WO2016192555A1 - 一种智能生物特征识别系统及识别方法 - Google Patents

一种智能生物特征识别系统及识别方法 Download PDF

Info

Publication number
WO2016192555A1
WO2016192555A1 PCT/CN2016/083328 CN2016083328W WO2016192555A1 WO 2016192555 A1 WO2016192555 A1 WO 2016192555A1 CN 2016083328 W CN2016083328 W CN 2016083328W WO 2016192555 A1 WO2016192555 A1 WO 2016192555A1
Authority
WO
WIPO (PCT)
Prior art keywords
zoom
assembly
biometric
component
information
Prior art date
Application number
PCT/CN2016/083328
Other languages
English (en)
French (fr)
Inventor
张宏鑫
Original Assignee
聚鑫智能科技（武汉）股份有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 聚鑫智能科技（武汉）股份有限公司 filed Critical 聚鑫智能科技（武汉）股份有限公司
Publication of WO2016192555A1 publication Critical patent/WO2016192555A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
    • G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
    • G06V40/18—Eye characteristics, e.g. of the iris
    • G06V40/193—Preprocessing; Feature extraction
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
    • G06V40/70—Multimodal biometrics, e.g. combining information from different biometric modalities

Definitions

  • the invention belongs to the field of intelligent security control, and in particular relates to an intelligent biometric recognition system and a recognition method.
  • biometric identification devices on the market have relatively simple identification methods. For example, face recognition devices cannot simultaneously realize iris recognition, or can not simultaneously realize voiceprint recognition and artificial intelligence voice communication.
  • the conventional zoom camera is large in size and cannot be in a limited space because it needs to adjust the distance of multiple sets of zoom lenses during the zooming process.
  • the traditional zoom camera needs to adjust the distance of multiple sets of zoom lenses during the zooming process, so the fast zoom cannot be achieved. If the face and the human eye are recognized at the same time, two slow focuss will occur, in actual use. It is very inconvenient to need people to cooperate and wait; although the fixed focus camera can achieve a small volume, it needs a good cooperation in the identification process to clearly and accurately identify, if it is not well matched, Identify errors.
  • the image acquisition camera of most biological visual recognition devices on the market is a fixed position, but according to the height of the person, if the distance is too high, it will be blocked by the body, if it is too short, it will not be captured, so it cannot be automatically adapted. Different people's height, the facial features are collected correctly.
  • An intelligent biometric recognition system comprising: a biometrics acquisition unit and an integrated circuit portion;
  • the integrated circuit part includes:
  • Microcontroller Controls the operation of the entire intelligent biometric system
  • Storage system used to store all data that needs to be stored, including the baseline data required for authentication;
  • Biological sign monitoring component used for real-time tracking of the moving state of an object having a biological sign within the range detected by the intelligent biometric system;
  • DSP image processor when the biological sign monitoring component detects that someone passes, starts the DSP image processor to perform image processing and turns on the biometric collecting unit to perform iris recognition on the human eye, and stereoscopically recognize the face; the DSP image processor will The identified information is processed and transmitted to the microcontroller for comparison with the reference data for identity verification;
  • Power management system used to power the entire intelligent biometric system
  • the optical part of the biometric collecting unit comprises a periscope zooming component and a MEMS zooming component
  • the mechanical part of the biometric collecting unit comprises a human eye-like high-speed zoom machine vision device
  • the periscope zoom component has a focal length Adjustment
  • MEMS zoom components are set in the periscope zoom assembly, including telephoto zoom lens assembly and MEMS fast zoom lens assembly
  • human eye high speed zoom machine vision device for long distance zoom and MEMS fast of telephoto zoom lens assembly Quick zoom of the zoom lens assembly.
  • the human eye high speed zoom machine vision device comprises a movable outer cover, a longitudinal drive assembly, a movable inner cover, a lateral drive assembly and first and second transmission systems
  • the longitudinal drive assembly drives the movable outer cover around the longitudinal axis through the first transmission system
  • the lateral drive assembly drives the movable inner cover to rotate axially about the transverse axis through the second transmission system
  • the movable inner cover is disposed inside the movable outer cover
  • the longitudinal axis intersects the transverse axis.
  • the optical portion of the biometric collection unit further includes: a coated high-definition lens disposed in front of the periscope zoom component; and an infrared high-definition photosensitive chip disposed behind the periscope zoom component;
  • the photosensitive chip sends the imaging information to the biometric identification chip through the line tube and the signal and the power supply line for the identification operation;
  • the driving component control chip controls the longitudinal driving component and the lateral driving component to perform a plurality of degrees of freedom rotation of the human eye while automatically locking Biometrics are sent to the DSP image processor one by one.
  • the MEMS zoom assembly further includes a beam splitter disposed on the same optical axis and a polarizer disposed at 45°; the telephoto zoom lens assembly and the MEMS fast zoom lens assembly are disposed in the beam splitter and the polarizer
  • the MEMS zoom assembly further includes a glazing mirror positioned above the polarizer and at an angle of 45° to the polarizer.
  • the MEMS rapid zoom lens assembly comprises two plates of different thicknesses, one end surface being a curved surface and the other end surface being a plane, and the planar end faces of the two plates are opposite and have a gap, and when two plates are relatively moved, the focus occurs. The change thus achieves optical zooming by refraction.
  • the periscope zoom component comprises: a total reflection prism that converts vertical light into parallel rays and a zoom optical portion disposed at the output end of the total reflection prism, thereby achieving up and down adjustment of the focal length.
  • the intelligent biometric identification system further includes a fill light component and/or a smart dimming component, wherein
  • the fill light component is connected to the DSP image processor, and fills the light when it detects that the current environment needs to fill light;
  • the intelligent dimming component is connected to the DSP image processor for digital fill or wide dynamic range imaging under the control of the microcontroller.
  • the intelligent biometric recognition system further includes a voiceprint recognition module for recognizing the voice of the user.
  • the microcontroller is an MCU microcontroller
  • the biological sign monitoring component is a radar monitoring component
  • the integrated circuit part also includes:
  • a fill light component connected to the DSP image processor to fill light when it is detected that the current environment needs to fill light;
  • a smart dimming component connected to the DSP image processor.
  • the infrared high-definition photo sensor Before performing biometric identification, the infrared high-definition photo sensor first monitors whether the living environment of the living body determines whether it is in a visible light or non-visible environment and performs backlighting according to the current environment. Compensating or starting the fill light component for adding and subtracting light compensation; the intelligent dimming component comprises an HDR unit and a light-increasing and dimming unit, and before performing biometric recognition, the infrared high-definition photosensitive chip first monitors whether the living environment of the living body determines whether it is in visible light or not The non-visible environment is selected according to the current environment:
  • Treatment 1 When the environment is in visible light environment, the infrared high-definition sensor detects whether it needs to fill light in real time, and whether it is a backlight environment. If it needs to fill light, the infrared high-definition sensor will inform the MCU of the current environment of the microcontroller. The fill light intensity, the MCU microcontroller controls the intelligent dimming component to perform digital fill light to enable the biometric feature to be clearly collected; when the environment is in the backlight environment, the infrared high-definition light sensor informs the MCU microcontroller of the information, and the MCU microcontroller controls the smart tone.
  • the optical component turns on HDR for wide dynamic range imaging;
  • Treatment 2 When the environment is in a non-visible environment, the infrared high-definition sensor informs the MCU microcontroller of the information, and the MCU microcontroller controls the intelligent dimming component and the fill-light unit to increase or decrease the illuminance in real time to enable the biometric feature. Clear collection;
  • a signal output device the MCU microcontroller compares the information processed by the DSP processor with the information in the storage system. When correct, the signal output device controls the external device to be turned on; when the error occurs, no operation is performed or according to the user. Set to perform the expansion operation;
  • An intelligent video intercom system a biometrics collection unit including real-time camera and a voiceprint recognition module, the smart video intercom system being connected with the MCU microcontroller and the communication module, the information output module, and the storage system, and real-time Full-duplex for local or remote audio and video interaction and storage;
  • a voice pattern recognition module for identifying the user's voice and confirming whether it is consistent, and sending the result to the MCU microcontroller, and the MCU microcontroller determines whether to activate the open device according to the comparison result;
  • the voiceprint recognition module includes a microphone, a recording circuit, and a sound analysis processing component;
  • the MCU microcontroller retrieves audio information corresponding to the biological sign in the storage system after the biomarker is matched for sound matching; when the sound recognition is performed for the first time, the recording circuit stores the identified audio information to Storage system;
  • a communication module connected to the MCU microcontroller, being a wired and wireless network module;
