CN215642742U

CN215642742U - Multi-modal biological recognition module and multi-modal biological recognition device

Info

Publication number: CN215642742U
Application number: CN202121291707.4U
Authority: CN
Inventors: 李小成; 彭程
Original assignee: Beijing Eyes Intelligent Technology Co ltd; Shenzhen Aiku Smart Technology Co ltd; Beijing Eyecool Technology Co Ltd
Current assignee: Beijing Eyes Intelligent Technology Co ltd; Shenzhen Aiku Smart Technology Co ltd; Beijing Eyecool Technology Co Ltd
Priority date: 2021-06-08
Filing date: 2021-06-08
Publication date: 2022-01-25
Anticipated expiration: 2031-06-08

Abstract

The utility model discloses a multi-mode biological recognition module and multi-mode biological recognition equipment, which belong to the field of biological recognition, wherein the multi-mode biological recognition module comprises an FPGA chip with an image preprocessing AI module and a USB interface chip, wherein: the USB interface chip is connected with the FPGA chip and is used as a communication interface with an upper computer and a main power supply interface of the multi-mode biological recognition module; the FPGA chip is connected with a face lens, an iris lens, a near-infrared light supplement lamp, a motor and a distance sensor, and the iris lens and the face lens are driven to rotate by the motor. The utility model reduces the data volume transmitted by USB, has low image compression degree, reduces the consumption of upper computer resources, improves the identification accuracy, and has simple realization mode and low cost.

Description

Multi-modal biological recognition module and multi-modal biological recognition device

Technical Field

The utility model relates to the field of biological identification, in particular to a multi-mode biological identification module and multi-mode biological identification equipment.

Background

The iris recognition technology is one of human body biological recognition technologies, and compared with fingerprint recognition and face recognition, the iris recognition technology is a biological recognition mode with very high safety at present, and the safety is only second to DNA detection.

In practical application, most iris recognition devices are fixed at a certain height, and because the vertical field angle of the iris lens is small, users with high or low heights often need to adjust the acquisition height by themselves to realize iris image acquisition (for example, the users with high heights need to bend over and the users with low heights need to stand on their feet), so that not only is the user experience poor, but also the image quality is not in accordance with the requirements, and the recognition passing probability is reduced.

At present, a USB interface becomes a global universal device interface, and no matter windows, linux, Android and other mainstream platforms support by default, a USB plug-and-play mode can better realize the quick access of the device, so that a multi-mode biological identification module using a USB device form is a trend of biological identification device development.

However, at present, a multi-modal biometric module is manufactured into a USB device form and is connected with a windows, linux, Android and other upper computers through a USB as a lower computer, and the problems of high transmission compression rate of the lower computer, high requirement on the upper computer and low recognition accuracy of the upper computer generally exist, so that the multi-modal biometric module in the form of a USB interface does not exist in the prior art.

SUMMERY OF THE UTILITY MODEL

In order to solve the problems of high transmission compression ratio of a lower computer, high requirement on an upper computer and low identification accuracy of the upper computer in the prior art, the utility model provides the multi-mode biological identification module, which reduces the data volume transmitted through a USB, has low image compression degree, reduces the consumption of upper computer resources, improves the identification accuracy, and has simple implementation mode and low cost.

The technical scheme provided by the utility model is as follows:

a multimode biological recognition module comprises an FPGA chip with an image preprocessing AI module and a USB interface chip, wherein:

the USB interface chip is connected with the FPGA chip and is used as a communication interface with an upper computer and a main power supply interface of the multi-mode biological recognition module;

the FPGA chip is connected with a face lens, an iris lens, a near-infrared light supplement lamp, a motor and a distance sensor, and the iris lens and the face lens are driven to rotate by the motor.

Furthermore, the FPGA chip is respectively connected with the face lens and the iris lens through a first digital processing chip and a second digital processing chip.

Furthermore, the FPGA chip is connected with the near-infrared light supplementing lamp through a near-infrared light supplementing lamp driving circuit, and the second digital processing chip is connected with the near-infrared light supplementing lamp driving circuit.

