CN208903379U - A kind of anti-theft alarm system based on face and voice recognition - Google Patents

A kind of anti-theft alarm system based on face and voice recognition Download PDF

Info

Publication number
CN208903379U
CN208903379U CN201821331193.9U CN201821331193U CN208903379U CN 208903379 U CN208903379 U CN 208903379U CN 201821331193 U CN201821331193 U CN 201821331193U CN 208903379 U CN208903379 U CN 208903379U
Authority
CN
China
Prior art keywords
processing unit
operation processing
face
sensor
cloud
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN201821331193.9U
Other languages
Chinese (zh)
Inventor
薛振峰
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.)
Shenzhen Yiaikesi Information Technology Co Ltd
Original Assignee
Shenzhen Yiaikesi Information Technology Co Ltd
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 Shenzhen Yiaikesi Information Technology Co Ltd filed Critical Shenzhen Yiaikesi Information Technology Co Ltd
Priority to CN201821331193.9U priority Critical patent/CN208903379U/en
Application granted granted Critical
Publication of CN208903379U publication Critical patent/CN208903379U/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Landscapes

  • Alarm Systems (AREA)

Abstract

A kind of anti-theft alarm system based on face and voice recognition provided by the utility model, using multimedia terminal, cloud, network, operation processing unit, the structure of at least four sensors, at least two cameras, at least two microphones, local memory device, reserve battery, after the sensor, the camera, the microphone signal are carried out processing operation, as a result the cloud is sent to by network, a part of result is fed back to the operation processing unit after the cloud progress secondary calculating, realizes the controls such as alarm, auditory tone cues and switch;The technical issues of another part result is pushed to the mode of the multimedia terminal, and solution has a single function in the prior art, and precision is low, lacks dynamic identification, realizes that multi-environment adaptability is high, has recognizer ability, precision is high, the high technical effect of Dynamic Recognition.