  • a security system the security system is connected with the MCU microcontroller; the other Internet of Things monitoring devices can be matched, and the successfully paired device information is sent to the MCU microcontroller for security or alarm;
  • a shock sensor set in the security system, when receiving hammer, flaw and damage, will send information to the MCU microcontroller in time; the MCU microcontroller sends information to the user through the communication module or set other expansion operations And simultaneously activate the alarm device in the security system;
  • An information output module connected to the MCU microcontroller for displaying real-time video and processing information
  • An embedded background processing system connected to the MCU microcontroller for accepting information of the MCU microcontroller, and displaying the human-machine interactive system interface on the video display screen through the information output module; the embedded background processing system The information is pushed to the cloud platform through the communication module to facilitate access by the user terminal, and the signal output device and the security system can be directly pushed to the terminal device preset by the user through the cloud platform during the alarm;
  • the human eye high-speed zoom machine vision device includes an outer cover, a longitudinal drive assembly that drives the outer cover to be axially rotatable about the longitudinal axis, an inner cover, a lateral drive assembly that drives the inner cover to be axially rotatable about the transverse axis, and a a biometric monitoring assembly transmission system in the inner cover and a driving assembly control chip for controlling the longitudinal driving assembly and the lateral driving assembly; the biometric monitoring assembly transmission system is coupled to the optical portion of the biometric collecting unit;
  • the biometric monitoring assembly transmission system comprises an outer frame and a transparent outer cover; the transparent outer cover is hemispherical, the outer frame is open at one end, and the transparent outer cover and the outer frame form a closed main body casing having a cavity; the biometric collecting unit optical Partially disposed in the enclosed body casing; the lateral drive assembly and the longitudinal drive assembly are each a micro-motor; the outer cover or the inner cover is driven to rotate by the bearing and the drive rod, respectively;
  • the optical portion of the biometric collection unit further includes:
  • the drive component control chip controls the longitudinal drive component and the lateral drive component to perform a plurality of degrees of freedom rotation of the human eye while automatically locking the biometrics and transmitting them to the DSP image processor one by one, the DSP image processor processing and inputting into the MCU a high speed zoom assembly comprising the periscope zoom assembly, and the MEMS zoom assembly disposed within a periscope zoom assembly; the MEMS zoom assembly including a center set on a same optical axis a splitter, a second set of telephoto zoom lens assemblies, a first set of MEMS fast zoom lens assemblies, and a polarizer disposed at 45° to change the optical path; the polarizer is at an angle of 45° to the lens; MEMS The fast zoom lens assembly comprises: two
  • the periscope zoom assembly includes: a total reflection prism that converts vertical light into parallel rays, and a zoom optical portion disposed at an output end of the total reflection prism, thereby changing the original front and rear adjustment focal length to an up and down adjustment;
  • the zoom range is required to be short, especially when iris recognition is performed, the telephoto zoom lens assembly does not work, and only the MEMS fast zoom lens assembly works to quickly achieve high-speed zoom; when telephoto zoom is required, Especially when performing stereo modeling of the face and later face recognition, the telephoto zoom lens assembly works, and the MEMS fast zoom lens assembly does not work, enabling zooming operation over a large distance range, and the structure can be maximized. In the case of a scaled compression structure thickness, fast zoom and long distance zoom are achieved.
  • a method for identifying an intelligent biometric recognition system comprising:
  • a reference data collection step when first used, the biometric collection unit collects biometric features of the user and stores the data as reference data in the storage system;
  • Verification data collection step when the biological sign monitoring component detects that someone passes, start the DSP image processor to perform image processing and turn on the biometrics collecting unit to perform biometric data collection;
  • Verification step The DSP image processor processes the acquired data and transmits it to the microcontroller for comparison with the reference data for authentication.
  • the invention adopts a human eye-like high-speed zoom machine vision device, which can automatically adapt different heights, distances, and fast locks of a person's face, and performs high-speed zoom adjustment of the image to an optimal state without special cooperation. Face and eye features can be acquired correctly and quickly.
  • the intelligent dimming device when encountering a poor lighting environment, it will not be able to identify and identify errors. For example, in a backlighting environment, a strong light environment, an overexposed exposure environment, and a light and dark light handover environment, the intelligent dimming device will The picture acquired by the complex light source is adjusted to the optimal state, or the fill light component is activated to increase or decrease the light compensation according to the environment, thereby improving the recognition rate.
  • the three-dimensional face recognition is realized, and the double iris recognition and voiceprint recognition are used to make the biometric recognition result more secure and reliable, and the artificial intelligence voice communication; the embedded background processing system pushes the information to the cloud through the communication module.
  • the platform is convenient for user terminal access, and can realize other network surveillance camera screen access, other network sensor access, etc., the vision system realizes high-definition low-stream narrow-band video communication and video recording, remote information recording and backup.
  • FIG. 1 is a schematic structural view of an intelligent biometric identification system of the present invention.
  • FIG. 2 is a schematic structural view of a transmission system of the intelligent biometric identification system of the present invention.
  • FIG. 3 is a schematic structural view of an integrated circuit portion of an intelligent biometric identification system according to the present invention.
  • FIG. 4 is a schematic structural view of a MEMS zoom assembly in the present invention.
  • FIG. 5 is a schematic view showing the working principle of the MEMS rapid zoom lens group in the present invention.
  • FIG. 6 is a schematic diagram of a state in which a face is subjected to double iris recognition and three-dimensional face recognition according to the present invention.
  • the intelligent biometric identification system proposed by the present invention is shown in FIGS. 1 to 3. All modules of the intelligent biometric identification system are disposed on a main board, and the main board includes: a biometric collecting unit and an integrated circuit part; and a biometric collecting unit. Including mechanical parts and optical parts;
  • the integrated circuit part includes:
  • An MCU microcontroller controlling the operation of the entire intelligent biometric system, those skilled in the art know that the MCU microcontroller can also employ other types of microcontrollers;
  • a storage system for storing all data that needs to be stored; when first used, the intelligent biometric system collects facial features of the user and builds a stereo face model and stores it in the storage system as a stereo face reference information; intelligent biometrics The identification system collects user voiceprint information and stores it as voiceprint reference information in the storage system; the intelligent biometric recognition system collects the user's double iris information and stores it as iris reference information in the storage system.
  • a radar monitoring component for tracking the moving state of an object having a biological sign in a range detected by the intelligent biometric system in real time; those skilled in the art know that the radar monitoring component can also adopt other types of biological sign monitoring components;
  • a DSP image processor when the radar monitoring component detects that someone passes, starts the DSP image processor to perform image processing and starts the biometrics collecting unit, performs iris recognition on the human eye, and stereoscopically recognizes the human face; The information is input into the MCU microcontroller; the first time the biometric recognition is performed, the DSP image processor stores the identified biometric information (including the stereo face and the double iris) as biometric reference information to the storage system;
  • a fill light component connected to the DSP image processor to fill light when it is detected that the current environment needs to fill light;
  • a smart dimming component coupled to a DSP image processor, the smart dimming component comprising an HDR unit and a dimming unit.
  • the infrared high-definition photo sensor Before performing biometric identification, the infrared high-definition photo sensor first monitors whether the environment in which the living body is located determines whether it is in a visible or non-visible environment, and performs corresponding processing according to the current environment, for example, backlight compensation or activation of the dimming unit. Dimming compensation. Before performing biometric identification, the infrared high-definition sensor first monitors the environment in which the living being is located, determines whether it is in the visible and non-visible environment, and performs the following processing according to the current environment:
  • Treatment 1 When the environment is in visible light environment, the infrared high-definition sensor detects whether it needs to fill light in real time, and whether it is a backlight environment. If it needs to fill light, the infrared high-definition sensor will inform the MCU of the current environment of the microcontroller. The fill light intensity, the MCU microcontroller controls the intelligent dimming component to digitally fill the light through the addition and subtraction of the light unit to enable the biometric feature to be clearly collected; when the environment is a backlight environment, the infrared high-definition light sensor informs the MCU microcontroller, MCU micro The controller controls the intelligent dimming component to turn on the HDR unit for wide dynamic range imaging;
