 is a schematic diagram of a reset integrating and reading circuit for resetting integration and reading out charge (electron) voltages of the imaging pixel unit of the image sensor 105 in embodiment 1 of the present invention (203 is a photodiode, 205 is an analog-to-digital converter ADC, M1, M2, M3 are transistors, Vdd is a power supply, GND is ground, reset is a reset integrating control signal for resetting the integrated charge (electron) voltages, read is a reading control signal for reading out the charge (electron) voltages, and output is an analog-to-digital conversion quantized data output of the analog-to-digital converter ADC 205).

The specific principle process of resetting the integrating and readout circuit is as follows:

when resetting the integrated charge (electron) voltage, the reset integration control signal reset effectively turns on transistor M1, the incident photon 200 undergoes a photon-to-photon conversion into an accumulated charge (electron) via photodiode 203, at which time the read control signal read is deactivated and turns off transistor M3, no read occurs;

when the read control signal read is used to read out the charge (electron) voltage, the transistor M3 is turned on, the photodiode 203 accumulates the charge (electrons) and the charge (electrons) is output to the ADC205 through the transistors M2 and M3 to convert the quantized data output, and the reset integration control signal reset is disabled to turn off the transistor M1 and not accumulate the charge (electrons).

The ADC205 has a significance of not less than 8 bits with an analog-to-digital conversion quantization resolution; e.g. 8, 10, 12, etc., forming at least 2⁸＝256LSB，2¹⁰＝1024LSB，2¹²4096LSB quantization resolution.

The physical dimensions (PS) of each photodiode 203 imaging pixel cell in the imaging array of image sensor 105 that independently receives the RGB-IR wavelength channels satisfy the following condition: 1um/pixel ≦ PS ≦ 3um/pixel (microns per pixel);

the pixel cell photoelectrically converted values YR of the independently received R wavelength channels in the imaging array of the image sensor 105 are:

YR＝FF*QR*GAIN*EXP*ADCG*E*PSU (EQ3)

the pixel cells of the G wavelength channel in the imaging array of the image sensor 105 are independently received and photoelectrically converted to values YG of:

YG＝FF*QG*GAIN*EXP*ADCG*E*PSU (EQ4)

the value YB of the photoelectric conversion of the pixel cells of the B wavelength channel independently received in the imaging array of the image sensor 105 is:

YB＝FF*QB*GAIN*EXP*ADCG*E*PSU (EQ5)

the pixel cell photoelectrically converted value YIR of the IR wavelength channel independently received in the imaging array of image sensor 105 is:

YIR＝FF*QIR*GAIN*EXP*ADCG*E*PSU (EQ6)

wherein: ff (fill factor) mentioned above is a fill factor of the microlens 201(micro lenses);

EXP is the reset integration time or exposure time of the imaging array of the image sensor 105, in units: s seconds; EXP sync equals LED illumination source 106 radiance time;

GAIN is the digital and analog GAIN of the imaging array of image sensor 105, unitless;

ADCG is the ADC voltage analog-to-digital conversion quantization resolution of the imaging array of image sensor 105, unit: LSB/V, numerical bits per volt;

e is the radiance or radiance received by the imaging array of image sensor 105, in units: lux (lux) or mw/cm²(every milliwatt per square centimeter);

E＝C*β*I/WD²*cos²ψ*(1/FNO)² (EQ7)

wherein: i is the LED illumination source 106 radiation intensity in milliwatts per steradian (mw/sr); psi is the angle between the radiation position of the LED illumination source 106 and the optical axis 100 of the imaging system; WD is the focal working object distance of the optical imaging system; FNO is the numerical aperture of the optical imaging lens 102, i.e., the reciprocal of the relative pitch; beta is the biological tissue optical effect reflectivity of the imaging object (iris or human face) (the wavelength radiated by the LED lighting source is absorbed, reflected and scattered by the iris or human face biological tissue to generate the biological tissue optical effect reflectivity); c is the optical coefficient of the optical imaging system;

C＝1/16*cos⁴ω/(1+OM)²(EQ8) wherein: omega is the object space field angle of the incident light; OM is the optical magnification of the photoelectric imaging system;

PSU is the physical scale area unit ratio of each photodiode imaging pixel cell of the imaging array of image sensor 105; PSU ═ PS)/cm²；

QR, QG, QB, QIR are the photon-to-photon conversion constants for each imaging pixel element of the imaging array of image sensor 105 that independently receives a wavelength channel; the digital values YR, YG, YB, YIR photoelectrically converted by the pixel cells independently receiving the wavelength channels in the imaging array of the image sensor 105 are further output as the imaging image RAW pixel data I { YR, YG, YB, YIR }.

The image sensor 105 imaging array has at least 1920 x 1080 number of RGB-IR imaging pixel elements.

The RGB-IR imaging pixel cells of the imaging array of image sensor 105 have a 4-way 2 x 2 cross-space arrangement format.

Fig. 4 is a schematic diagram of a pixel cell 4 direction 2 × 2 cross-over spacing arrangement format of an imaging array of image sensor 105RGB-IR wavelength channels according to embodiment 1 of the present invention;

fig. 4 shows that the RGB-IR wavelength channels are repeatedly formed in a 2 x 2 cross-space arrangement every 4 directions. The pixels of the same wavelength channels RGB-IR of the imaging array of the image sensor 15 adopt a 4-direction cross-interval sampling mode, that is, pixels Pixel _ SC of the same wavelength channel are in the current direction, pixels Pixel _ SH of the same wavelength channel are in the horizontal direction, pixels Pixel _ SV of the same wavelength channel are in the vertical direction, and pixels Pixel _ SD of the same wavelength channel are in the diagonal direction. Reference is made in detail to the 4 same wavelength channel pixels indicated in the schematic diagram 5.

