CN108520217A - A kind of face identification and Activity recognition system - Google Patents

A kind of face identification and Activity recognition system Download PDF

Info

Publication number
CN108520217A
CN108520217A CN201810264880.1A CN201810264880A CN108520217A CN 108520217 A CN108520217 A CN 108520217A CN 201810264880 A CN201810264880 A CN 201810264880A CN 108520217 A CN108520217 A CN 108520217A
Authority
CN
China
Prior art keywords
module
processor
electrically connected
recognition system
activity recognition
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN201810264880.1A
Other languages
Chinese (zh)
Inventor
朱旭辉
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Individual
Original Assignee
Individual
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 Individual filed Critical Individual
Priority to CN201810264880.1A priority Critical patent/CN108520217A/en
Publication of CN108520217A publication Critical patent/CN108520217A/en
Pending legal-status Critical Current

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06VIMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V40/00Recognition of biometric, human-related or animal-related patterns in image or video data
    • G06V40/10Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
    • G06V40/16Human faces, e.g. facial parts, sketches or expressions
    • G06V40/161Detection; Localisation; Normalisation
    • G06V40/166Detection; Localisation; Normalisation using acquisition arrangements
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N23/00Cameras or camera modules comprising electronic image sensors; Control thereof
    • H04N23/70Circuitry for compensating brightness variation in the scene
    • H04N23/74Circuitry for compensating brightness variation in the scene by influencing the scene brightness using illuminating means

Abstract

The invention belongs to recognitions of face and Activity recognition technical field, more particularly to a kind of face identification and Activity recognition system, including power module, processor and human-computer interaction module, memory and the picture processing chip being electrically connected respectively with processor, picture processing chip is connected with infrared camera and visible image capturing head, further includes the communication module, optical sensor and light compensating lamp being connect respectively with processor;Power module is electrically connected processor, human-computer interaction module, memory, infrared camera, visible image capturing head, communication module, optical sensor and light compensating lamp.Present invention camera shooting imaging effect is more preferable, and comprehensive two kinds of image recognition technologys make better recognition effect, and safety higher improves the search efficiency and accuracy rate when comparing, and identification method is various, easy to operate, and experience effect is good.