  • Treatment 2 When the environment is in a non-visible environment, the infrared high-definition sensor informs the MCU microcontroller of the information, and the MCU microcontroller controls the addition and subtraction unit of the intelligent dimming component to increase or decrease the illuminance in real time to make the optical device Can be clearly collected;
  • a signal output device the MCU microcontroller compares the information processed by the DSP image processor with the stereo face reference information stored in the storage system. When correct, the MCU microcontroller controls the external device to be turned on by the signal output device. Or the MCU microcontroller performs subsequent verification operations; when an error occurs, no operation is performed or an expansion operation is performed according to user settings;
  • An intelligent video intercom system a biometrics collection unit including real-time camera and a voiceprint recognition module, the smart video intercom system being connected with the MCU microcontroller and the communication module, the information output module, and the storage system, and real-time Full-duplex for local or remote audio and video interaction and storage;
  • the voiceprint recognition module is configured to identify the user voice and confirm whether the voiceprint reference information stored in the storage system is met, and send the comparison result to the MCU microcontroller, and the MCU microcontroller determines whether to open the external connection according to the comparison result.
  • the voiceprint recognition module comprises a microphone, a recording circuit and a sound analysis processing component; after the biometric matching of the three-dimensional human face, the MCU microcontroller combines the voiceprint information collected by the voiceprint recognition module with the voiceprint reference in the storage system.
  • Information (audio information) is used for sound matching; when the voice recognition is performed for the first time, the recording circuit stores the recognized audio information as voiceprint reference information to the storage system;
  • a communication module connected to the MCU microcontroller, being a wired and wireless network module;
  • a security system the security system is connected with the MCU microcontroller; the other Internet of Things monitoring devices can be matched, and the successfully paired device information is sent to the MCU microcontroller for security or alarm;
  • a shock sensor set in the security system, when receiving hammer, flaw and damage, will send information to the MCU microcontroller in time; the MCU microcontroller sends information to the user through the communication module or set other expansion operations And simultaneously activate the alarm device in the security system;
  • An information output module connected to the MCU microcontroller for displaying real-time video and processing information
  • An embedded background processing system connected to the MCU microcontroller for accepting information of the MCU microcontroller, and displaying the human-machine interactive system interface on the video display screen through the information output module; the embedded background processing system Pass through The letter module pushes the information to the cloud platform for convenient access by the user terminal, and the signal output device and the security system can be directly pushed to the terminal device preset by the user through the cloud platform when the alarm is alarmed;
  • a power management system capable of supplying power to a rechargeable battery, DC power, or AC for powering the entire system.
  • Figure 1 The mechanical part of the biometric collection unit is shown in Figure 1, which includes:
  • a human eye high speed zoom machine vision device comprising a housing 10, a longitudinal drive assembly for driving the housing to be axially rotatable about a longitudinal axis, an inner cover 8, and a lateral drive assembly for driving the inner cover to be axially rotatable about a transverse axis a biometric monitoring assembly transmission system disposed within the inner cover and a drive assembly control chip for controlling the longitudinal drive assembly and the lateral drive assembly through the overall control cable 13; the biometric acquisition unit optical portion and the biometric monitoring total Connected to the drive system;
  • the biometric monitoring assembly transmission system includes an outer frame 12 and a transparent outer cover 9; the transparent outer cover is hemispherical, the outer frame is open at one end, and the transparent outer cover and the outer frame form a closed main body casing having a cavity through the mounting plate 14.
  • the optical component of the biometric collecting unit is disposed in the closed body casing; the lateral driving component and the longitudinal driving component are both a micro high speed motor 1; respectively, the outer cover or the inner cover is rotated by the bearing 2 and the driving rod.
  • Two high-speed motors drive two movable covers to rotate left and right and up and down, so that the rotation of the human eye can be realistically simulated.
  • the optical component of the biometric collection unit includes:
  • the imaging information is sent to the biometric chip on the integrated circuit board 11 through the line tube 7 and the signal and power supply line 15;
  • the coated high-definition lens 6 is a wide-angle lens;
  • the driving component control chip controls the longitudinal driving component and the lateral driving component to perform human eye-like
  • the rotation of multiple degrees of freedom simultaneously automatically locks the biometrics and sends them to the DSP image processor one by one, and the DSP image processor processes and inputs the MCU microcontroller;
  • the high speed zoom assembly includes a periscope zoom assembly and a MEMS zoom assembly disposed within the periscope zoom assembly;
  • the MEMS zoom assembly includes a splitter, a telephoto zoom lens assembly, a MEMS fast zoom lens assembly, and a polarizer disposed at 45° to change the optical path, which are disposed on the same optical axis;
  • the MEMS zoom assembly further includes a glazing mirror located above the polarizer and at an angle of 45° to the polarizer;
  • the MEMS rapid zoom lens assembly comprises: two plates of different thicknesses and one end surface being curved, the planar end faces of the two plates are opposite and have a gap, and when the two plates move relative to each other, the focus changes to realize optical zoom by refraction;
  • the periscope zoom assembly includes: a total reflection prism that converts vertical light into parallel rays, and a zoom optical portion disposed at an output end of the total reflection prism, thereby changing the original front and rear adjustment focal length to an up and down adjustment;
  • the zoom range is required to be short, especially when iris recognition is performed, the telephoto zoom lens assembly does not work, and only the MEMS fast zoom lens assembly works to quickly achieve high-speed zoom; when telephoto zoom is required, Especially when performing stereo modeling of the face and later face recognition, the telephoto zoom lens assembly works, and the MEMS fast zoom lens assembly does not work, enabling zooming operation over a large distance range, and the structure can be maximized. In the case of a scaled compression structure thickness, fast zoom and long distance zoom are achieved.
  • the specific mechanical structure of the biometric collecting unit in the present invention a micro high speed motor 1, a movable bearing 2, a zoom lens 3, a high definition infrared imaging chip 4, and a high speed variable assembly 5 , coated high-definition lens 6, line tube 7, inner hemisphere movable cover 8, transparent plexiglass cover 9, outer hemisphere cover 10, integrated circuit board 11, outer casing 12, total control cable 13, mounting plate 14, signal and power supply line 15
  • Two high-speed motors control the two movable covers to rotate left and right and up and down, so that the rotation of the human eye can be realistically simulated.
  • the user When using for the first time, the user needs to stand in front of the biometrics collection unit to collect the reference information; the user needs to swing the head to the left and right to enable the human eye biometrics collection unit to collect the facial features of the user and build the stereoscopic face model and store it;
  • the infrared high-definition photo sensor first monitors the environment in which the living body is located to determine whether it is in the visible light and non-visible environment and performs the following processing according to the current environment:
  • Treatment 1 When the environment is in visible light environment, the infrared high-definition sensor detects whether it needs to fill light in real time, and whether it is a backlight environment. If it needs to fill light, the infrared high-definition sensor will inform the MCU of the current environment of the microcontroller. The fill light intensity, the MCU microcontroller controls the intelligent dimming component to digitally fill the light through the addition and subtraction of the light unit to enable the biometric feature to be clearly collected; when the environment is a backlight environment, the infrared high-definition light sensor informs the MCU microcontroller, MCU micro The controller controls the intelligent dimming component to turn on the HDR unit for wide dynamic range imaging;
  • Treatment 2 When the environment is in a non-visible environment, the infrared high-definition sensor informs the MCU microcontroller of the information, and the MCU microcontroller controls the addition and subtraction unit of the intelligent dimming component to increase or decrease the illuminance in real time to make the optical device Can be clearly collected;
  • the invention uses a high-speed zoom method, as shown in FIG. 4 to FIG. 5, combining the MEMS zoom component and the periscope zoom component; if the zoom distance is required to be short, especially for iris recognition.
  • the telephoto zoom lens assembly does not work (the long-distance iris recognition area in Figure 6), only the MEMS fast zoom lens assembly works to quickly achieve high-speed zoom; when telephoto zoom is required, especially for the face Stereoscopic modeling and later face recognition, the telephoto zoom lens assembly works, the MEMS fast zoom lens assembly does not work, (the stereo face recognition area in Figure 6), enabling zooming over a large distance range action.
  • the user also needs to perform identification recording storage of the iris
  • the user can also perform voiceprint acquisition and record storage through the voiceprint recognition module; the specific embodiments described herein are merely illustrative of the spirit of the present invention.
  • a person skilled in the art can make various modifications or additions to the specific embodiments described or in a similar manner, without departing from the spirit of the invention or as defined by the appended claims. The scope.