The image sensor 105 according to embodiment 1 of the present invention may be further reduced in size by using a package such as a Bare Die (COB), a shellt CSP, a NeoPAC CSP, or a TSV CSP.

The LED illumination light source (106RGB and 106IR-LED) according to embodiment 1 of the present invention includes: RGB and IR imaging wavelengths of independent radiation. Further, the RGB-LED illumination source (106RGB) has: the RGB imaging wavelengths of the radiation mix to form white visible light.

The LED illumination source 106 is comprised of a semiconductor light emitting diode that is physically configured in the same manner as a semiconductor photodiode, but in an opposite manner, by applying a current to cause a photon-hole pair at the semiconductor PN junction to produce a photon-to-electron quantum conversion that radiates a photon 200 outward.

Further, the LED illumination source (106RGB and 106IR-LED) according to embodiment 1 of the present invention includes: a convex lens or a concave mirror that controls the half-peak radiation angle. The half-peak radiation angle omega satisfies the following conditions:

Ω≥FOV；

the FOV is a full field angle of the imaging system;

FOV≥2*arctan((DI*PS)/(2*EFL))；

wherein: EFL is the equivalent focal length of the optical imaging lens 102; DI is the number of image plane diagonal pixel cells of the imaging array of image sensor 105; PS is the physical dimension of the pixel cells of the imaging array of image sensor 105;

the LED is theoretically a Lambert point light source radiating light at an angle of 360 degrees, and the light radiated by the LED point light source can be converged by adopting a convex lens or a concave reflector to play a role in controlling the half-peak radiation angle of an LED illumination light source. The convex lens may be made of an optical matrix material such as optical grade PMMA, optical grade PC, etc., and the concave mirror may be made of a high reflectivity metal matrix material.

The LED illumination light source (106RGB and 106IR-LED) according to embodiment 1 of the present invention includes: one or more different radiation angle positions for optimizing the imaging field of view and imaging quality effects of the optoelectronic imaging system and providing in vivo detection of optical reflections at different positions of the cornea. E.g. using different radiation angle positions (left Psrl, right Psrr, left and right Psrl & Psrr) to the left and/or right of the optical axis 100 of the imaging system.

The LED illumination light source (106RGB and 106IR-LED) according to embodiment 1 of the present invention includes: the continuous or pulsed radiation time and radiation intensity, synchronized with the image sensor 105, are used to jointly optimize the imaging quality effect of the optoelectronic imaging system. The LED illumination sources (106RGB and 106 IR-LEDs) may be further reduced in size using packaging such as SMD surface mount devices.

The optical filter (101 or 104) according to embodiment 1 of the present invention includes: filtering RGB and IR imaging wavelengths, transmitting light within the RGB and IR imaging wavelength ranges, reflecting and/or absorbing light outside the RGB and IR imaging wavelength ranges.

Further, the optical filter (101 or 104) described in embodiment 1 of the present invention has:

the light filtering rate Fi in the RGB and IR imaging wavelength ranges is less than or equal to 10.0 percent,

the light filtering rate Fo outside the RGB and IR imaging wavelength ranges is more than or equal to 99.9 percent;

or equivalent

The light transmittance Ti in the RGB and IR imaging wavelength ranges is more than or equal to 90.0 percent,

the light transmittance To outside the RGB and IR imaging wavelength ranges is less than or equal To 0.1 percent.

The optical filter (101 or 104) can be realized by performing surface multilayer coating on optical matrix materials such as optically transparent glass, colored glass, optical plastics and the like, and the thickness of the optical filter (101 or 104) is less than or equal to 0.3mm, furthermore, as an equivalent understanding of the invention, the optical filter (101 or 104) can be equivalently replaced by performing multilayer coating on the surface of the optical imaging lens 102 as an optical matrix.

The optical imaging lens 102 according to embodiment 1 of the present invention includes: physical refraction focuses the RGB and IR imaging wavelengths. Further, the optical imaging lens 102 according to embodiment 1 of the present invention has the following characteristics for RGB and IR imaging wavelengths:

the maximum surface reflectivity Rmax is less than or equal to 1.0 percent, and the average surface reflectivity Ravg is less than or equal to 0.35 percent;

or equivalent

The minimum surface transmittance Tmin is more than or equal to 99.0 percent, and the average surface transmittance Tavg is more than or equal to 99.65 percent.

The optical imaging lens 102 can be realized by performing multilayer antireflection or antireflection coating on the surface of an aspheric optical plastic such as optical-grade PMMA (polymethyl methacrylate), optical-grade PC (polycarbonate) and other optical matrix materials; the optical plastic injection molding process can be realized by 3-5P aspheric optical plastic, and the total optical length of the TTL is less than or equal to 6 mm.

The optical imaging lens has: focal length EFL, numerical aperture FNO satisfy:

3mm≤EFL≤6mm，2.0≤FNO≤4.0。

the optical imaging lens 102 is configured as a fixed focal length lens, including any one of a liquid-driven lens, a liquid crystal-driven lens, a VCM voice coil-driven lens, a MEMS-driven lens, an EDOF wavefront phase modulation lens, or a wafer-level micro-array lens.

The liquid driving lens comprises a fixed focusing lens, a liquid lens and a voltage driver for controlling the liquid lens;

the liquid crystal driving lens comprises a fixed focusing lens, a liquid crystal lens and a voltage driver for controlling the liquid crystal lens;

the liquid driving lens and the liquid crystal driving lens realize the automatic focusing function by changing the diopter of incident light, namely optical power adjustment.