Furthermore, the near-infrared light supplementing lamps are multiple groups, each group of near-infrared light supplementing lamps is connected with the FPGA chip through a near-infrared light supplementing lamp driving circuit, and the multiple groups of near-infrared light supplementing lamp driving circuits are connected with the second digital processing chip.

Furthermore, the multi-modal biological recognition module further comprises an ambient light supplementary lighting lamp, and the FPGA chip is connected with the ambient light supplementary lighting lamp through an ambient light supplementary lighting circuit.

Further, the multi-modal biological recognition module further comprises a loudspeaker, a magnetic coding chip and/or a plurality of status indicator lamps, the FPGA chip is respectively connected with the motor and the loudspeaker through a motor driving circuit and an audio decoding driving chip, and the FPGA chip is respectively connected with the magnetic coding chip and the plurality of status indicator lamps.

Furthermore, the multi-modal biological recognition module further comprises an auxiliary power supply interface, and the auxiliary power supply interface is connected with the FPGA chip.

Furthermore, when the FPGA chip does not detect that the auxiliary power supply interface supplies power, the FPGA chip controls the motor and the near-infrared light supplementing lamp to work in a mutual exclusion manner, and when the FPGA chip detects that the auxiliary power supply interface supplies power, the FPGA chip controls the motor and the near-infrared light supplementing lamp to work simultaneously.

Further, a power supply circuit is arranged between the main power supply interface and the auxiliary power supply interface and the power input interface of the multi-modal biological recognition module, and the power supply circuit comprises a diode, a voltage stabilizing diode, a capacitor, a first PMOS (P-channel metal oxide semiconductor) tube, a third PMOS tube, a first resistor and a fifth resistor, wherein:

the main power supply interface is grounded through a fifth resistor, the auxiliary power supply interface is grounded through a third resistor and a fourth resistor which are sequentially connected in series, the main power supply interface is connected with the anode of the diode, and the cathode of the diode is connected with the power supply input interface of the multi-mode biological recognition module;

the main power supply interface is connected with the S pole of the second PMOS tube, the negative pole of the diode is connected with the S pole of the third PMOS tube, the D poles of the second PMOS tube and the third PMOS tube are connected with the G pole of the first PMOS tube through a first resistor, and the G poles of the second PMOS tube and the third PMOS tube are connected between a third resistor and a fourth resistor;

the S pole of the first PMOS tube is connected with the power input interface of the multi-modal biological recognition module, the D pole of the first PMOS tube is connected with the auxiliary power supply interface, and the G pole of the first PMOS tube is grounded through a second resistor;

the power input interface of the multi-modal biological recognition module is grounded through a voltage stabilizing diode and a capacitor which are connected in parallel, wherein the anode of the voltage stabilizing diode is connected with the power input interface of the multi-modal biological recognition module, and the cathode of the voltage stabilizing diode is grounded.

The utility model provides a multimode biological identification equipment, includes host computer and aforementioned multimode biological identification module, wherein multimode biological identification module's USB interface chip with the host computer is connected.

The utility model has the following beneficial effects:

according to the utility model, the acquired images in the form of video streams are preprocessed by the FPGA chip with the image preprocessing AI module, the images meeting the requirements are screened out and then uploaded to the upper computer, and each image received by the upper computer is a relatively clear image. On one hand, because the image which does not meet the requirement is removed, the data volume transmitted through the USB interface chip is reduced, the degree of image compression is reduced as much as possible on the premise of the USB2.0 transmission bandwidth requirement, and the loss of image details is reduced as much as possible. On the other hand, dependence of the multi-modal biometric identification module on an upper computer and consumption of upper computer resources are greatly reduced, and hardware configuration requirements on the upper computer are reduced, and the hardware configuration requirements of the upper computer are much lower than those of a common dual-mode face lens or an iris lens, so that the overall equipment cost is further reduced; and because the upper computer and the multi-modal biological recognition module are not required to perform frequent instruction operation, the stability and the adaptability of the equipment are improved.

According to the position of the positioned human eyes, a control motor in the multi-mode biological recognition module automatically rotates the human face lens and the iris lens to the proper positions for acquiring iris images, so that the system can be adapted to users with different heights independently, the cooperation of the users is not needed, and the user experience is good.