Landscapes

  • Engineering & Computer Science (AREA)
  • Human Computer Interaction (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Multimedia (AREA)
  • Theoretical Computer Science (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Health & Medical Sciences (AREA)
  • General Health & Medical Sciences (AREA)
  • Ophthalmology & Optometry (AREA)
  • Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
  • Image Input (AREA)

Abstract

一种智能生物特征识别系统，该智能生物特征识别系统包括：生物特征采集单元和集成电路部分；其中，集成电路部分包括微控制器、存储系统、生物体征监控组件、DSP图像处理器和电源管理系统；所述生物特征采集单元的光学部分包括潜望式变焦组件和MEMS变焦组件，所述生物特征采集单元的机械部分包括仿人眼高速变焦机器视觉装置；仿人眼高速变焦机器视觉装置用于实现长焦变焦镜片组件的长距离变焦和MEMS快速变焦镜片组件的快速变焦。该智能生物特征识别系统可以自动适配不同人身高、远近距离，快速锁定人的面部，进行高速变焦调整图像为最佳状态，无需特意配合就能正确快速的采集到面部及眼部特征。

Description

一种智能生物特征识别系统及识别方法 技术领域
本发明属于智能安全控制领域，尤其是涉及一种智能生物特征识别系统及识别方法。
背景技术
目前市面上大多数生物体识别设备识别方式比较单一，比如人脸识别设备就无法同时实现虹膜的识别，或者无法同时实现声纹识别及人工智能语音交流。
目前市面上的生物体视觉识别装置大多都是定焦摄像头或传统式变焦摄像机，传统式变焦摄像机由于在变焦过程中需要对多组变焦镜片的距离进行调整因此体积较大，无法在有限的空间中安装；另外传统式变焦摄像机在变焦过程中需要对多组变焦镜片的距离进行调整因此无法实现快速变焦，如果同时对人脸和人眼识别，就会出现两次缓慢对焦，在实际使用过程中需要人来配合并等待，非常不方便；定焦摄像机虽然可以做到很小的体积，但是在识别过程中需要人来很好的配合才能清晰和准确识别，如果稍微配合的不好就会识别错误。
目前市面上的生物体视觉识别装置遇到较差的采光环境时，就会无法识别和识别错误，比如在(逆光环境下、强光环境下、补光曝光过度的环境下、明暗光交接环境下)无法实现智能化的调光。
目前市面上大多数生物体视觉识别设备的图像采集摄像头为固定位置，但根据人的身高不同，近距离的话：太高就会被身体所遮挡，太矮则拍摄不到，所以无法自动适配不同人身高，正确采集到面部特征。
发明内容
本发明的上述技术问题主要是通过下述技术方案得以解决：
一种智能生物特征识别系统，该智能生物特征识别系统包括：生物特征采集单元和集成电路部分；
其中，集成电路部分包括：
微控制器：控制整个智能生物特征识别系统的的运作；
存储系统：用于存储所有需要存储的数据，包括身份验证所需的基准数据；
生物体征监控组件：用于实时跟踪智能生物特征识别系统所检测范围内具有生物体征的物体的移动状态；
DSP图像处理器：在生物体征监控组件检测到有人通过时，启动DSP图像处理器进行图像处理并开启生物特征采集单元对人眼进行虹膜识别，同时对人脸进行立体识别；DSP图像处理器将识别后的信息处理后传送到微控制器与基准数据进行比较以进行身份验证；
电源管理系统：用于为整个智能生物特征识别系统供电；
所述生物特征采集单元的光学部分包括潜望式变焦组件和MEMS变焦组件，所述生物特征采集单元的机械部分包括仿人眼高速变焦机器视觉装置；其中，潜望式变焦组件对焦距进行上下调整；MEMS变焦组件设置在潜望式变焦组件内，包括长焦变焦镜片组件和MEMS快速变焦镜片组件；仿人眼高速变焦机器视觉装置用于实现长焦变焦镜片组件的长距离变焦和MEMS快速变焦镜片组件的快速变焦。
进一步地，该仿人眼高速变焦机器视觉装置包括活动外罩、纵向驱动组件、活动内罩、横向驱动组件以及第一和第二传动系统，纵向驱动组件通过第一传动系统驱动活动外罩绕纵向轴轴向旋转，横向驱动组件通过第二传动系统驱动活动内罩绕横向轴轴向旋转，活动内罩设于活动外罩内部，且纵向轴与横向轴相交。
进一步地，所述生物特征采集单元的光学部分还包括：设置在所述潜望式变焦组件前方的镀膜高清镜片以及设置在所述潜望式变焦组件后方的红外高清感光芯片；所述红外高清感光芯片将成像信息通过线路管和信号及供电线路发送给生物识别芯片进行所述识别操作；驱动组件控制芯片控制纵向驱动组件和横向驱动组件进行仿人眼的多个自由度的旋转同时自动锁定生物特征并将其逐个发送给DSP图像处理器。
进一步地，所述MEMS变焦组件还包括中心在同一光轴上设置的分光器以及呈45°设置的偏光镜；所述长焦变焦镜片组件和MEMS快速变焦镜片组件设置在分光器和偏光镜之间；所述MEMS变焦组件还包括采光镜，采光镜位于所述偏光镜上方且与偏光镜呈45°角度。
进一步地，所述MEMS快速变焦镜片组件包括两个不同厚度的且一端面为曲面另一端面为平面的板块，两个板块的平面端面相对且有间隙，两个板块发生相对移动时，焦点发生变化从而通过折射实现光学变焦。
进一步地，所述潜望式变焦组件包括：将垂直光线变为平行光线的全反射棱镜以及设置在全反射棱镜输出端的变焦光学部分，从而实现焦距的上下调整。
进一步地，智能生物特征识别系统还包括补光组件和/或智能调光组件，其中，
补光组件与DSP图像处理器连接，在侦测到当前环境需要补光时进行补光；
智能调光组件与DSP图像处理器连接，在微控制器的控制下进行数码补光或进行宽动态范围成像。
进一步地，智能生物特征识别系统还包括声纹识别模块，用于对用户的声音进行识别。
进一步地，微控制器为MCU微控制器；生物体征监控组件为雷达监控组件；
集成电路部分还包括：
一补光组件：与DSP图像处理器连接，在侦测到当前环境需要补光时进行补光；
一智能调光组件：与DSP图像处理器连接，在进行生物特征识别前，红外高清感光芯片首先监测生物体所处环境判断是否处于可见光和非可见光环境并根据当前所处环境进行选择处理进行逆光补偿或启动补光组件进行增减光补偿；所述智能调光组件包括HDR单元以及增减光单元，在进行生物特征识别前，红外高清感光芯片首先监测生物体所处环境判断是否处于可见光和非可见光环境并根据当前所处环境进行选择处理：
处理一：当所处环境是可见光环境时，红外高清感光芯片实时侦测是否需要补光，以及是否为逆光环境，若需要补光，红外高清感光芯片将信息告知MCU微控制器当前所处环境的补光强度，MCU微控制器控制智能调光组件进行数码补光使生物特征能够清晰采集；当环境为逆光环境时，红外高清感光芯片将信息告知MCU微控制器，MCU微控制器控制智能调光组件开启HDR进行宽动态范围成像；
处理二：当所处环境是非可见光环境时，红外高清感光芯片将信息告知MCU微控制器，MCU微控制器控制智能调光组件和补光单元对补光设备进行实时增减光照度,使生物特征能够清晰采集；
一信号输出装置：MCU微控制器将经过DSP处理器进行处理后的信息与存储系统中的信息进行比对，正确时，信号输出装置控制外接设备开启；错误时，不进行任何操作或根据用户设定进行扩展操作；
一智能可视对讲系统：包括实时摄像的生物特征采集单元以及声纹识别模块，该智能可视对讲系统与所述MCU微控制器和通信模块、信息输出模块、存储系统连接，可实时全双工进行本地或异地音视频交互并存储；