The VCM voice coil driving lens comprises a fixed focusing lens, a VCM voice coil and a current driver for controlling the VCM voice coil;

the VCM voice coil driving lens realizes an automatic focusing function by changing optical back focus and optical image distance adjustment.

The MEMS (micro-electro-mechanical system) driving lens comprises a fixed focusing lens, a MEMS lens and an electrostatic driver for controlling the MEMS lens.

The MEMS driving lens realizes the automatic focusing function by changing the optical position of the MEMS lens.

The wafer level array micro-lens realizes the 3D full-depth-of-field reconstruction function through micro-lens array Computational Imaging (Computational Imaging).

The EDOF wave-front phase modulation lens comprises a lens and a wave-front phase modulation optical element;

the EDOF wave-front phase modulation is modulated by the wave-front phase modulation optical element, and then the inverse filtering demodulation reconstruction is carried out to realize the function of expanding the depth of field.

The EDOF wavefront phase modulation lens has the advantages of low cost, small volume, simple structure, no complex driving and the like. Therefore, the embodiment 1 of the present invention is preferably described in detail by taking an EDOF wavefront phase modulation lens as an example, which is used for imaging, and ensures that the EDOF wavefront phase modulation lens has a depth of field (depth of field) range more than 10 times that of a conventional optical imaging system under the condition of maximizing the luminous flux, and simultaneously simplifies the design of the field of view (view of field) and aberration correction of the optical system.

The wavefront phase modulating optical element acts as a phase pupil between the lenses.

Pupil phase modulation function Φ (x, y) defining the wavefront phase modulating optical element to have odd symmetry:

Φ(-x，-y)＝-Φ(x，y)

wherein: m and N are orders, and alpha mn is a numerical coefficient.

In practical application, in consideration of the requirements of numerical calculation, complexity of practical manufacturing and the like, the specific embodiment 1 of the present invention generally adopts a low order with an order less than 9, for example, adopts 7, 5 and 3 as orders.

The wavefront phase modulation optical element of embodiment 1 of the present invention can be designed and manufactured by a micron-sized aspheric injection molding method, which can reduce the cost and has a simple structure, and is easy for mass production.

The wavefront phase modulation optical system has an optical point spread function PSF (u, v; theta)

PSF(u，v；θ)＝|h(u，v；θ)|²

Wherein: p (x, y) is a pupil function of the optical system,

p (x, y) ═ 1, when the integration parameter (x, y) is included in the pupil range;

p (x, y) ═ 0, when the integration parameter (x, y) is not included in the pupil range;

the pupil function can equally well be expressed as a domain area range of two-dimensional fixed integration, i.e. the domain area integration range defining 2-dimensional fixed integration is the pupil range. (x, y) are points of the pupil plane and (u, v) are points of the image plane.

Theta is a diffraction wave aberration or defocus parameter; λ is imaging wavelength, f is equivalent focal length of optical system, d_oDistance from the entrance pupil plane to the object plane, d_iIs the exit pupil plane to image plane distance, A is the pupil area, and Zernike (x, y) is the Zernike aberration function of the optical system;

in fact, considering that the optical system has spherical characteristics, the two-dimensional integral can be equivalently expressed by polar integral. The point spread function PSF (u, v; theta) is even symmetric according to the definition of the pupil phase modulation function phi (x, y).

A pupil phase modulation function Φ (x, y) of a wavefront phase modulation optical system having a Modulation Transfer Function (MTF) and diffraction-aberration (diffraction-aberration) space/frequency domain combination optimized satisfies a condition: the diffraction wave aberration optimization degree J is globally minimized, and J is 0 in theory.

Wherein: the diffracted wave aberration optimality J is determined by the following definition:

wherein: the [ -theta 0, theta 0] is the symmetric range of the diffraction wave aberration or defocus parameter specified in practical application;

meanwhile, according to the optical theory, the wavefront phase modulation optical system has the Fourier transform pair of which the optical transfer function OTF (s, t; theta) is PSF (u, v; theta), and the following reasoning is carried out:

the modulation transfer function optimization M is determined by the following definition:

from the above definitions and inferences, it can be demonstrated that the pupil phase modulation function Φ (x, y) has a spatial/frequency domain combined optimization of the modulation transfer function and the diffracted wave aberration under the condition that the global minimization of the diffracted wave aberration optimization degree J is satisfied. And in the theoretical condition that J is 0, the wavefront phase is a fixed constant relative to the diffraction wave aberration, and the original image can be restored through simple digital demodulation processing.

The image plane image O (u, v) imaged by the image sensor 105 is restored by digital signal processing image demodulation, and as a result, the original digital image I (x, y) is reconstructed. The digital signal processing image demodulation and recovery specifically comprises the following steps:

I(x，y)＝H(u，v)*g(u，v)＝∫∫H(x-u，y-v)g(u，v)du dv

wherein H (u, v) ═ 0(u, v) -N (u, v);

0(u, v) is an image plane image imaged by the image sensor 105, N (u, v) is an equivalent noise function of the photoelectric imaging system, and g (u, v) ═ F^-1(1/MTF (s, t)), i.e., inverse Fourier transform of the inverse of MTF (s, t), MTF (s, t) is a predetermined Modulation Transfer Function (MTF) function of the wavefront phase modulation optical system, which represents a 2-dimensional function convolution integral.

Since the MTF (s, t) is determined for the predetermined optical system, g (u, v) is also determined, and the convolution scale of g (u, v) is also tightly-supported, the more recent equivalent noise function N (u, v) is also determined for the predetermined optical system. Therefore, the above-mentioned demodulation recovery of the digital signal processing image can be expressed in a mathematical discrete form, and the specific embodiment 1 of the present invention can optimize the integer code to be implemented in real time by the digital signal processing device such as FPGA or DSP, or implemented in real time by the software algorithm of the processor chip 109.