Description

A kind of face identification and Activity recognition system
Technical field
The invention belongs to recognitions of face and Activity recognition technical field, and in particular to a kind of face identification is known with behavior Other system.
Background technology
With application of the face recognition technology in mobile phone, also more and more occasions start to apply recognition of face and row at present For motion analysis technology.Existing face recognition technology mainly applies infrared image identification technology or visible light identification technology, Image recognition analysis is carried out, and is all that can only identify face or human action, product is basically used for security protection, attendance and door lock In.Technology has the following problems at present:
One, there are two big defects for existing infrared image technology:One is infrared quicker to ambient temperature variation Sense will be unable to effectively to detect eyes and move the other is infrared have non-penetrative to glass, therefore when driver wears glasses Make, may cause to judge by accident;And only identified with visible light, then it is influenced seriously by ambient light, on the big ground of insufficient light, variation Side, it is difficult to identify;
Two, existing face recognition technology can only identify success or not, identify and successfully then complete to verify, can not be by tight It snugly alarms when anxious situation, and contactless action directive can not be completed;
Three, existing visible light identification technology, it is difficult to which the sham face such as automatic identification video, picture and action are easy to crack;
Four, the prior art is difficult to balance between the two in identification misclassification rate and reject rate, especially in twins' identification Simple recognition of face is easy accidentally to know;
Five, existing product technology is designed exploitation both for certain application scenarios, must cooperation button, touch screen, voice Prompt etc. is operated, and complicated for operation, is carried out personalized exploitation and is carried out the difficulty height of the system integration;
Six, the prior art uses static state 1:N or 1:1 identification, it is desirable that verification personnel must reach certain distance and could identify, know Other speed is slow, and experience effect is bad.
Invention content
In order to solve the above problem of the existing technology, present invention aims at providing, a kind of recognition accuracy is high, knows Other mode is various, and easy to operate, recognition speed is fast, the good face identification of experience effect and Activity recognition system.
The technical solution adopted in the present invention is:
A kind of face identification and Activity recognition system, including processor and the power module that is electrically connected with processor, Further include the human-computer interaction module being respectively electrically connected with processor and power module, memory, picture processing chip, communication mould Block, optical sensor and light compensating lamp, picture processing chip are connected with the infrared camera being electrically connected respectively with power module and visible Light video camera head.
It may further be preferable that the power module includes accumulator and external alternating current, accumulator and external alternating current are logical Cross power supply adaptor electrical connection processor, human-computer interaction module, memory, picture processing chip, communication module, optical sensor, Infrared camera visible image capturing head and light compensating lamp.
It is usually powered using external alternating current, uses storage battery power supply, accumulator that can make system when having a power failure in addition With certain retardance, i.e., in the case of accidental power failure still can normal operation within a certain period of time, be unlikely to immediately without Method uses, therefore accumulator is used as stand-by power supply, greatly improves security performance.
It is even furthermore preferable that the human-computer interaction module is touch screen and button.
Touch screen, button, mobile phone, computer and tablet are currently used human-computer interaction module, of low cost, operation letter It is single.
It is even furthermore preferable that the communication module includes wireless communication module and/or wire communication module, the nothing Line communication module includes one or several in gsm module, 3G module, 4G modules, bluetooth module and WIFI module;It is described wired Communication module includes one or two of serial communication module and USB module.
Connection can be established between outside by being belonged to using communication module guarantee, facilitated and operated and backed up, improve peace Quan Xing, be also convenient in time and silently alarm, further increase safety.
It is even furthermore preferable that the light compensating lamp includes the visible light being respectively electrically connected with processor and power module Light compensating lamp and infrared light compensating lamp.
Visible light light compensating lamp carries out light filling for visible image capturing head, and infrared light compensating lamp is mended for infrared camera Light ensures that visible image capturing head and infrared camera all have good light under any pattern, big in insufficient light, variation Place, be also unlikely to be influenced by ambient light, in insufficient light, the big place of variation, ensure the accuracy judged.
It is even furthermore preferable that the processor is also electrically connected with speech horn, speech horn is electrically connected with power module It connects.
Speech horn can carry out voice prompt when carrying out face characteristic acquisition and behavior acquisition, can be used for usually Voice prompt, it is convenient and practical.
It is even furthermore preferable that the processor is arm processor.
Arm processor is the first item risc microcontroller of the low-power consumption cost of Acorn Co., Ltds of Britain design, volume Small, low-power consumption, low cost, high-performance;It supports Thumb (16)/ARM (32) double instruction set, 8/16 can be compatible with well Position device;Register is largely used, instruction execution speed is faster;Addressing system is flexibly simple, and execution efficiency is high.
Beneficial effects of the present invention are:
Present invention uses light compensating lamp and optical sensor, the variation of induced environment light automatically adjusts red in insufficient light The start and stop of outer light compensating lamp and visible light light compensating lamp, and automatic regulating lightness keep camera shooting imaging effect more preferable;The present invention is using infrared Camera and visible image capturing head, comprehensive two kinds of image recognition technologys make better recognition effect;In the processor of the present invention Default multiple facial expressions actions, can be used as special command and password, may be implemented to automatically control, the functions such as concealed alarm, carry Height identification authentication security;The present invention uses News Search alignments, improves search efficiency and accuracy rate when comparing;This hair It is bright compared with prior art, recognition accuracy is high, and identification method is various, easy to operate, and recognition speed is fast, and experience effect is good, more Safety.
Description of the drawings
Fig. 1 is the system block diagram of the present invention.
Specific implementation mode
Below in conjunction with the accompanying drawings and specific embodiment does further explaination to the present invention.
Embodiment 1:
As shown in Figure 1, a kind of face identification and Activity recognition system, including processor and it is electrically connected with processor Power module further includes the human-computer interaction module being respectively electrically connected with processor and power module, memory, image procossing core Piece, communication module, optical sensor and light compensating lamp, picture processing chip are connected with the infrared photography being electrically connected respectively with power module Head and visible image capturing head.
Preferably, the power module includes accumulator and external alternating current, and accumulator and external alternating current are suitable by power supply Orchestration is electrically connected processor, human-computer interaction module, memory, picture processing chip, communication module, optical sensor, infrared photography Head visible image capturing head and light compensating lamp.In accumulator electric-quantity deficiency, can also be charged a battery with power supply adaptor.