一声纹识别模块：用于辨认用户声音并确认是否符合，并将结果发送给MCU微控制器，由MCU微控制器根据比对结果决定是否启动开启装置；声纹识别模块包括话筒、录音电路以及声音分析处理组件；MCU微控制器在生物体征匹配后调取存储系统中的与该生物体征对应的音频信息进行声音匹配；第一次进行声音识别时，录音电路将识别后的音频信息存储到存储 系统；
一通信模块：与MCU微控制器连接，为有线和无线网络模块；
一安防系统：所述安防系统与MCU微控制器连接；可匹配其它物联网监控设备，将配对成功的设备信息及时发送给MCU微控制器进行设防或报警；
一震动传感器：设置在安防系统中，在接收到锤、砸和破坏的时候，将及时发送信息给MCU微控制器；所述MCU微控制器通过通信模块发送信息给用户或设定其他扩展操作并同时启动安防系统中的报警装置；
一信息输出模块：与MCU微控制器连接，用于显示实时视频及处理信息；
一嵌入式后台处理系统：与MCU微控制器连接，用于接受MCU微控制器的信息，把人机交互式系统界面通过信息输出模块，显示在视频显示屏上；所述嵌入式后台处理系统透过通信模块将信息推送到云平台，方便用户终端访问，所述信号输出装置和安防系统报警时可直接通过云平台推送到用户预设的终端设备；
所述仿人眼高速变焦机器视觉装置包括一个外罩、驱动外罩能够绕纵向轴轴向旋转的纵向驱动组件、一个内罩、驱动内罩能够绕横向轴轴向旋转的横向驱动组件、一个设置在内罩内的生物特征监控总成传动系统以及一个用于控制纵向驱动组件和横向驱动组件的驱动组件控制芯片；所述生物特征监控总成传动系统与生物特征采集单元的光学部分连接；
所述生物特征监控总成传动系统包括一个外框和一个透明外罩；所述透明外罩为半球形，外框一端开口，透明外罩与外框组成具有空腔的封闭主体外壳；生物特征采集单元光学部分设置在封闭主体外壳内；横向驱动组件和纵向驱动组件均为一个微型电机；分别通过轴承和驱动杆驱动外罩或内罩旋转；
所述生物特征采集单元的光学部分还包括：
广角镜片、设置在镀膜高清镜片后方的用于采集视频信息的高速变焦总成、以及设置在高速变焦总成后方的红外高清感光芯片；所述红外高清感光芯片将成像信息发送给生物识别芯片；驱动组件控制芯片控制纵向驱动组件和横向驱动组件进行仿人眼的多个自由度的旋转同时自动锁定生物特征并将其逐个发送给DSP图像处理器，所述DSP图像处理器进行处理后输入MCU微控制器；所述高速变焦总成包括所述潜望式变焦组件，以及设置在潜望式变焦组件内的所述MEMS变焦组件；所述MEMS变焦组件包括中心在同一光轴上的依次设置的分光器、第二组长焦变焦镜片组件、第一组MEMS快速变焦镜片组件、以及呈45°设置的用来改变光路的偏光镜；所述偏光镜上方与采光镜呈45°角度；MEMS快速变焦镜片组件包括：两个不同厚度的且一端面为曲面的板块，两个板块的平面端面相对且有间隙，两个板块发生相对移动 时，焦点变化从而通过折射实现光学变焦；
所述潜望式变焦组件包括：一个将垂直光线变为平行光线的全反射棱镜，以及设置在全反射棱镜输出端的变焦光学部分，从而把原来前后调整焦距变成上下调整；
若是在需要变焦范围短的情况，特别是在进行虹膜识别的时候，长焦变焦镜片组件不工作，仅MEMS快速变焦镜片组件进行工作，能够快速实现高速变焦的工作；在需要长焦变焦时，特别是进行脸部的立体建模以及以后进行脸部识别的时候，长焦变焦镜片组件进行工作，MEMS快速变焦镜片组件不工作，能够实现较大距离范围内的变焦动作，该结构能够在最大尺度的压缩结构厚度的情况下，实现快速变焦和长距离变焦。
一种如前任一项所述的智能生物特征识别系统的识别方法，该方法包括：
基准数据采集步骤：首次使用时，所述生物特征采集单元对用户的生物特征进行采集，并作为基准数据存储在存储系统中；
验证数据采集步骤：在生物体征监控组件检测到有人通过时，启动DSP图像处理器进行图像处理并开启生物特征采集单元进行生物特征数据采集；
验证步骤：DSP图像处理器将采集的数据进行处理后传送到微控制器与基准数据进行比较以进行身份验证。
本发明的有益效果：
首先，在机械和光学部分，本发明采用仿人眼高速变焦机器视觉装置，可以自动适配不同人身高、远近距离，快速锁定人的面部，进行高速变焦调整图像为最佳状态，无需特意配合就能正确快速的采集到面部及眼部特征。
其次，遇到较差的采光环境时，就会无法识别和识别错误，比如在(逆光环境下、强光环境下、补光曝光过度的环境下、明暗光交接环境下)智能调光装置将复杂光源采集到的画面调整到最佳状态，或启动补光组件根据环境需要进行增减光补偿，提高识别率。
另外，在系统结构上，实现了立体人脸识别并配合双虹膜识别和声纹识别使生物体识别结果更安全可靠，人工智能语音交流；嵌入式后台处理系统透过通信模块将信息推送到云平台，方便用户终端访问，并能实现其它网络监控摄像头画面接入、其它网络传感器接入等，视觉系统实现高清晰低码流的窄带视频通信和视频录制，远程信息记录和备份。
附图说明
图1为本发明的智能生物特征识别系统的结构示意图。
图2为本发明中智能生物特征识别系统的传动系统的结构示意图。
图3为本发明中智能生物特征识别系统的集成电路部分的结构示意图。
图4为本发明中的MEMS变焦组件结构示意图。
图5为本发明中的MEMS快速变焦镜片组工作原理示意图。
图6为本发明中人脸进行双虹膜识别和立体人脸识别的状态示意图。
具体实施方式
下面通过实施例，并结合附图，对本发明的技术方案作进一步具体的说明。
一、首先介绍一下本发明的硬件结构：
本发明提出的智能生物特征识别系统如图1至3所示，该智能生物特征识别系统的所有模块均设置在一个主板上，该主板包括：生物特征采集单元和集成电路部分；生物特征采集单元包括机械部分和光学部分；
其中，集成电路部分包括：
一MCU微控制器：控制整个智能生物特征识别系统的的运作，本领域技术人员知晓该MCU微控制器也可以采用其他类型的微控制器；
一存储系统：用于存储所有需要存储的数据；首次使用时，智能生物特征识别系统采集用户脸部特征并建立立体人脸模型后存储在存储系统中，作为立体人脸基准信息；智能生物特征识别系统采集用户声纹信息并作为声纹基准信息存储在存储系统中；智能生物特征识别系统采集用户双虹膜信息并作为虹膜基准信息存储在存储系统中。
一雷达监控组件：用于实时跟踪智能生物特征识别系统所检测范围内具有生物体征的物体的移动状态；本领域技术人员知晓该雷达监控组件也可以采用其他类型的生物体征监控组件；
一DSP图像处理器：在雷达监控组件检测到有人通过时，启动DSP图像处理器进行图像处理并开启生物特征采集单元，对人眼进行虹膜识别，同时对人脸进行立体识别；并将识别后的信息输入MCU微控制器；第一次进行生物特征识别时，DSP图像处理器将识别后的生物特征信息(包括立体人脸、双虹膜)作为生物特征基准信息存储到存储系统；
一补光组件：与DSP图像处理器连接，在侦测到当前环境需要补光时进行补光；
一智能调光组件：与DSP图像处理器连接，所述智能调光组件包括HDR单元以及增减光单元。在进行生物特征识别前，红外高清感光芯片首先监测生物体所处环境判断是否处于可见光和非可见光环境，并根据当前所处环境进行相应的处理，例如，逆光补偿或启动增减光单元进行增减光补偿。在进行生物特征识别前，红外高清感光芯片首先监测生物体所处环境，判断是否处于可见光和非可见光环境并根据当前环境进行以下处理：
处理一：当所处环境是可见光环境时，红外高清感光芯片实时侦测是否需要补光，以及是否为逆光环境，若需要补光，红外高清感光芯片将信息告知MCU微控制器当前所处环境的补光强度，MCU微控制器控制智能调光组件通过增减光单元进行数码补光使生物特征能够清晰采集；当环境为逆光环境时，红外高清感光芯片将信息告知MCU微控制器，MCU微控制器控制智能调光组件开启HDR单元进行宽动态范围成像；