The embodiment 1 of the present invention is attributed to the fact that the iris recognition photoelectric imaging system and the front photoelectric imaging system have different optical imaging requirements, imaging wavelength, pixel spatial resolution, optical magnification, optical spatial resolution, and focusing work object distance range.

The iris identification photoelectric imaging system has the following optical imaging requirements:

the imaging wavelength WI of the iris recognition photoelectric imaging system meets the following requirements:

WI is more than or equal to 800nm and less than or equal to 900nm or more than or equal to 750nm and less than or equal to 850 nm;

the focusing working object distance WD of the iris recognition photoelectric imaging system satisfies the following conditions:

10cm≤WD≤30cm。

the optical magnification OM (optical magnification) of the iris recognition photoelectric imaging system should satisfy:

OM＝PS*PSR

wherein the following steps:

PS is the physical dimension of each imaging pixel unit of the image sensor;

PSR is the pixel space resolution of the iris recognition photoelectric imaging system;

The front photoelectric imaging system has the following optical imaging requirements:

the imaging wavelength WI of the front photoelectric imaging system meets the following conditions:

WI is more than or equal to 400nm and less than or equal to 700nm or WI is more than or equal to 400nm and less than or equal to 650nm

The focusing working object distance WD of the front photoelectric imaging system satisfies the following conditions:

30cm≤WD≤100cm。

the optical magnification om (optical magnification) of the front photoelectric imaging system should satisfy:

OM＝PS*PSR

wherein the following steps:

PS is the physical dimension of each imaging pixel unit of the image sensor;

PSR is the pixel spatial resolution of the front photoelectric imaging system;

The invention discloses a front-mounted photoelectric imaging method, which comprises the following steps:

1. processor chip 109 controls LED current driver 108 to drive LED illumination source 106(106RGB) to produce radiation in a continuous or synchronized pulse pattern of RGB imaging wavelengths;

2. after RGB imaging wavelength filtering and physical refractive focusing, the imaging array of the image sensor 105 independently receives 3RGB wavelength channels for global frame mode or rolling line mode reset integration (exposure) and readout;

3. the processor chip 109 respectively obtains original RAW pixel data I { YR, YG, YB } of the imaging image output by 3 channels with the same RGB wavelength in the imaging array;

4. the processor chip 109 drives the image sensor 105, the LED illumination light source 106 and the optical imaging lens 102 to focus according to the original RAW pixel data I { YR, YG, YB } of the imaging image and the photoelectric conversion relationship of the pixel unit, thereby implementing feedback control;

5. the processor chip 109 performs interpolation reconstruction between the original RAW data I { YR, YG, YB } pixels of 3 identical RGB wavelength channels in the imaging array;

6. the processor chip 109 outputs an interpolated reconstructed image I { r, g, b }, each pixel containing RGB pixel values, respectively.

To explain further, in the above-described steps, the imaging array of the image sensor 105 is N × M RGB-IR imaging units, the RAW data I { YR, YG, YB } of 3 same RGB wavelength channels are each (N/2) × (M/2) number of imaging units, and the (N/2) × (M/2) number of imaging unit pixels of each same wavelength channel are interpolated and reconstructed into N × M number of pixels. And (N/2) × (M/2) pixels passing through the same wavelength channel are respectively interpolated to reconstruct N × M pixels, namely, each pixel respectively comprises RGB pixel values.

To explain further, the above-mentioned photoelectric conversion relationship of the pixel unit in step 4 includes equations EQ3, EQ4, EQ 5. The processor chip 109 may feedback control the reset integration time, digital and analog gain settings, and feedback control the LED current driver 108 to drive the radiation intensity, radiation angular position, and radiation time of the LED illumination source 106 for improving the imaging quality according to the imaged image RAW pixel data I { YR, YG, YB } output by the image sensor 105 and the corresponding formulas EQ3, EQ4, EQ 5.

The focusing of the optical imaging lens 102 is realized by calculating the focal quality value feedback control of the original RAW pixel data I { YR, YG, YB } of an imaging image, and the focusing working object distance WD of the front photoelectric imaging system is at least 30cm-100 cm. Conventional auto-focus methods such as focus quality maximum peak iterative search may be employed.

The processor chip 109 may control the LED current driver 108 to drive the radiation intensity of the LED illumination source 106RGB according to the current ambient light level through the light sensor (according to the use situation, such a separate additional device may be disposed on the processor chip 109 by the current known technology, or such a light sensor function may be implemented by purchasing a corresponding processor chip on the market). Furthermore, in embodiment 1 of the present invention, if the light sensor determines that the current ambient light brightness is greater than 500-.

Further, the processor chip 109 can perform optical black level correction BLC, RGB channel auto white balance AWB, RGB channel color matrix correction CCM, lens edge shading correction lens shading correction, auto exposure feedback control AEC, auto gain feedback control AGC, and the like of the image sensor by the RAW pixel data of the imaged image output from the image sensor 105.