The collected images and the corresponding distance information are uploaded to the upper computer, and each image can correspondingly bind the identification of the current distance information, so that the upper computer can adjust the recognition algorithm according to the distance in a self-adaptive mode, the recognition accuracy is improved, and the use experience is greatly increased.

The USB interface chip and the FPGA chip are arranged into two independent chips, so that the realization mode is simple and the cost is low.

Drawings

FIG. 1 is a schematic diagram of a multimodal biometric module of the present invention;

FIG. 2 is a schematic diagram of near-infrared fill light and its driving method;

fig. 3 is a schematic diagram of a power supply circuit.

Detailed Description

In order to make the technical problems, technical solutions and advantages of the present invention more apparent, the following detailed description is given with reference to the accompanying drawings and specific embodiments.

The iris identification device aims to solve the problem that the iris identification device cannot be adaptive to the height of a user due to the fact that the vertical field angle of an iris lens is small. The iris recognition device is provided with a face lens and an iris lens (namely a multi-mode biological recognition module), wherein on one hand, a processor in the iris recognition device can detect the face position in a face image, the angle of the face lens and the iris lens needing to rotate is calculated by utilizing the pixel difference value between the detected face position coordinate and a preset reference coordinate, the face lens and the iris lens are rotated to the optimal collection position according to the rotation angle, the height of a user can be automatically judged, and the user experience is improved; on the other hand, the human face color preview image can be displayed on the screen of the iris recognition equipment for man-machine interaction; on the other hand, the face color image can be combined with the collected iris image to perform multi-modal biological recognition.

The multimode biological identification module is made into a USB device shape and is used as a lower computer to be connected with upper computers such as windows, linux, Android and the like through a USB, and the following problems exist:

1. the multimode biological recognition module is used as a lower computer and is connected with upper computers such as windows, linux and Android through a USB, all active control of the multimode biological recognition module is initiated by the upper computers, and the dependence on the upper computers is high.

The lower computer directly sends all the acquired iris images and face images to the upper computer in a video streaming mode, and the mode has a large burden on the processing of the USB bandwidth, cables and the application of the upper computer.

The theoretical bandwidth of the USB2.0 is 480Mbps, and video streams of two paths of lenses of an iris and a human face are compressed in a certain proportion to meet the bandwidth requirement of the USB 2.0; image compression causes image minutiae loss (the current conclusion is that the image compression ratio is in direct proportion to the loss rate of the image minutiae), and the accuracy of biological identification is influenced.

2. The lower computer has no autonomous processing capability, all data processing is performed on the upper computer, and in order to realize the height judgment, the face recognition process and the iris recognition process of the user, the upper computer needs to consume a large amount of memory and resources for calculation, so that the upper computer which is configured in the market in an older mode is not friendly, and the situation of system or application program jamming and collapse is easily caused.

3. Because of the limitation of current mainstream suppliers, the iris lens mostly adopts a fixed focus lens, the identification range is limited by the depth of field range, and the farther the human body is away from the focus, the more fuzzy the picture is. The existing multi-modal biological recognition module cannot judge the position relation between the acquisition position of the iris image and the focus, so that an upper computer only adopts the same algorithm to process, and the recognition accuracy is low.

In order to solve the above problems, the present invention provides a multi-modal biometric module, and the specific embodiments thereof are as follows.

Example 1:

the embodiment of the present invention provides a multi-modal biometric module 100, as shown in fig. 1 to 3, which includes an FPGA chip 1 with an image preprocessing AI module and a USB interface chip 2, wherein:

the USB interface chip 2 is connected with the FPGA chip 1, and the USB interface chip 2 is used as a communication interface with the upper computer 200 and a main power supply interface of the multi-modal biometric identification module 100.

The FPGA chip 1 is connected with a face lens 3, an iris lens 4, a near-infrared light supplement lamp 5, a motor 6 and a distance sensor 7, and the iris lens 4 and the face lens 3 are driven to rotate through the motor 6.