Preferably, the human-computer interaction module is touch screen and button, can be easily mobile phone, computer, tablet etc., be Currently used human-computer interaction module, of low cost, easy to operate, preferably computer herein.
Preferably, the communication module includes wireless communication module and/or wire communication module, the wireless communication module Including one or several in gsm module, 3G module, 4G modules, bluetooth module and WIFI module;The wire communication module packet Include one or two of serial communication module and USB module.I.e. communication module can be gsm module, 3G module, 4G modules, Bluetooth module, WIFI module, serial communication module or USB module can be serial communication module or combination, the GSM of USB module The combination of module and 3G module, the combination of 4G modules and WIFI module, the combination of bluetooth module and WIFI module, WIFI module and Combination of USB module etc., preferred communication module is the combination of WIFI module and USB module herein.
Preferably, the light compensating lamp includes the visible light light compensating lamp being respectively electrically connected with processor and power module and red Outer light compensating lamp.The processor is also electrically connected with speech horn, and speech horn is electrically connected with power module.
Preferably, the processor is arm processor, or other processors.
The course of work of the present invention is as follows:
1, personnel register:By the prompt of human-computer interaction module, according to infrared camera and visible image capturing head to face Collection apparatus is carried out, under the control of a processor compiles the face characteristic of extraction by personnel after picture processing chip is handled Code corresponds and is that form stores in memory.
2, according to requirements, order setting recognition mode is issued by human-computer interaction module:
1) personal identification pattern:Personal identification state, nonrecognition action are constantly in after startup.
2) action recognition pattern:Handle action recognition pattern, nonrecognition personnel identity always after.
3) recognition mode is combined:I.e. recognizable personnel can recognize that personnel's action, output personnel identity coding are carried out with it again Action coding.
4) expression password identification pattern:First personnel's face similarity identification is passed through, and facial expression is acted and moved Make that expression password that sequencing also will originally be set with the personnel is consistent to be just identified by.
The identifiable facial expression of face, which acts, includes:It is simple eye closure reach certain time length, magnify mouth reach certain time length, Discharge tongue reaches certain time length, left and right shakes the head and reaches certain number, left-hand rotation head, right-hand rotation head, come back, bow, smiling, rubbing one's eyes Eyeball eats, opens eyes wide etc. and setting one-to-one action coding.These face feelings actions may be set to be coerced by danger When and special command operate when condition adjudgement, carry out the corresponding command subsequent action.
3, light compensating lamp brightness is automatically adjusted after equipment starts, and is worked according to the recognition mode of setting:
1) when personal identification pattern:When detecting face, according to the image of infrared camera and visible image capturing head into Row handles and completes currently to identify the face characteristic acquisition of personnel, is compared with the characteristic in memory, works as similarity When reaching the threshold values of setting, judge that personal identification passes through, store and exports personnel's coding by various communication modes;Work as similarity Less than setting threshold values when, judge that personal identification does not pass through, by various communication modes output identification not by result.
2) when action recognition pattern:When detecting face, according to the image of infrared camera and visible image capturing head into Row handles and completes currently to identify upper part of the body collection apparatus, is compared with the characteristic of previous image, analyzes changing features Position and variation tendency are compared according to stored part variation and variation tendency feature in program, when similarity reaches When the threshold values of setting, exports corresponding action by different communication modes and encode;When threshold values of the similarity less than setting, after Continuous description image.
3) when combining recognition mode:When detecting face, according to the image of infrared camera and visible image capturing head into Row handles and completes currently to identify upper part of the body collection apparatus, the other identification of advanced pedestrian, if personnel identity identification does not pass through Side exports the result of recognition failures.After personnel identity is identified by, action recognition is carried out, action recognition passes through, and exports personnel couple It should encode and act corresponding coding;Action recognition does not pass through, then exports personnel and correspond to coding.
4) expression password identification pattern:The facial expression action password and action that setting keeper verifies in systems Sequencing can set 1 facial expression action or be combined as password by the sequence of multiple actions.First to personnel face when identification Portion's similarity is identified, after similarity passes through, by identification personnel facial expression action and its sequencing whether with original Setting is consistent, if personnel's face similarity, facial expression act, the sequencing of action is all consistent, is judged as that identification is logical It crosses, output personnel's coding.When similarity identification, action or action sequencing and former setting are inconsistent, then judge to fail, output Recognition failures result.
4, when identifying personnel identity, using News Search, mode is as follows:
When human body close to when, infrared camera and visible image capturing head have recognized face Partial Feature, i.e., by memory It is one group that middle similarity personnel in the top, which encode,;In not stopping near procedure, by certain time length and close to distance length It is compared several times, forms several groups similarity personnel in the top and encode array N1;By the personnel of this several groups array Coded data is compared, and number sequence is encoded by the personnel repeated, and the forward recomposition that will sort is new array M2, Personnel are close to after certain distance, the final feature aspect ratio pair corresponding with the personnel of array M2 coding extracted, when similar When spending highest and reaching threshold values, output personnel's coding.When aspect ratio corresponding with the personnel of M2 coding is to being not up to threshold values, then The personnel characteristics other than M2 are compared, when reaching threshold values if there is personnel characteristics' similarity, output personnel's coding, if still do not reached When to threshold values, judge that personal identification does not pass through, the result not communicated by the output identification of various communication modes.
Present invention uses light compensating lamp and optical sensor, the variation of induced environment light automatically adjusts red in insufficient light The start and stop of outer light compensating lamp and visible light light compensating lamp, and automatic regulating lightness keep camera shooting imaging effect more preferable;The present invention is using infrared Camera and visible image capturing head, comprehensive two kinds of image recognition technologys make better recognition effect;In the processor of the present invention Default multiple facial expressions actions, can be used as special command and password, may be implemented to automatically control, the functions such as concealed alarm, carry Height identification authentication security;The present invention uses News Search alignments, improves search efficiency and accuracy rate when comparing;This hair It is bright compared with prior art, recognition accuracy is high, and identification method is various, easy to operate, and recognition speed is fast, and experience effect is good, more Safety.
The present invention is not limited to above-mentioned optional embodiment, anyone can show that other are each under the inspiration of the present invention The product of kind form.Above-mentioned specific implementation mode should not be understood the limitation of pairs of protection scope of the present invention, protection of the invention Range should be subject to be defined in claims, and specification can be used for interpreting the claims.