The iris identification photoelectric imaging method comprises the following steps:

1. the processor chip 109 controls the LED current driver 108 to drive the LED illumination source 106(106IR) to generate radiation at an IR imaging wavelength in a continuous or synchronized pulse pattern;

2. after IR imaging wavelength filtering and physical refractive focusing, the image sensor 105 imaging array independently receives IR wavelength channels for global frame mode or rolling row mode reset integration (exposure) and readout;

3. the processor chip 109 acquires original RAW pixel data I { YIR } of an imaging image output by the same IR wavelength channel in the imaging array;

4. the processor chip 109 drives the image sensor 105, the LED illumination light source 106 and the optical imaging lens 102 to focus according to the original RAW pixel data I { YIR } of the imaging image and the photoelectric conversion relation of the pixel units, so as to realize feedback control;

5. the processor chip 109 interpolates and reconstructs the original RAW data I { YIR } pixels of the same IR wavelength channel in the imaging array;

6. the processor chip 109 outputs an interpolated reconstructed image I { ir }.

To explain further, in the above steps, the imaging array of the image sensor 105 is N × M RGB-IR imaging units, the RAW data I { YIR } of the same IR wavelength channel is (N/2) × (M/2) number of imaging units, and the (N/2) × (M/2) number of imaging units of the same IR wavelength channel are interpolated and reconstructed into N × M number of pixels. And (N/2) × (M/2) pixels passing through the same IR wavelength channel are interpolated to reconstruct N × M number of pixels.

To explain further, the step-4 pixel cell photoelectric conversion relationship described above includes the formula EQ 6. The processor chip 109 may feedback control the reset integration time, digital and analog gain settings of the image sensor 105, feedback control of the LED current driver 108 to drive the radiation intensity, radiation angle position, and radiation time of the LED illumination source 106 for improved imaging quality based on the imaged RAW image pixel data output by the image sensor 105 and the formula EQ 6. The focusing of the optical imaging lens 102 is realized by calculating the focal quality value feedback control of the original RAW pixel data I { YIR } of an imaging image, and the focusing working object distance WD of the iris recognition photoelectric imaging system is at least 10cm-30 cm. Conventional auto-focus methods such as focus quality maximum peak iterative search may be employed.

Further, the processor chip 109 can perform optical black level correction BLC, automatic exposure feedback control AEC, automatic gain feedback control AGC of the image sensor by the imaged image RAW pixel data output by the image sensor 105.

As a simplified example of the same understanding as the specific embodiment 1 of the present invention, the iris recognition photoelectric imaging method includes the following steps:

1. the processor chip 109 controls the LED current driver 108 to drive the LED illumination source (106IR) to generate radiation at an IR imaging wavelength in a continuous or synchronized pulse pattern;

4. the processor chip 109 drives the image sensor 105, the LED illumination light source (106IR) and the optical imaging lens 102 to focus according to the photoelectric conversion relation between the original RAW pixel data I { YIR } of the imaging image and the pixel unit, so as to realize feedback control;

5. the processor chip 109 outputs RAW data I { YIR } pixels for the same IR wavelength channel in the imaging array.

The simplified example is to be understood as the embodiment 1 of the present invention, and the iris recognition photoelectric imaging method removes the interpolation reconstruction step between the I { YIR } pixels of the original RAW data.

The interpolation reconstruction described in embodiment 1 of the present invention uses an interpolation algorithm for original RAW data of 4-direction neighboring pixels between pixels of the same wavelength channel in an imaging array.

The interpolation algorithm includes the tradition:

nearest neighbor interpolation, Linear interpolation, bilinear interpolation, bicubic interpolation, Spline interpolation, etc.

Considering that image textures such as an iris or a human face have natural continuity characteristics, the invention provides a faster and more effective interpolation algorithm based on the correlation among image pixels, and referring to the schematic diagram of fig. 5, the method comprises the following steps:

1. sampling pixel values of imaging image original RAW to-be-interpolated pixel data 4 direction crossing intervals output by the same wavelength channel, wherein the pixel values are respectively as follows: the Pixel of the same wavelength channel in the current direction Pixel _ SC, the Pixel of the same wavelength channel in the horizontal direction Pixel _ SH, the Pixel of the same wavelength channel in the vertical direction Pixel _ SV, the Pixel of the same wavelength channel in the diagonal direction Pixel _ SD;

the same wavelength channel pixels 4-way interleaved sampling is due to the fact that the pixel cells of the same wavelength channel of the imaging array are arranged in a 4-way 2 x 2 interleaved format.

2. Calculating the interpolation of 4 direction adjacent pixels of the Pixel data to be interpolated, Pixel _ C, Pixel _ H, Pixel _ V and Pixel _ D:

pixel _ C of the current direction is Pixel _ SC;

adjacent Pixel interpolation Pixel _ H in the horizontal direction is (Pixel _ SH + Pixel _ SC)/2;

the neighboring Pixel interpolation Pixel _ V in the vertical direction is (Pixel _ SV + Pixel _ SC)/2;

a diagonal direction adjacent Pixel interpolation Pixel _ D ═ 4 (Pixel _ SH + Pixel _ SV + Pixel _ SD + Pixel _ SC);

3. and (3) circulating the step 1 to the step 2, traversing and calculating all original RAW pixel data to be interpolated in the imaged image to form final complete interpolated image data.

As an equivalent understanding, the interpolation algorithm for the 4-direction neighboring pixels can be generalized as well.

The invention provides a high-safety living body detection method for an iris forgery prevention artifact, which has real-time detection capability for the iris forgery prevention artifact and is used for ensuring the safety of biological identification, and the method comprises the following steps:

one or more of the following should be used:

a real-time detection method for optical activity characteristics of biological tissues generated by RGB-IR imaging wavelength radiation;

a real-time detection method for pupil iris diameter change rate biological tissue activity characteristics generated by RGB-IR imaging wavelength radiation;

3, a real-time detection method for the optical reflection position of the cornea generated by RGB-IR imaging wavelength radiation;

4. real-time detection method for eyeball physiological motion activity characteristics.