处理二：当所处环境是非可见光环境时，红外高清感光芯片将信息告知MCU微控制器，MCU微控制器控制智能调光组件的增减光单元对补光设备进行实时增减光照度，使生物特征能够清晰采集；
一信号输出装置：MCU微控制器将经过DSP图像处理器进行处理后的信息与存储系统中存储的立体人脸基准信息进行比对，正确时，MCU微控制器通过信号输出装置控制外接设备开启或MCU微控制器进行后续的验证操作；错误时，不进行任何操作或根据用户设定进行扩展操作；
一智能可视对讲系统：包括实时摄像的生物特征采集单元以及声纹识别模块，该智能可视对讲系统与所述MCU微控制器和通信模块、信息输出模块、存储系统连接，可实时全双工进行本地或异地音视频交互并存储；
其中，声纹识别模块用于辨认用户声音并确认是否符合存储系统中存储的声纹基准信息，并将比对结果发送给MCU微控制器，由MCU微控制器根据比对结果决定是否开启外接设备。优选地，声纹识别模块包括话筒、录音电路以及声音分析处理组件；MCU微控制器在立体人脸的生物特征匹配后，将声纹识别模块采集的声纹信息与存储系统中的声纹基准信息(音频信息)进行声音匹配；第一次进行声音识别时，录音电路将识别后的音频信息作为声纹基准信息存储到存储系统；
一通信模块：与MCU微控制器连接，为有线和无线网络模块；
一安防系统：所述安防系统与MCU微控制器连接；可匹配其它物联网监控设备，将配对成功的设备信息及时发送给MCU微控制器进行设防或报警；
一震动传感器：设置在安防系统中，在接收到锤、砸和破坏的时候，将及时发送信息给MCU微控制器；所述MCU微控制器通过通信模块发送信息给用户或设定其他扩展操作并同时启动安防系统中的报警装置；
一信息输出模块：与MCU微控制器连接，用于显示实时视频及处理信息；
一嵌入式后台处理系统：与MCU微控制器连接，用于接受MCU微控制器的信息，把人机交互式系统界面通过信息输出模块，显示在视频显示屏上；所述嵌入式后台处理系统透过通 信模块将信息推送到云平台，方便用户终端访问，所述信号输出装置和安防系统报警时可直接通过云平台推送到用户预设的终端设备；
一电源管理系统：能够为可充电蓄电池、直流供电或交流供电，用于给整个系统进行供电。
生物特征采集单元机械部分如图1所示，该部分包括：
一个仿人眼高速变焦机器视觉装置，该装置包括一个外罩10、驱动外罩能够绕纵向轴轴向旋转的纵向驱动组件、一个内罩8、驱动内罩能够绕横向轴轴向旋转的横向驱动组件、一个设置在内罩内的生物特征监控总成传动系统以及一个用于通过总控制排线13控制纵向驱动组件和横向驱动组件的驱动组件控制芯片；生物特征采集单元光学部分与生物特征监控总成传动系统连接；
所述生物特征监控总成传动系统包括一个外框12和一个透明外罩9；所述透明外罩为半球形，外框一端开口，透明外罩与外框通过安装板14组成具有空腔的封闭主体外壳；生物特征采集单元光学部分设置在封闭主体外壳内；横向驱动组件和纵向驱动组件均为一个微型高速电机1；分别通过轴承2和驱动杆驱动外罩或内罩旋转。两个高速电机分别驱动两个活动罩进行左右和上下旋转，从而能够逼真的模拟人眼的转动。
所述生物特征采集单元光学部分包括：
镀膜高清镜片6、设置在镀膜高清镜片6后方的用于采集视频信息的高速变焦总成5、以及设置在高速变焦总成后方的变焦镜片3和红外高清感光芯片4；所述红外高清感光芯片将成像信息通过线路管7和信号及供电线路15发送给集成电路板11上的生物识别芯片；镀膜高清镜片6为广角镜片；驱动组件控制芯片控制纵向驱动组件和横向驱动组件进行仿人眼的多个自由度的旋转同时自动锁定生物特征并将其逐个发送给DSP图像处理器，所述DSP图像处理器进行处理后输入MCU微控制器；
所述高速变焦总成包括一个潜望式变焦组件，以及设置在潜望式变焦组件内的MEMS变焦组件；
所述MEMS变焦组件如图4所示，包括中心在同一光轴上的依次设置的分光器、长焦变焦镜片组件、MEMS快速变焦镜片组件以及呈45°设置的用来改变光路的偏光镜；所述MEMS变焦组件还包括采光镜，采光镜位于所述偏光镜上方且与偏光镜呈45°角度；
MEMS快速变焦镜片组件包括：两个不同厚度的且一端面为曲面的板块，两个板块的平面端面相对且有间隙，两个板块发生相对移动时，焦点发生变化从而通过折射实现光学变焦；
所述潜望式变焦组件包括：一个将垂直光线变为平行光线的全反射棱镜，以及设置在全反射棱镜输出端的变焦光学部分，从而把原来前后调整焦距变成上下调整；
若是在需要变焦范围短的情况，特别是在进行虹膜识别的时候，长焦变焦镜片组件不工作，仅MEMS快速变焦镜片组件进行工作，能够快速实现高速变焦的工作；在需要长焦变焦时，特别是进行脸部的立体建模以及以后进行脸部识别的时候，长焦变焦镜片组件进行工作，MEMS快速变焦镜片组件不工作，能够实现较大距离范围内的变焦动作，该结构能够在最大尺度的压缩结构厚度的情况下，实现快速变焦和长距离变焦。
如图1至图2所示，为本发明中的所述生物特征采集单元的具体机械结构，微型高速电机1、活动轴承2、变焦镜片3、高清感红外成像芯片4、高速变总成5、镀膜高清镜片6、线路管7、内半球活动罩8、透明有机玻璃罩9、外半球罩10、集成电路板11、外壳12、总控制排线13、安装板14、信号及供电线路15；两个高速电机分别控制两个活动罩进行左右和上下旋转，从而能够逼真的模拟人眼的转动。
二、下面介绍一下本发明的工作方法：
生物特征识别过程：
首次使用时，用户需要站在生物特征采集单元前进行基准信息采集；用户需要适当的左右摆动头部使仿人眼生物特征采集单元能够采集用户脸部特征并建立立体人脸模型后存储起来；
如图3所示，进行生物特征识别前，需要进行光环境侦测；红外高清感光芯片首先监测生物体所处环境判断是否处于可见光和非可见光环境并根据当前环境进行以下处理：
处理一：当所处环境是可见光环境时，红外高清感光芯片实时侦测是否需要补光，以及是否为逆光环境，若需要补光，红外高清感光芯片将信息告知MCU微控制器当前所处环境的补光强度，MCU微控制器控制智能调光组件通过增减光单元进行数码补光使生物特征能够清晰采集；当环境为逆光环境时，红外高清感光芯片将信息告知MCU微控制器，MCU微控制器控制智能调光组件开启HDR单元进行宽动态范围成像；
处理二：当所处环境是非可见光环境时，红外高清感光芯片将信息告知MCU微控制器，MCU微控制器控制智能调光组件的增减光单元对补光设备进行实时增减光照度，使生物特征能够清晰采集；
本发明所使用的是一种高速变焦方法，如图4至图5所示，将MEMS变焦组件和潜望式变焦组件结合在一起；若需要变焦距离短的时候，特别是在进行虹膜识别的时候，长焦变焦镜片组件不工作(图6中的远距离虹膜识别区域)，仅MEMS快速变焦镜片组件进行工作，能够快速实现高速变焦的工作；在需要长焦变焦时，特别是进行脸部的立体建模以及以后进行脸部识别的时候，长焦变焦镜片组件进行工作，MEMS快速变焦镜片组件不工作，(图6中的立体人脸识别区域)，能够实现较大距离范围内的变焦动作。
另外，用户也需要进行虹膜的识别记录存储；
最后，用户也能够通过声纹识别模块进行声纹采集并记录存储；本文中所描述的具体实施例仅仅是对本发明精神作举例说明。本发明所属技术领域的技术人员可以对所描述的具体实施例做各种各样的修改或补充或采用类似的方式替代，但并不会偏离本发明的精神或者超越所附权利要求书所定义的范围。
尽管本文较多地使用了微型高速电机1、活动轴承2、变焦镜片3、高清感红外成像芯片4、高速变总成5、镀膜高清镜片6、线路管7、内半球活动罩8、透明有机玻璃罩9、外半球罩10、集成电路板11、外壳12、总控制排线13、安装板14、信号及供电线路15等术语，但并不排除使用其它术语的可能性。使用这些术语仅仅是为了更方便地描述和解释本发明的本质；把它们解释成任何一种附加的限制都是与本发明精神相违背的。