Claims (7)

1. a kind of face identification and Activity recognition system, it is characterised in that:It is electrically connected including processor and with processor Power module further includes the human-computer interaction module being respectively electrically connected with processor and power module, memory, image procossing core Piece, communication module, optical sensor and light compensating lamp, picture processing chip are connected with the infrared photography being electrically connected respectively with power module Head and visible image capturing head.
2. a kind of face identification according to claim 1 and Activity recognition system, it is characterised in that:The power supply mould Block includes accumulator and external alternating current, and accumulator and external alternating current are electrically connected processor, human-computer interaction by power supply adaptor Module, memory, picture processing chip, communication module, optical sensor, infrared camera visible image capturing head and light compensating lamp.
3. a kind of face identification according to claim 1 and Activity recognition system, it is characterised in that:The man-machine friendship Mutual module is touch screen and button.
4. a kind of face identification according to claim 1 and Activity recognition system, it is characterised in that:The communication mould Block includes wireless communication module and/or wire communication module, and the wireless communication module includes gsm module, 3G module, 4G moulds It is one or several in block, bluetooth module and WIFI module;The wire communication module includes serial communication module and USB module One or two of.
5. a kind of face identification according to claim 1 and Activity recognition system, it is characterised in that:The light compensating lamp Including the visible light light compensating lamp being respectively electrically connected with processor and power module and infrared light compensating lamp.
6. a kind of face identification and Activity recognition system, feature according to any one of claim 1-5 exist In:The processor is also electrically connected with speech horn, and speech horn is electrically connected with power module.
7. a kind of face identification according to claim 1 or 2 and Activity recognition system, it is characterised in that:The place Reason device is arm processor.
CN201810264880.1A 2018-03-28 2018-03-28 A kind of face identification and Activity recognition system Pending CN108520217A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201810264880.1A CN108520217A (en) 2018-03-28 2018-03-28 A kind of face identification and Activity recognition system

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201810264880.1A CN108520217A (en) 2018-03-28 2018-03-28 A kind of face identification and Activity recognition system

Publications (1)

Publication Number Publication Date
CN108520217A true CN108520217A (en) 2018-09-11

Family

ID=63430665

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201810264880.1A Pending CN108520217A (en) 2018-03-28 2018-03-28 A kind of face identification and Activity recognition system

Country Status (1)

Country Link
CN (1) CN108520217A (en)

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109698899A (en) * 2019-02-20 2019-04-30 重庆工程职业技术学院 A kind of double mode machine visual lens and control method
CN109823306A (en) * 2019-02-22 2019-05-31 广东远峰汽车电子有限公司 Car door unlocking method, device, system and readable storage medium storing program for executing
CN110191292A (en) * 2019-06-28 2019-08-30 联想(北京)有限公司 Information processing method and its system, medium and electronic equipment
CN110287794A (en) * 2019-05-24 2019-09-27 华纳科技(深圳)有限公司 A kind of production line face intelligent identifying system
CN114565993A (en) * 2022-02-24 2022-05-31 深圳福鸽科技有限公司 Intelligent lock system based on face recognition and cat eye