The flow processing speed of the real-time detection method for detecting the living body of the anti-counterfeiting artifact of the iris is higher than the image acquisition frame rate; the image acquisition frame rate is 120fps, 90fps, 60fps and 30fps, and the reliability of the iris anti-counterfeiting object in-vivo detection method is higher as the image acquisition frame rate is higher.

The invention discloses a real-time detection method for optical activity characteristics of biological tissues generated by RGB-IR imaging wavelength radiation, which comprises the following steps:

1. the processor chip 109 controls the LED current driver 108 to drive the LED illumination sources 106(106RGB and 106IR) to generate RGB imaging wavelength radiation and IR imaging wavelength radiation in real time;

2. the processor chip 109 acquires real-time imaging images IRGB and IIR output by the RGB wavelength channel and the IR wavelength channel of the imaging array of the image sensor 105 in real time;

3. the processor chip 109 respectively calculates the contrast ratios Csk, Csi, Cip, Csip and Ckip of the RGB imaging image IRGB and the IR imaging image IIR in step 2 in real time, which are IRGB _ Csk, IRGB _ Csi, IRGB _ Cip, IRGB _ Csip, IRGB _ Ckip, IIR _ Csk, IIR _ Csi, IIR _ ip, IIR _ Csip and IIR _ Ckip;

wherein:

csk is the contrast between the skin region and the iris region;

csi is the contrast between the scleral area and the iris area;

cip is the contrast between the iris region and the pupil region;

csip is the mutual contrast between the scleral region, the iris region and the pupil region;

ckip is the skin area, the mutual contrast between the iris area and the pupil area;

Csk＝S(Iskin)/S(Iiris)；

Csi＝S(Isclera)/S(Iiris)；

Cip＝S(Iiris)/S(Ipupil)；

Csip＝(S(Isclera)-S(Iiris))/(S(Iiris)-S(Ipupil))；

Ckip＝(S(Iskin)-S(Iiris))/(S(Iiris)-S(Ipupil))；

ipoul represents the pupil region pixel;

iiris denotes iris area pixels;

isclera denotes scleral region pixels;

iskin represents a skin region pixel;

the function S is a pixel statistical evaluation function of a corresponding region, and the method adopted by the pixel statistical evaluation function comprises the following steps: histogram statistics, frequency statistics, mean statistics, weighted mean statistics, median statistics, energy value statistics, variance statistics, space-frequency domain filters, and the like; the corresponding region pixel statistical evaluation function S of the present invention is not limited to the above example, and other methods should be equally understood.

4. The processor chip 109 calculates the image contrast activity change rates Fsk and Fsi, Fip, Fsip, Fkip of the RGB imaging wavelength radiation and the IR imaging wavelength radiation, respectively, in real time;

wherein:

Fsk＝IRGB_Csk/IIR_Csk*100％；

Fsi＝IRGB_Csi/IIR_Csi*100％；

Fip＝IIR_Cip/IRGB_Cip*100％；

Fsip＝IRGB_Csip/IIR_Csip*100％；

Fkip＝IRGB_Ckip/IIR_Ckip*100％；

5. according to the preset value of the optical activity characteristic of the biological tissue radiated by the RGB-IR imaging wavelength and the corresponding change rate of the activity contrast of the data values Fsk, Fsi, Fip, Fsip and Fkip in the step 4, judging that any one or more of the conditions Fsk is more than 300 percent, Fsi is more than 300 percent, Fip is more than 300 percent, Fsip is more than 900 percent and Fkip is more than 900 percent, and realizing the real-time detection of the living iris state.

Furthermore, due to the different absorption and scattering optical activity characteristics of the iris melanocytes of different races on the radiation of RGB-IR imaging wavelength, the preset values of the optical activity characteristics of biological tissues are different, for example, the light iris race with too few melanocytes has the corresponding change rate of the activity contrast of the independent B wavelength channel and the independent IR wavelength channel, and the dark iris race with too many melanocytes has the corresponding change rate of the activity contrast of the independent B wavelength channel and the independent IR wavelength channel. Therefore, it is understood that the above-mentioned predetermined value of the optical activity characteristic of the biological tissue may have a different range of variation. However, regardless of the above range of variation, it was determined that iris biological tissue has different dynamic activity contrast change rates for biological tissue optical activity characteristics generated by radiation of different RGB-IR imaging wavelengths to reflect iris liveness.

Fig. 6 is a schematic diagram of defining a contrast region of an iris image according to embodiment 1 of the present invention. As shown in the schematic 6 notation, where Isclera, Iiris, Ipupil, Iskin defines:

1 is a pupil area Ipoul which represents the pixels of the pupil area;

2 denotes iris area pixels for iris area Iiris;

3, representing the sclera area pixel by the sclera area Isclera;

4 skin area Iskin represents skin area pixels;

the method for detecting the pupil iris diameter change rate biological activity characteristics generated by the RGB-IR imaging wavelength radiation in the specific embodiment of the invention comprises the following steps:

1. the processor chip 109 controls the LED current driver 108 to drive the LED illumination sources 106(106RGB and 106IR) to generate RGB and/or IR imaging wavelength radiation with different intensities dil, con and time Δ t, respectively, in real time, so as to stimulate the pupil to actively expand and contract biological tissues;

2. the processor chip 109 respectively obtains real-time imaging images Idil and Icon under the conditions of different radiation time delta t and intensity dil output by the RGB-IR wavelength channel of the imaging array of the image sensor 105 and con;

3. the processor chip 109 respectively calculates the ratio rho data of the pupil to iris diameter of the iris image in the imaging image Idil and the iris in the step 2 in real time, wherein the ratio rho data is rhodil and rhocon;

ρ＝Dpupil/Diris，

the Dpupil is the length of a pupil diameter pixel;

the Diris is the length of an iris diameter pixel;

4. the processor chip 109 calculates in real time the corresponding activity rate of change Δ ρ ═ p dil- ρ con)/Δ t × 100%:

5. and (4) judging the condition that the Delta rho is more than 10% according to preset values of activity expansion and contraction of biological tissues generated by stimulating the pupil in real time under different intensities and time conditions and the corresponding activity change rate of the data value Delta rho in the step 4, and realizing the real-time detection of the living iris state.