Claims (10)

  1. 一种智能生物特征识别系统，其特征在于，该智能生物特征识别系统包括：生物特征采集单元和集成电路部分；
    其中，集成电路部分包括：
    微控制器：控制整个智能生物特征识别系统的的运作；
    存储系统：用于存储所有需要存储的数据，包括身份验证所需的基准数据；
    生物体征监控组件：用于实时跟踪智能生物特征识别系统所检测范围内具有生物体征的物体的移动状态；
    DSP图像处理器：在生物体征监控组件检测到有人通过时，启动DSP图像处理器进行图像处理并开启生物特征采集单元对人眼进行虹膜识别，同时对人脸进行立体识别；DSP图像处理器将识别后的信息处理后传送到微控制器与基准数据进行比较以进行身份验证；
    电源管理系统：用于为整个智能生物特征识别系统供电；
    所述生物特征采集单元的光学部分包括潜望式变焦组件和MEMS变焦组件，所述生物特征采集单元的机械部分包括仿人眼高速变焦机器视觉装置；其中，潜望式变焦组件对焦距进行上下调整；MEMS变焦组件设置在潜望式变焦组件内，包括长焦变焦镜片组件和MEMS快速变焦镜片组件；仿人眼高速变焦机器视觉装置用于实现长焦变焦镜片组件的长距离变焦和MEMS快速变焦镜片组件的快速变焦。
  2. 根据权利要求1所述的智能生物特征识别系统，其特征在于，该仿人眼高速变焦机器视觉装置包括活动外罩、纵向驱动组件、活动内罩、横向驱动组件以及第一和第二传动系统，纵向驱动组件通过第一传动系统驱动活动外罩绕纵向轴轴向旋转，横向驱动组件通过第二传动系统驱动活动内罩绕横向轴轴向旋转，活动内罩设于活动外罩内部，且纵向轴与横向轴相交。
  3. 根据权利要求2所述的智能生物特征识别系统，其特征在于，所述生物特征采集单元的光学部分还包括：设置在所述潜望式变焦组件前方的镀膜高清镜片以及设置在所述潜望式变焦组件后方的红外高清感光芯片；所述红外高清感光芯片将成像信息通过线路管和信号及供电线路发送给生物识别芯片进行所述识别操作；驱动组件控制芯片控制纵向驱动组件和横向驱动组件进行仿人眼的多个自由度的旋转同时自动锁定生物特征并将其逐个发送给DSP图像处理器。
  4. 根据权利要求1所述的智能生物特征识别系统，其特征在于，所述MEMS变焦组件还包括中心在同一光轴上设置的分光器以及呈45°设置的偏光镜；所述长焦变焦镜片组件和 MEMS快速变焦镜片组件设置在分光器和偏光镜之间；所述MEMS变焦组件还包括采光镜，采光镜位于所述偏光镜上方且与偏光镜呈45°角度。
  5. 根据权利要求1所述的智能生物特征识别系统，其特征在于，所述MEMS快速变焦镜片组件包括两个不同厚度的且一端面为曲面另一端面为平面的板块，两个板块的平面端面相对且有间隙，两个板块发生相对移动时，焦点发生变化从而通过折射实现光学变焦。
  6. 根据权利要求1所述的智能生物特征识别系统，其特征在于，所述潜望式变焦组件包括：将垂直光线变为平行光线的全反射棱镜以及设置在全反射棱镜输出端的变焦光学部分，从而实现焦距的上下调整。
  7. 根据权利要求1所述的智能生物特征识别系统，其特征在于，智能生物特征识别系统还包括补光组件和/或智能调光组件，其中，
    补光组件与DSP图像处理器连接，在侦测到当前环境需要补光时进行补光；
    智能调光组件与DSP图像处理器连接，在微控制器的控制下进行数码补光或进行宽动态范围成像。
  8. 根据权利要求1所述的智能生物特征识别系统，其特征在于，智能生物特征识别系统还包括声纹识别模块，用于对用户的声音进行识别。
  9. 根据权利要求1所述的智能生物特征识别系统，其特征在于，微控制器为MCU微控制器；生物体征监控组件为雷达监控组件；
    集成电路部分还包括：
    一补光组件：与DSP图像处理器连接，在侦测到当前环境需要补光时进行补光；
    一智能调光组件：与DSP图像处理器连接，在进行生物特征识别前，红外高清感光芯片首先监测生物体所处环境判断是否处于可见光和非可见光环境并根据当前所处环境进行选择处理进行逆光补偿或启动补光组件进行增减光补偿；所述智能调光组件包括HDR单元以及增减光单元，在进行生物特征识别前，红外高清感光芯片首先监测生物体所处环境判断是否处于可见光和非可见光环境并根据当前所处环境进行选择处理：
    处理一：当所处环境是可见光环境时，红外高清感光芯片实时侦测是否需要补光，以及是否为逆光环境，若需要补光，红外高清感光芯片将信息告知MCU微控制器当前所处环境的补光强度，MCU微控制器控制智能调光组件进行数码补光使生物特征能够清晰采集；当环境为逆光环境时，红外高清感光芯片将信息告知MCU微控制器，MCU微控制器控制智能调光组件开启HDR进行宽动态范围成像；
    处理二：当所处环境是非可见光环境时，红外高清感光芯片将信息告知MCU微控制器， MCU微控制器控制智能调光组件和补光单元对补光设备进行实时增减光照度,使生物特征能够清晰采集；
    一信号输出装置：MCU微控制器将经过DSP处理器进行处理后的信息与存储系统中的信息进行比对，正确时，信号输出装置控制外接设备开启；错误时，不进行任何操作或根据用户设定进行扩展操作；
    一智能可视对讲系统：包括实时摄像的生物特征采集单元以及声纹识别模块，该智能可视对讲系统与所述MCU微控制器和通信模块、信息输出模块、存储系统连接，可实时全双工进行本地或异地音视频交互并存储；
    一声纹识别模块：用于辨认用户声音并确认是否符合，并将结果发送给MCU微控制器，由MCU微控制器根据比对结果决定是否启动开启装置；声纹识别模块包括话筒、录音电路以及声音分析处理组件；MCU微控制器在生物体征匹配后调取存储系统中的与该生物体征对应的音频信息进行声音匹配；第一次进行声音识别时，录音电路将识别后的音频信息存储到存储系统；
    一通信模块：与MCU微控制器连接，为有线和无线网络模块；
    一安防系统：所述安防系统与MCU微控制器连接；可匹配其它物联网监控设备，将配对成功的设备信息及时发送给MCU微控制器进行设防或报警；
    一震动传感器：设置在安防系统中，在接收到锤、砸和破坏的时候，将及时发送信息给MCU微控制器；所述MCU微控制器通过通信模块发送信息给用户或设定其他扩展操作并同时启动安防系统中的报警装置；
    一信息输出模块：与MCU微控制器连接，用于显示实时视频及处理信息；
    一嵌入式后台处理系统：与MCU微控制器连接，用于接受MCU微控制器的信息，把人机交互式系统界面通过信息输出模块，显示在视频显示屏上；所述嵌入式后台处理系统透过通信模块将信息推送到云平台，方便用户终端访问，所述信号输出装置和安防系统报警时可直接通过云平台推送到用户预设的终端设备；
    所述仿人眼高速变焦机器视觉装置包括一个外罩、驱动外罩能够绕纵向轴轴向旋转的纵向驱动组件、一个内罩、驱动内罩能够绕横向轴轴向旋转的横向驱动组件、一个设置在内罩内的生物特征监控总成传动系统以及一个用于控制纵向驱动组件和横向驱动组件的驱动组件控制芯片；所述生物特征监控总成传动系统与生物特征采集单元的光学部分连接；
    所述生物特征监控总成传动系统包括一个外框和一个透明外罩；所述透明外罩为半球形，外框一端开口，透明外罩与外框组成具有空腔的封闭主体外壳；生物特征采集单元光学部分 设置在封闭主体外壳内；横向驱动组件和纵向驱动组件均为一个微型电机；分别通过轴承和驱动杆驱动外罩或内罩旋转；
    所述生物特征采集单元的光学部分还包括：
    广角镜片、设置在镀膜高清镜片后方的用于采集视频信息的高速变焦总成、以及设置在高速变焦总成后方的红外高清感光芯片；所述红外高清感光芯片将成像信息发送给生物识别芯片；驱动组件控制芯片控制纵向驱动组件和横向驱动组件进行仿人眼的多个自由度的旋转同时自动锁定生物特征并将其逐个发送给DSP图像处理器，所述DSP图像处理器进行处理后输入MCU微控制器；所述高速变焦总成包括所述潜望式变焦组件，以及设置在潜望式变焦组件内的所述MEMS变焦组件；所述MEMS变焦组件包括中心在同一光轴上的依次设置的分光器、第二组长焦变焦镜片组件、第一组MEMS快速变焦镜片组件、以及呈45°设置的用来改变光路的偏光镜；所述偏光镜上方与采光镜呈45°角度；MEMS快速变焦镜片组件包括：两个不同厚度的且一端面为曲面的板块，两个板块的平面端面相对且有间隙，两个板块发生相对移动时，焦点变化从而通过折射实现光学变焦；
    所述潜望式变焦组件包括：一个将垂直光线变为平行光线的全反射棱镜，以及设置在全反射棱镜输出端的变焦光学部分，从而把原来前后调整焦距变成上下调整；
    若是在需要变焦范围短的情况，特别是在进行虹膜识别的时候，长焦变焦镜片组件不工作，仅MEMS快速变焦镜片组件进行工作，能够快速实现高速变焦的工作；在需要长焦变焦时，特别是进行脸部的立体建模以及以后进行脸部识别的时候，长焦变焦镜片组件进行工作，MEMS快速变焦镜片组件不工作，能够实现较大距离范围内的变焦动作，该结构能够在最大尺度的压缩结构厚度的情况下，实现快速变焦和长距离变焦。
  10. 一种如权利要求1至9中任一项所述的智能生物特征识别系统的识别方法，其特征在于，该方法包括：
    基准数据采集步骤：首次使用时，所述生物特征采集单元对用户的生物特征进行采集，并作为基准数据存储在存储系统中；
    验证数据采集步骤：在生物体征监控组件检测到有人通过时，启动DSP图像处理器进行图像处理并开启生物特征采集单元进行生物特征数据采集；
    验证步骤：DSP图像处理器将采集的数据进行处理后传送到微控制器与基准数据进行比较以进行身份验证。
PCT/CN2016/083328 2015-05-29 2016-05-25 一种智能生物特征识别系统及识别方法 WO2016192555A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201510288145.0 2015-05-29
CN201510288145.0A CN104992141B (zh) 2015-05-29 2015-05-29 基于双虹膜、立体人脸和声纹识别的智能生物特征监控总成及监控方法

Publications (1)

Publication Number Publication Date
WO2016192555A1 true WO2016192555A1 (zh) 2016-12-08

Family

ID=54303954

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2016/083328 WO2016192555A1 (zh) 2015-05-29 2016-05-25 一种智能生物特征识别系统及识别方法