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101976356A (en) * 2010-09-30 2011-02-16 惠州市华阳多媒体电子有限公司 Face identification method for surfing identification policy, and identification system thereof
DE102009022571A1 (en) * 2009-05-25 2011-12-15 Peter G. Stolzenberg Person double images and photographic holography arrangement for use in e.g. passport, for automatically obtaining access to money, has normal color or black-white-photo and infrared-black-white-photo integrated in scale
CN102915434A (en) * 2012-09-26 2013-02-06 上海交通大学 Face recognition system based on low-consumption embedded platform
US20160162673A1 (en) * 2014-12-05 2016-06-09 Gershom Kutliroff Technologies for learning body part geometry for use in biometric authentication
CN107766785A (en) * 2017-01-25 2018-03-06 丁贤根 A kind of face recognition method

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
DE102009022571A1 (en) * 2009-05-25 2011-12-15 Peter G. Stolzenberg Person double images and photographic holography arrangement for use in e.g. passport, for automatically obtaining access to money, has normal color or black-white-photo and infrared-black-white-photo integrated in scale
CN101976356A (en) * 2010-09-30 2011-02-16 惠州市华阳多媒体电子有限公司 Face identification method for surfing identification policy, and identification system thereof
CN102915434A (en) * 2012-09-26 2013-02-06 上海交通大学 Face recognition system based on low-consumption embedded platform
US20160162673A1 (en) * 2014-12-05 2016-06-09 Gershom Kutliroff Technologies for learning body part geometry for use in biometric authentication
CN107766785A (en) * 2017-01-25 2018-03-06 丁贤根 A kind of face recognition method

Cited By (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109698899A (en) * 2019-02-20 2019-04-30 重庆工程职业技术学院 A kind of double mode machine visual lens and control method
CN109698899B (en) * 2019-02-20 2024-04-05 重庆工程职业技术学院 Dual-mode machine vision lens
CN109823306A (en) * 2019-02-22 2019-05-31 广东远峰汽车电子有限公司 Car door unlocking method, device, system and readable storage medium storing program for executing
CN110287794A (en) * 2019-05-24 2019-09-27 华纳科技(深圳)有限公司 A kind of production line face intelligent identifying system
CN110191292A (en) * 2019-06-28 2019-08-30 联想(北京)有限公司 Information processing method and its system, medium and electronic equipment
CN110191292B (en) * 2019-06-28 2020-09-25 联想(北京)有限公司 Information processing method, system, medium, and electronic device
CN114565993A (en) * 2022-02-24 2022-05-31 深圳福鸽科技有限公司 Intelligent lock system based on face recognition and cat eye
CN114565993B (en) * 2022-02-24 2023-11-17 深圳福鸽科技有限公司 Intelligent lock system based on face recognition and cat eye

Similar Documents

Publication Publication Date Title
CN108520217A (en) A kind of face identification and Activity recognition system
CN104933344B (en) Mobile terminal user identity authentication device and method based on multi-biological characteristic mode
Lwin et al. Automatic door access system using face recognition
CN105700363A (en) Method and system for waking up smart home equipment voice control device
CN104217480A (en) Intelligent entrance guard system based on fingerprint and finger vein joint recognition
CN203366348U (en) Face recognition system and entrance guard system
EP3610738B1 (en) Method and system for unlocking electronic cigarette
CN106682607A (en) Offline face recognition system and offline face recognition method based on low-power-consumption embedded and infrared triggering
CN205354143U (en) Intelligent door access control system
CN204791017U (en) Mobile terminal users authentication device based on many biological characteristics mode
CN205427946U (en) Hotel moves in management system
WO2018210048A1 (en) Fpga-based hardware iris recognition system and method, and readable storage medium
CN103336950A (en) Human face identification method and system
CN106446992A (en) System and method for contrasting identity information based on mobile terminal
US20200394390A1 (en) Apparatus and method for vehicle driver recognition and applications of same
CN114220130A (en) Non-contact gesture and palm print and palm vein fused identity recognition system and method
CN107292217A (en) A kind of Dual-factor identity authentication system based on recognition of face Yu vena metacarpea technology
CN106846552A (en) A kind of intelligent entrance guard control system
CN106503532A (en) A kind of mouse and its using method based on fingerprint recognition
CN109003367A (en) Control method and device, storage medium, the terminal of gate inhibition
CN204204058U (en) A kind of based on fingerprint and the intelligent access control system referring to combined with intravenous identification
CN207440783U (en) Hardware iris authentication system based on FPGA
CN212809322U (en) Hand vein face multi-mode identification entrance guard attendance all-in-one
CN113724421B (en) Fingerprint module, fingerprint identification system, control method and intelligent lock
US10824850B2 (en) Body information analysis apparatus capable of indicating shading-areas

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
RJ01 Rejection of invention patent application after publication

Application publication date: 20180911

RJ01 Rejection of invention patent application after publication