Fig. 7 is a diagram illustrating the pupil and iris diameter of an iris image defined according to the present invention. As indicated in the schematic 7 notation, where dpipil, Diris defines:

the Dpupil is the length of a pupil diameter pixel;

the Diris is the length of an iris diameter pixel;

the invention relates to a cornea optical reflection position detection method generated by RGB-IR imaging wavelength radiation, which comprises the following steps:

1. the processor chip 109 controls the LED current driver 108 to drive the LED illumination source 106(106RGB or 106IR) to generate RGB and/or IR imaging wavelength radiation under different position conditions of the left side Psrl, the right side Psrr and the left and right 2 sides Psrl & Psrr, respectively, in real time, to form corneal optical reflection points at different positions;

2. the processor chip 109 respectively acquires real-time imaging images Isr output by the RGB-IR wavelength channels of the imaging array of the image sensor 105 in real time;

3. the processor chip 109 respectively calculates the corneal optical reflection point position data Psr of the imaging image Isr in the step 2 in real time;

4. according to preset values of the LED illumination light source 106 under different position conditions and the corneal optical reflection point position Psr calculated in the step 3, whether the corneal optical reflection point position Psr meets the corresponding LED illumination light source position condition is judged:

if the position of the LED illumination light source is Psrl, Psr ═ Psrl should be met;

if the position of the LED illumination light source is Psrr, Psr should be equal to Psrr;

if the position of the LED illumination light source is Psrl & Psrr, Psr & Psrr should be satisfied;

the real-time detection of the living iris state is realized.

FIG. 8 is a schematic diagram of the optical reflection points of the present invention defining different positions of the cornea of an iris image. As indicated in the scheme 8, where Psrl, Psrr, Psrl & Psrr define:

the Psrl is a cornea optical reflection point at the left position generated by the LED illumination light source;

the Dsrr is a corneal optical reflection point at the right position generated by the LED illumination light source;

the Psrl & Psrr is a cornea optical reflection point which is generated by the LED illumination light source at the left and right 2-side positions.

Further, it is understood that, the same or equivalent, the method for detecting the corneal optical reflection position by using RGB-IR imaging wavelength radiation according to the present invention, if RGB-IR imaging wavelength radiation is not generated, the corresponding imaging image does not have the corneal optical reflection point position, so that the method for detecting the iris living body state in real time by using the method for generating and/or not generating RGB-IR imaging wavelength radiation in the corresponding imaging image and/or not generating the corneal optical reflection point position can be equivalently used.

The invention discloses a real-time detection method for activity characteristics of eyeball physiological movement, which comprises the following steps of detecting eyelid movement activity characteristics generated by eyeball physiological movement in real time:

1. the processor chip 109 controls the LED current driver 108 to drive the LED illumination source 106(106RGB or 106IR) to generate RGB-IR imaging wavelength radiation in real time;

2. the processor chip 109 acquires a real-time imaging image Iem output by the RGB-IR wavelength channel of the imaging array of the image sensor 105 in real time;

3. the processor chip 109 calculates eyelid movement characteristic degree data FM generated by the eyeball physiological movement of the imaging image Iem in the step 2 in real time;

wherein:

the degree of eyelid movement characteristics FM generated by the physiological movement of the eyeball is defined as:

EM＝Visual_Iris/All_Iris*100％；

the All _ Iris is the pixel number of the whole area of the Iris in the imaging image Iem;

visual _ Iris is the number of pixels of an effective area of an Iris formed by eyelid movement in the imaging image Iem;

4. calculating an activity change rate value delta EM of the eyelid movement characteristic degree EM generated by the physiological movement of the eyeballs in real time;

wherein:

the activity change rate value Δ EM of the degree of eyelid movement characteristics generated by the physiological movement of the eyeball is defined as: the absolute difference of EM between consecutive images Iem;

5. and (4) judging the condition that the delta EM is more than 10% according to the preset value of the activity change rate of the eyelid movement characteristic degree generated by the eyeball physiological movement and the activity change rate value delta EM of the eyelid movement characteristic degree data EM generated by the eyeball physiological movement calculated in the step 4, and realizing the real-time detection of the iris living body state.

Fig. 9 is a schematic diagram illustrating the degree of eyelid movement characteristics generated by the physiological movement of the eyeball according to the present invention. As indicated in the diagram of fig. 9, the dotted line All _ Iris in the imaged image represents the number of pixels in the entire area of the Iris, and the solid line Visual _ Iris represents the number of pixels in the effective area of the Iris.