The FPGA chip with the image preprocessing AI module is used as a main chip of the multi-mode biological recognition module, so that the multi-mode biological recognition module has simple image preprocessing capability. The human face image and the iris image in the form of video streams collected by the human face lens and the iris lens are simply preprocessed by the image preprocessing AI module, wherein the preprocessing comprises height judgment, image definition judgment, simple human face positioning or human eye positioning and the like. After the preprocessing, clear images meeting the requirements are transmitted to a USB interface chip and output to an upper computer through the USB interface chip, and the images uploaded by the USB interface chip are all clear images after screening.

In the utility model, if the FPGA chip integrates the USB2.0 interface output, the realization mode is complex and the cost is high, so the USB interface chip and the FPGA chip are arranged into two independent chips, the FPGA chip is a main processing chip and is used for mounting all functional blocks, the USB interface chip is an external communication interface, and the USB interface chip and the FPGA chip are connected through a DVP protocol.

In one embodiment, the raw data output by the FPGA chip can be mjpg compressed and then uploaded to an upper computer through a USB2.0 interface, and the upper computer downloads instructions to the multimodal biological recognition module 100 in the same way.

Whether an object exists in front of the multi-modal biometric identification module can be identified through the distance sensor, and the accurate distance (error +/-3 cm) between each object and the multi-modal biometric identification module can be measured. After the distance sensor detects the object, whether the object is a human body model is judged according to the image preprocessing AI module. And after the human body model is determined, carrying out human face detection and positioning the human eye position according to the human body model. The distance from human eyes to the multi-mode biological recognition module can be calculated according to the pythagorean theorem by measuring the accurate distance between the human body and the multi-mode biological recognition module and the rotation angle of the current motor.

After the position of the human eyes is located, the motor is controlled to automatically rotate the human face lens and the iris lens to proper positions for collecting iris images, and the human face image and the iris images are collected. The face image, the iris image and the distance information of the face image and the iris image are uploaded to the upper computer together, so that the upper computer can adaptively adjust the recognition algorithm according to the distance (for example, the algorithm parameters are adjusted according to the distance or different algorithms are used for different distances), the recognition accuracy is improved, and the use experience is greatly improved.

The FPGA chip is connected with the face lens and the iris lens, if the FPGA chip is directly connected behind image sensors (sensors) of the face lens and the iris lens, the effect debugging of parameters of the face lens and the iris lens needs to directly call ip core of isp of the FPGA, and the method is complex in implementation mode and high in cost.

In order to solve the above problem, the FPGA chip 1 is connected to the face lens 3 and the iris lens 4 through a first digital processing chip (first DSP)8 and a second digital processing chip 9 (second DSP), respectively.

The sensors of the face lens and the iris lens are respectively connected to a digital processing chip (DSP), the sensors are debugged on the DSP, and the DSP outputs raw data to the FPGA.

As an improvement of the embodiment of the present invention, as shown in fig. 2, the FPGA chip 1 is connected to the near-infrared light supplement lamp 5 through a near-infrared light supplement lamp driving circuit 10, and the second digital processing chip 9 is connected to the near-infrared light supplement lamp driving circuit 10.

Furthermore, the near-infrared light supplement lamps 5 can be in multiple groups, each group of near-infrared light supplement lamps 5 is connected with the FPGA chip 1 through a near-infrared light supplement lamp driving circuit 10, and the multiple groups of near-infrared light supplement lamp driving circuits 10 are connected with a DSP (namely, the second digital processing chip 9).

Taking two groups of near-infrared light supplement lamps shown in fig. 2 as an example, two groups of near-infrared light supplement lamps (D1 and D2 are one group, and D3 and D4 are one group) are arranged on the left side and the right side of the iris lens, the two groups of near-infrared light supplement lamps are respectively controlled by 2 IO ports (IO1 and IO2) of the FPGA chip, and the exposure control is performed by a second digital processing chip of the iris camera, the hardware schematic diagram is shown in fig. 2, and the working process is as follows:

the FPGA chip controls the on and off of the left and right groups of near-infrared light supplementing lamps through IO1 and IO2 respectively, the problem of imaging light spots of iris images of people wearing glasses can be reduced as much as possible through the rapid on and off of the left and right groups of near-infrared light supplementing lamps, and in addition, PWM signals output by the second digital processing chip are overturned along with the frame rate of the PWM signals, so that the exposure time can be controlled accurately. According to the distance information fed back by the distance sensor, the FPGA chip adjusts the duty ratio of the PWM signal to enable light supplement to be more intelligent.