Country Status (2)

Country Link
CN (1) CN104992141B (zh)
WO (1) WO2016192555A1 (zh)

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106778715A (zh) * 2017-03-23 2017-05-31 合肥芯福传感器技术有限公司 相控阵生物识别信息采集装置及采集方法
CN107993339A (zh) * 2017-12-28 2018-05-04 联合华通（天津）科技有限公司 带有人脸识别功能的智能ic卡终端机
CN108725208A (zh) * 2018-08-14 2018-11-02 北京中润惠通科技发展有限公司 一种受电弓滑板检测监测装置
CN111768659A (zh) * 2020-05-15 2020-10-13 四川科华天府科技有限公司 一种基于ar互动教学设备的变焦系统
CN112052844A (zh) * 2020-08-31 2020-12-08 深圳市商汤科技有限公司 补光方法及装置、电子设备和存储介质

Families Citing this family (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104992141B (zh) * 2015-05-29 2017-02-22 聚鑫智能科技（武汉）股份有限公司 基于双虹膜、立体人脸和声纹识别的智能生物特征监控总成及监控方法
CN106407950A (zh) * 2016-09-30 2017-02-15 维沃移动通信有限公司 一种人脸信息识别方法及移动终端
CN107239770A (zh) * 2017-02-22 2017-10-10 全普光电科技(上海)有限公司 安全认证方法、安全认证系统及电子设备
CN107391983B (zh) 2017-03-31 2020-10-16 创新先进技术有限公司 一种基于物联网的信息处理方法及装置
CN109501724A (zh) * 2017-09-15 2019-03-22 南京志超汽车零部件有限公司 一种同时基于脸部识别和虹膜识别的车辆防盗系统
CN108492518A (zh) * 2018-03-01 2018-09-04 上海市保安服务总公司 智能安防系统
CN108492231A (zh) * 2018-05-22 2018-09-04 集美大学 基于学生行为数据的课堂效果信息的监控系统
CN110427740B (zh) * 2019-07-30 2021-06-15 深圳市智微智能软件开发有限公司 一种加密系统
CN114511916B (zh) * 2022-04-19 2022-06-17 亿慧云智能科技（深圳）股份有限公司 基于人脸识别的终端设备控制方法、控制装置及终端设备

Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1618079A (zh) * 2002-01-16 2005-05-18 虹膜技术公司 利用立体人脸识别的虹膜识别系统和方法
US20050270386A1 (en) * 2004-05-28 2005-12-08 Hirofumi Saitoh Method and apparatus for authentication utilizing iris
CN1932840A (zh) * 2005-09-16 2007-03-21 中国科学技术大学 基于虹膜和人脸的多模态生物特征身份识别系统
WO2009036103A1 (en) * 2007-09-10 2009-03-19 Retica Systems, Inc. Long distance multimodal biometric system and method
JP2010079609A (ja) * 2008-09-26 2010-04-08 Brother Ind Ltd 個人認証装置、個人認証プログラム、および個人認証装置を備えたインターホンシステム
CN101814130A (zh) * 2009-02-19 2010-08-25 中国科学院自动化研究所 用摄像机阵列的虹膜识别装置和多模态生物特征识别方法
CN103593647A (zh) * 2013-10-21 2014-02-19 王晓鹏 一种生物特征成像的方法与设备
CN104992141A (zh) * 2015-05-29 2015-10-21 张宏鑫 基于双虹膜、立体人脸和声纹识别的智能生物特征监控总成及监控方法

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2005258860A (ja) * 2004-03-12 2005-09-22 Matsushita Electric Ind Co Ltd 複数認証方法及びその装置
CN101432192A (zh) * 2006-03-10 2009-05-13 通用电气安全股份有限公司 集成验证和筛查系统
CN101520838A (zh) * 2008-02-27 2009-09-02 中国科学院自动化研究所 自动跟踪和自动变焦的虹膜图像获取方法
CN201628970U (zh) * 2010-04-20 2010-11-10 北京森博克智能科技有限公司 基于多生物特征和rfid的嵌入式多模态身份认证便携终端
CN202887216U (zh) * 2012-11-19 2013-04-17 成都芯软科技发展有限公司 一种基于网络数据传输的生物信息采集识别系统

Patent Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1618079A (zh) * 2002-01-16 2005-05-18 虹膜技术公司 利用立体人脸识别的虹膜识别系统和方法
US20050270386A1 (en) * 2004-05-28 2005-12-08 Hirofumi Saitoh Method and apparatus for authentication utilizing iris
CN1932840A (zh) * 2005-09-16 2007-03-21 中国科学技术大学 基于虹膜和人脸的多模态生物特征身份识别系统
WO2009036103A1 (en) * 2007-09-10 2009-03-19 Retica Systems, Inc. Long distance multimodal biometric system and method
JP2010079609A (ja) * 2008-09-26 2010-04-08 Brother Ind Ltd 個人認証装置、個人認証プログラム、および個人認証装置を備えたインターホンシステム
CN101814130A (zh) * 2009-02-19 2010-08-25 中国科学院自动化研究所 用摄像机阵列的虹膜识别装置和多模态生物特征识别方法
CN103593647A (zh) * 2013-10-21 2014-02-19 王晓鹏 一种生物特征成像的方法与设备
CN104992141A (zh) * 2015-05-29 2015-10-21 张宏鑫 基于双虹膜、立体人脸和声纹识别的智能生物特征监控总成及监控方法

Cited By (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106778715A (zh) * 2017-03-23 2017-05-31 合肥芯福传感器技术有限公司 相控阵生物识别信息采集装置及采集方法
CN107993339A (zh) * 2017-12-28 2018-05-04 联合华通（天津）科技有限公司 带有人脸识别功能的智能ic卡终端机
CN107993339B (zh) * 2017-12-28 2024-01-05 联合华通（天津）科技有限公司 带有人脸识别功能的智能ic卡终端机
CN108725208A (zh) * 2018-08-14 2018-11-02 北京中润惠通科技发展有限公司 一种受电弓滑板检测监测装置
CN108725208B (zh) * 2018-08-14 2023-04-07 北京中润惠通科技发展有限公司 一种受电弓滑板检测监测装置
CN111768659A (zh) * 2020-05-15 2020-10-13 四川科华天府科技有限公司 一种基于ar互动教学设备的变焦系统
CN111768659B (zh) * 2020-05-15 2022-08-30 四川科华天府科技有限公司 一种基于ar互动教学设备的变焦系统
CN112052844A (zh) * 2020-08-31 2020-12-08 深圳市商汤科技有限公司 补光方法及装置、电子设备和存储介质

Also Published As

Publication number Publication date
CN104992141B (zh) 2017-02-22
CN104992141A (zh) 2015-10-21

Similar Documents

Publication Publication Date Title
WO2016192555A1 (zh) 一种智能生物特征识别系统及识别方法
WO2017049922A1 (zh) 一种图像信息采集装置、图像采集方法及其用途
CN102855471B (zh) 远距离虹膜智能成像装置及方法
WO2016112630A1 (zh) 一种影像识别系统及方法
CN101051349B (zh) 采用主动视觉反馈的双目虹膜采集设备
WO2015172514A1 (zh) 图像采集装置和方法
US20050084137A1 (en) System and method for iris identification using stereoscopic face recognition
CN104268968B (zh) 具有虹膜识别功能的可视门禁、安全门锁和安全门
CN103049744A (zh) 门禁用识别设备及虹膜识别设备
CN204948210U (zh) 一种图像信息采集装置
CN113221865A (zh) 单摄像头双眼虹膜图像采集方法及装置
CN101369311A (zh) 一种采用主动视觉反馈的小型化虹膜识别模块
CN103530617B (zh) 虹膜对准与采集装置
TWI618970B (zh) 自動對焦拍照系統及電子裝置
CN108965683A (zh) 智能人脸识别摄像头
CN202632316U (zh) 远距离非侵入式虹膜图像获取系统
CN106101562A (zh) 一种根据眼睛动作拍照的照相设备
CN106408695A (zh) 一种虹膜识别智能锁具的装置和方法
CN107430276A (zh) 头戴式显示设备
CN218497546U (zh) 识别模组和门锁控制系统
CN204515275U (zh) 后置摄像头式后视眼镜
CN209044219U (zh) 一种基于眼球追踪技术的放大式眼镜
CN205983621U (zh) 一种虹膜识别智能锁具的装置
CN1752990A (zh) 便携式虹膜图像采集装置
CN2768661Y (zh) 便携式虹膜图像采集装置

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 16802478

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 16802478

Country of ref document: EP

Kind code of ref document: A1

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 30/05/2018)

122 Ep: pct application non-entry in european phase

Ref document number: 16802478

Country of ref document: EP

Kind code of ref document: A1