The invention discloses a real-time detection method for activity characteristics of eyeball physiological motion, which comprises the following steps of detecting the activity characteristics of off-axis strabismus generated by the eyeball physiological motion in real time:

2. the processor chip 109 acquires Ieg a real-time imaging image output by the RGB-IR wavelength channels of the imaging array of the image sensor 105 in real-time;

3. the processor chip calculates off-axis strabismus characteristic degree data EG generated by the physiological eyeball motion of the imaging image Ieg in step 2 in real time;

wherein:

the degree of off-axis strabismus characteristic EG produced by the physiological movement of the eyeball is defined as:

EG＝S_Iris/L_Iris*100％；

s _ Iris is the short axis pixel length of the Iris formed by off-axis squint in imaged image Ieg;

l _ Iris is the Iris long axis pixel length formed by off-axis squint in imaged image Ieg;

4. calculating an activity change rate value delta EG of the off-axis strabismus characteristic degree EG generated by the physiological movement of the eyeball in real time;

wherein:

the activity change rate value delta EG of the degree of off-axis strabismus characteristic generated by the physiological movement of the eyeball is defined as: absolute difference in EG between consecutive images Ieg;

5. and (4) judging the condition that the delta EG is more than 10% according to the preset value of the activity change rate of the off-axis strabismus characteristic degree generated by the physiological movement of the eyeball and the activity change rate value delta EG of the off-axis strabismus characteristic degree data EG generated by the physiological movement of the eyeball calculated in the step 4, and realizing the detection of the living iris state.

Fig. 10 is a schematic diagram illustrating the degree of the activity of the off-axis strabismus physiological movement generated by the physiological movement of the eyeball according to the present invention. As indicated in the schematic diagram 10, S _ Iris in the imaged image represents the Iris short axis pixel length and L _ Iris represents the Iris long axis pixel length.

The specific embodiments and features described herein may be implemented within the same or equivalent understanding as well, such as imaging wavelength range changes, image sensor changes, LED illumination source changes, optical filter changes, optical imaging lens changes, optical path changes, device substitutions, and the like.

Finally, it is also noted that the above-mentioned lists merely illustrate a few specific embodiments of the invention. It is obvious that the invention is not limited to the above embodiments, but that many variations are possible. All modifications which can be derived or suggested by a person skilled in the art from the disclosure of the present invention are to be considered within the scope of the invention.

Claims

1. A human face/iris anti-counterfeit living body detection system based on RGB-IR imaging is characterized by comprising:

an image sensor having an imaging array configured as RGB-IR wavelength channels with independent receive functionality;

an LED illumination source configured to have an RGB-IR radiation imaging wavelength, different radiation intensities and radiation times, one or more different radiation angular positions;

the processor chip is configured to control the radiation RGB-IR imaging wavelength, the radiation intensity, the radiation time and the radiation angle position of the LED illumination light source in real time, acquire the RGB-IR imaging image output by the independent wavelength channel of the image sensor in real time and detect and calculate the living body characteristics of the imaging image in real time;

wherein the LED illumination source and the image sensor are configured to have synchronized continuous or pulsed radiation time and radiation intensity;

the system adopts a real-time detection method of the activity characteristics of the physiological movement of the eyeballs to realize the in-vivo detection of the anti-counterfeiting object;

the method for realizing the anti-counterfeiting object in-vivo detection comprises the step of detecting the eyelid movement activity characteristics generated by the physiological movement of eyeballs in real time, and specifically comprises the following steps:

a. generating RGB-IR imaging wavelength radiation in real time;

b. acquiring a real-time imaging image Iem output by an RGB-IR wavelength channel of an imaging array of an image sensor in real time;

c. calculating eyelid movement characteristic degree data EM generated by the eyeball physiological movement of the imaging image Iem in the step b in real time;

d. calculating an activity change rate value (delta EM) of the eyelid movement characteristic degree EM generated by the physiological movement of the eyeball in real time;

e. judging whether a preset value condition is met or not according to an activity change rate preset value of eyelid movement characteristic degrees generated by eyeball physiological movement and an activity change rate value (Delta EM) of eyelid movement characteristic degree data EM generated by the eyeball physiological movement calculated in the step d, and if so, judging that the iris is in a living body state;

in the step c:

the degree EM of the eyelid movement characteristics generated by the physiological movement of the eyeball is defined as:

EM＝Visual_Iris/All_Iris*100％；

and d, defining the activity change rate value (delta EM) of the eyelid movement characteristic degree in the step d as: absolute difference of EM between the continuous images Iem in the step c;

and e, judging the condition as delta EM being more than 10% by the preset value of the activity change rate value delta EM of the eyelid movement characteristic degree EM in the step e.

2. The face/iris anti-counterfeit living body detection system according to claim 1, wherein the method for realizing anti-counterfeit living body detection comprises the step of detecting the activity characteristic of off-axis strabismus generated by eyeball physiological motion in real time, and specifically comprises the following steps:

A. generating RGB-IR imaging wavelength radiation in real time;

B. acquiring Ieg a real-time imaging image output by the RGB-IR wavelength channels of the imaging array of the image sensor in real-time;

C. calculating off-axis strabismus characteristic degree data EG generated by the physiological eyeball motion of the imaging image Ieg in the step B in real time;

D. calculating an activity change rate value (delta EG) of the off-axis strabismus characteristic degree EG generated by the physiological movement of the eyeball in real time;

E. judging whether a preset value condition is met or not according to an activity change rate preset value of the degree of the off-axis strabismus characteristic generated by the physiological movement of the eyeball and an activity change rate value (delta EG) of the off-axis strabismus characteristic degree data EG generated by the physiological movement of the eyeball calculated in the step D, and if so, judging that the iris is in a living body state;

in the step C:

the degree EG of off-axis strabismus characteristics generated by the physiological movement of the eyeball is defined as:

EG＝S_Iris/L_Iris*100％；

and D, defining an activity change rate value delta EG of the off-axis strabismus characteristic degree in the step D as: absolute difference in EG between consecutive images Ieg in step C;

and E, judging the condition that the preset value of the activity change rate value delta EG of the off-axis strabismus characteristic degree EG is more than 10 percent.