The duty ratio adjusting method of the PWM signal is as follows:

when the distance information is too small (namely smaller than the focal distance, for example, the distance is 30-40 cm), the duty ratio of the PWM signal is adjusted to be low, the intensity of near infrared light is reduced, and overexposure is avoided.

When the distance information is moderate (namely near the focus distance, for example, the distance is 40-70 cm), the duty ratio of the PWM signal is adjusted to be moderate.

When the distance information is too large (namely, the distance is larger than the focal distance, for example, the distance is 70-80 cm), the duty ratio of the PWM signal is increased, the intensity of near infrared light is increased, and the light supplement for the face is increased.

In order to adapt to different working environments, the multi-modal biological recognition module further comprises an ambient light supplementary lighting lamp 12, the FPGA chip 1 is connected with the ambient light supplementary lighting lamp 12 through an ambient light supplementary lighting circuit 11, light is supplemented through the ambient light supplementary lighting lamp when ambient light is dark, and the ambient light supplementary lighting lamp is generally a white light lamp.

The FPGA chip 1 is connected with the motor 6 through a motor driving circuit 13, and the enabling of the motor driving circuit is controlled by the FPGA chip.

In order to realize richer functions, the multi-modal biological recognition module can further comprise a loudspeaker, a magnetic coding chip and/or a plurality of status indicator lamps. The FPGA chip 1 is connected with a loudspeaker 15 through an audio decoding driving chip 14, and the FPGA chip 1 is respectively connected with a magnetic coding chip 16 and a plurality of status indicator lamps 17.

As an improvement of the embodiment of the present invention, the multi-modal biometric identification module may further include an auxiliary power supply interface 18(V _ ASS), and the auxiliary power supply interface 18 is connected to the FPGA chip 1. The auxiliary power supply interface 18 is used for externally connecting an auxiliary power supply to supply power for the multimode biological identification module in an auxiliary manner. The IO port of the FPGA chip 1 is connected to the auxiliary power supply interface V _ ASS through a configuration resistor, and can detect whether the auxiliary power supply interface 18 is inserted into the auxiliary power supply.

When the FPGA chip 1 does not detect that the auxiliary power supply interface 18 supplies power, the power is supplied only through the main power supply interface (namely the USB interface chip 2), and the FPGA chip 1 controls the motor 6 and the near-infrared light supplement lamp 5 to work in a mutually exclusive mode.

The power supply capacity of the standard USB2.0 interface is only 5V/500mA, and if no external power supply is provided and only the power supply is provided through the USB interface chip 2, the USB equipment needs to pay special attention that the peak current does not exceed the power supply capacity of the USB2.0 interface. However, in the multi-modal biometric identification module, the power consumption of the FPGA chip, the iris lens, the face lens, the near-infrared light supplement lamp, the stepping motor and the like is high, and the USB current overload is easily caused.

Therefore, in order to prevent the USB current overload, when the FPGA chip 1 does not detect that the auxiliary power supply interface 18 supplies power, and only supplies power through the main power supply interface, the FPGA chip 1 controls the motor 6 and the near-infrared light supplement lamp 5 to perform mutual exclusion operation.

Under the condition that the USB interface chip supplies power but no other auxiliary power, the FPGA chip 1 controls the enabling of the motor driving circuit and the near-infrared light supplementing lamp driving circuit in sequence. The method comprises the steps of firstly, turning on a motor driving circuit to enable, controlling the motor to rotate by using electricity, stopping the motor driving circuit immediately to enable when the human face lens and the iris lens are detected to rotate to proper positions, and turning on a near-infrared light supplementing lamp driving circuit to enable to supplement light. Therefore, the iris image can be captured only after a delay of about 20-50 ms.

The utility model designs independent power-on control circuits for each power-using component, and mutually excludes power-on of the motor and power-on of near-infrared light supplement and the like, thereby avoiding USB current overload from the source.

When the FPGA chip 1 detects that the auxiliary power supply interface 18(V _ ASS) supplies power, the FPGA chip 1 controls the motor 6 and the near-infrared light supplement lamp 5 to work simultaneously.

Whether the motor rotates and whether near-infrared light filling lamp light filling starts simultaneously depends on whether the multimode biological identification module is externally connected with auxiliary power supply, if the FPGA chip detects that auxiliary power supply has been inserted, the FPGA chip can control the motor and the near-infrared light filling lamp to be simultaneously opened, so that the iris lens can be opened when the FPGA chip rotates, and iris images can be rapidly collected after the lens rotates to a proper position.

In order to realize the main power supply and the auxiliary power supply of the multi-modal biometric identification module, a power supply circuit is arranged between a main power supply interface V _ USB (namely, a power supply provided by the USB interface chip 2) and an auxiliary power supply interface 18(V _ ASS) and a power supply input interface VSYS _5V of the multi-modal biometric identification module 100 (which provides power supply input for the multi-modal biometric identification module).

As shown in fig. 3, the power supply circuit includes a diode D1, a zener diode D2, a capacitor C, first to third PMOS transistors Q1, Q2, Q3, Q1, Q2, Q3 of the same type, first to fifth resistors R1, R2, R3, R4, R5, wherein:

the main power supply interface V _ USB is grounded through a fifth resistor R5, the auxiliary power supply interface V _ ASS is grounded through a third resistor R3 and a fourth resistor R4 which are sequentially connected in series, the main power supply interface V _ USB is connected with the anode of a diode D1, and the cathode of the diode D1 is connected with a power input interface VSYS _5V of the multi-mode biological recognition module.

The main power supply interface V _ USB is connected with the S pole of a second PMOS tube Q2, the negative pole of a diode D1 is connected with the S pole of a third PMOS tube Q3, the D poles of the second PMOS tube Q2 and the third PMOS tube Q3 are connected with the G pole of a first PMOS tube Q1 through a first resistor R1, and the G poles of the second PMOS tube Q2 and the third PMOS tube Q3 are connected between a third resistor R3 and a fourth resistor R4;

the S pole of the first PMOS tube Q1 is connected with a power input interface VSYS _5V of the multi-modal biometric identification module, the D pole of the first PMOS tube Q1 is connected with an auxiliary power supply interface V _ ASS, and the G pole of the first PMOS tube Q1 is grounded through a second resistor R2.

The power input interface VSYS _5V of the multi-modal biological recognition module is grounded through a voltage stabilizing diode D2 and a capacitor C which are connected in parallel, wherein the anode of the voltage stabilizing diode D2 is connected with the power input interface VSYS _5V of the multi-modal biological recognition module, and the cathode of the voltage stabilizing diode is grounded.

When only the main power supply interface V _ USB supplies power, the V _ USB outputs 4.3-4.4V voltage at VSYS _5V after passing through D1, Vgs is smaller than a conduction threshold value for Q2 and can be conducted, Vgs is also smaller than the conduction threshold value for Q3 and can be conducted, then the V _ USB passes through S, D electrodes of Q2 and Q3 in sequence to reach VSYS _5V, and the PMOS conduction voltage drop can be ignored, so that the VSYS _5V voltage can be pulled to 5V, and power is supplied to the multi-mode biological identification module.

When only the auxiliary power supply interface V _ ASS supplies power, the S voltage of the Q1 is about 4.4V due to the existence of the body diode of the Q1, Vgs is smaller than a conduction threshold value, Q1 is conducted, VSYS _5V is enabled to output 5V, and power is supplied to the multi-mode biological recognition module.

Similarly, when the main power supply interface V _ USB and the main power supply and auxiliary power supply interface V _ ASS supply power simultaneously, the resistors R1, R2, R3 and R4 are arranged, so that the Q1, the Q2 and the Q3 are simultaneously turned on, and due to the existence of the D2, the voltage of any interface of the main power supply interface V _ USB and the auxiliary power supply interface V _ ASS is higher, so that the voltage can be limited within the acceptable range of the back-sink voltage, and a safe voltage is provided for the multimodal biological identification module.

In summary, the present invention has the following advantages:

1. the multimode biological identification module with automatic height rotation, multi-angle near-infrared light supplement and iris and face double lenses is realized under the power supply of a single USB.

2. The image preprocessing AI module of the FPGA is utilized to carry out image preprocessing on the multi-mode biological recognition module, so that the data volume of USB transmission is reduced, the compression rate of image transmission is reduced, and the loss of image details is reduced; the dependence of the multi-modal biological recognition module on the upper computer and the consumption of upper computer resources are reduced, the configuration requirement on the upper computer is reduced, and the cost is reduced.

3. The motor rotates the self-adaptation height user of face camera lens iris camera lens, need not user's cooperation, and user experience is good.

4. The acquired images and the distance information thereof are uploaded to the upper computer, and the upper computer adjusts the recognition algorithm in a self-adaptive mode according to the distance, so that the recognition accuracy is improved.

5. The USB interface chip and the FPGA chip are arranged into two independent chips, so that the realization mode is simple and the cost is low.

Example 2:

the embodiment of the utility model provides a multi-modal biological recognition device, which comprises an upper computer 200 and a multi-modal biological recognition module 100 described in the embodiment 1, wherein a USB interface chip 2 of the multi-modal biological recognition module 100 is connected with the upper computer 200, as shown in FIG. 1.

For a brief description, the corresponding contents in embodiment 1 can be referred to for the sake of brevity, and details are not repeated herein.

While the foregoing is directed to the preferred embodiment of the present invention, it will be understood by those skilled in the art that various changes and modifications may be made without departing from the spirit and scope of the utility model as defined in the appended claims.

Claims

1. The multimode biological recognition module is characterized by comprising an FPGA chip with an image preprocessing AI module and a USB interface chip, wherein:

2. The multimodal biometric module of claim 1, wherein the FPGA chip is connected to the face lens and the iris lens via a first digital processing chip and a second digital processing chip, respectively.

3. The multi-modal biometric identification module of claim 2, wherein the FPGA chip is connected to the near-infrared fill light via a near-infrared fill light driving circuit, and the second digital processing chip is connected to the near-infrared fill light driving circuit.

4. The multi-modal biological recognition module of claim 3, wherein the plurality of groups of near-infrared light supplement lamps are connected to the FPGA chip through a near-infrared light supplement lamp driving circuit, and the plurality of groups of near-infrared light supplement lamp driving circuits are connected to the second digital processing chip.

5. The module of claim 1, further comprising an ambient light fill-in light, wherein the FPGA chip is connected to the ambient light fill-in light via an ambient light fill-in light circuit.

6. The module of claim 5, further comprising a speaker, a magnetic encoding chip and/or a plurality of status indicators, wherein the FPGA chip is connected to the motor and the speaker via a motor driving circuit and an audio decoding driving chip, respectively, and the FPGA chip is connected to the magnetic encoding chip and the plurality of status indicators, respectively.

7. The module of any of claims 1-6, further comprising an auxiliary power interface, wherein the auxiliary power interface is coupled to the FPGA chip.

8. The module of claim 7, wherein the FPGA chip controls the motor and the near-infrared fill light to operate in a mutually exclusive manner when the FPGA chip does not detect the power supply from the auxiliary power supply interface, and controls the motor and the near-infrared fill light to operate simultaneously when the FPGA chip detects the power supply from the auxiliary power supply interface.

9. The module of claim 8, wherein a power supply circuit is disposed between the main power supply interface and the auxiliary power supply interface and the power input interface of the module, and the power supply circuit comprises a diode, a zener diode, a capacitor, a first PMOS transistor to a third PMOS transistor, and a first resistor to a fifth resistor, wherein:

10. A multi-modal biometric device, comprising an upper computer and the multi-modal biometric module of any one of claims 1 to 9, wherein a USB interface chip of the multi-modal biometric module is connected to the upper computer.