Description

A kind of anti-theft alarm system based on face and voice recognition
Technical field
The utility model relates to anti-theft alarm system field, in particular to a kind of based on face and voice recognition Anti-theft alarm system.
Background technique
Conventional images identify that product obtains image by camera, identification comparison are carried out by cloud or local, to sentence Disconnected personnel whether before typing personnel's image;For realizing the identification etc. of enabling, attendance and unauthorized person.
Usually cloud identification or local identification choosing is not first, there are two types of identifications to be used in combination;Lack a variety of alarm mechanisms; Do not increase speech recognition;There is no multi-cam identification;Accuracy of identification is not high, lacks the adaptive algorithm under a variety of environment It practises;Lack Dynamic Recognition judgement.
Summary of the invention
The utility model provides a kind of anti-theft alarm system based on face and voice recognition, using multimedia terminal, cloud End, operation processing unit, at least four sensors, at least two cameras, at least two microphones, is locally stored and sets network Standby, reserve battery, the operation processing unit and the cloud pass through network dynamic coded communication, the multimedia terminal and institute State cloud by network carry out dynamic encryption communication, the operation processing unit respectively with the sensor, the camera, institute Microphone, the local memory device, reserve battery connection are stated, the camera is recognition of face camera, the fortune Calculate processing unit by the sensor, the camera, the microphone acquisition come data carry out processing and will handle Data afterwards export the cloud, and the cloud feeds back to a part of result the data progress secondary calculating received described Operation processing unit realizes the controls such as alarm, auditory tone cues and switch, and it is whole that another part result is pushed to the multimedia End, the mode with host personalityization prompting, guest close friend's prompting, stranger's early warning and warning function, solves in the prior art It has a single function, precision is low, lacks dynamic the technical issues of identifying.
This burglar alarm based on face and voice recognition that the utility model provides to solve above-mentioned technical problem System, including multimedia terminal, cloud, network, operation processing unit, at least four sensors, at least two cameras, at least Two microphones, local memory device, reserve batteries, the operation processing unit and the cloud are encrypted logical by network dynamic Letter, the multimedia terminal and the cloud by network progress dynamic encryption communication, the operation processing unit respectively with institute Sensor, the camera, the microphone, the local memory device, reserve battery connection are stated, the camera is Recognition of face camera, the operation processing unit by the sensor, the camera, the microphone acquisition come Data handle and data export the cloud by treated, and the cloud carries out secondary calculating handle to the data received A part of result feeds back to the operation processing unit, realizes the controls such as alarm, auditory tone cues and switch, another part knot Fruit is pushed to the multimedia terminal, has host personalityization prompting, guest close friend's prompting, stranger's early warning and warning function.
The operation processing unit includes CPU and at least two data-interfaces, and the operation processing unit is to the sensing Device parameter is configured, read sensor detection data simultaneously analyzed, the operation processing unit to the camera into Row setting, and the image information of the camera is read, the operation processing unit reads the data of the microphone, carries out sound Frequency analysis and calculating, the operation processing unit control lamp, loudspeaker, switch etc., and export some peripheral hardwares and can identify Signal or waveform, the operation processing unit carries out the data of the sensor, photo, video, audio data preliminary Calculation processing.
The sensor includes infrared sensor, temperature check sensor, human body sensor, light sensor, institute It is logical with the operation processing unit respectively to state infrared sensor, temperature check sensor, human body sensor, light sensor Letter, the sensor acquisition setting data value are simultaneously exported setting data value to the operation processing unit, the setting data Value includes the infrared, temperature check, human body sensing, light.
The camera at least there are two, the camera is communicated with the operation processing unit respectively, at least two institutes Camera is stated towards different, camera the acquisition different angle, the image data of different time points, while multiple angles can also To collect the human body image of multidimensional, more fully image and video data are provided for the operation processing unit.
The microphone at least there are two, the microphone position is non-conterminous, at least two microphones acquisitions These data are sent to operation processing unit processing by the audio data in multiple and different orientation of detected object.
The local memory device is data storage device, and the local memory device and the operation processing unit are logical Letter, the local memory device storage setting parameter, local image recognition algorithm, legal or illegal face and sound ginseng Number.
The reserve battery be Rechargeable battery, the reserve battery using lithium battery, nickel-metal hydride battery, lead-acid battery its One of.
The dynamic encryption is dynamic dual encryption algorithm, and the dynamic encryption is to the operation processing unit to the cloud End, the cloud are encrypted to the data between the multimedia terminal.
The utility model has the advantages that having multi-environment Adaptive Technology possessed by the utility model, temperature, light sensor pair are used The environment of recognition of face is detected, and starts the infrared light supply of different wave length according to detecting result, after obtaining image data, is called Corresponding environment learning algorithm carries out calculating identification before;Learning ability with recognizer, for high temperature, low temperature, high humidity, A variety of environment such as drying, daytime, night, rainy day, outdoor interior carry out actual measuring and calculating study, it is ensured that, can be with after algorithm learns Calculating identification is carried out to different ambient image and voice data;The processing of unstable networks situation, Cloud Server timing update For the black and white sample data of algorithm and user setting to operation processing unit, operation processing unit can be not connected to Cloud Server In the case of, independent operating realizes that sensor, image, the local of sound save, it calculates and comparison, while according to parameter preset, into Portable lighter light, sound and output control;Anti- cheating processing, for using photo, making up, block the cheatings such as face, using more The data of camera, multi-microphone and human body physical sign sensor carry out particular algorithm processing, and quickly which kind of judgement positioning belongs to and make Control is alarmed and exported accordingly to disadvantage behavior according to default.
Detailed description of the invention
Fig. 1 is the anti-theft alarm system schematic diagram described in the utility model based on face and voice recognition.
Fig. 2 is operation processing unit connection schematic diagram described in the utility model.
Fig. 3 is user's registration flow chart described in the utility model.
Fig. 4 is user's login process figure described in the utility model.
Fig. 5 is black and white lists typing schematic diagram described in the utility model.
Fig. 6 is distal end functional schematic described in the utility model.
Fig. 7 is the server capability schematic diagram in cloud described in the utility model.
Fig. 8 is the encryption flow figure of dynamic encryption described in the utility model.
Specific embodiment
In conjunction with above-mentioned Detailed description of the invention specific embodiments of the present invention.
There is Fig. 1 it is found that it includes more matchmakers that the utility model, which provides a kind of anti-theft alarm system based on face and voice recognition, Body terminal, cloud, network, operation processing unit, at least four sensors, at least two cameras, at least two microphones, sheet Ground stores equipment, reserve battery, and the operation processing unit and the cloud pass through network dynamic coded communication, the multimedia Terminal and the cloud by network progress dynamic encryption communication, the operation processing unit respectively with the sensor, described Camera, the microphone, the local memory device, reserve battery connection, the camera are recognition of face camera shooting Head, the operation processing unit by the sensor, the camera, the microphone acquisition come data handled And data export the cloud by treated, the cloud carries out secondary calculating to the data received, and a part of result is anti- It feeds the operation processing unit, realizes the controls such as alarm, auditory tone cues and switch, another part result is pushed to described Multimedia terminal has host personalityization prompting, guest close friend's prompting, stranger's early warning and warning function.
As shown in Figure 2, the operation processing unit includes CPU, at least two data transmission interfaces, at least two output controls Interface processed, the operation processing unit are configured the sensor parameters, read the data of sensor detection and are divided Analysis, the operation processing unit is configured the camera, and reads the image information of the camera, including video and Picture data, the operation processing unit read the data of the microphone, carry out audio analysis and calculating, the calculation process Unit controls lamp, loudspeaker, switch etc., and exports signal or waveform that some peripheral hardwares can identify, at the operation Reason unit carries out preliminary calculation processing to the data of the sensor, photo, video, audio data, without network or net It is compared in the case that network communication has some setbacks with local black and white lists data;It is selected in the case where having network according to user setting It selects the local still network that compares to compare, black and white lists data is acquired by the sensor, the camera, the microphone, Such as face, the data of sound while being saved in local, also save a server for arriving cloud, the output control interface Output including lamplight pointing, alarming horn, speech horn, switch and some signals;It can be to setting, warning, intercommunication and defeated Some signals, such as switch, waveform etc. that peripheral hardware needs out.
The sensor includes infrared sensor, temperature check sensor, human body sensor, light sensor, institute It is logical with the operation processing unit respectively to state infrared sensor, temperature check sensor, human body sensor, light sensor Letter, the sensor acquisition setting data value are simultaneously exported setting data value to the operation processing unit, the setting data Value includes the infrared, temperature check, human body sensing, light, these signals cooperate, can play and accurately detect people's In the presence of and anti-cheating function.
The camera at least there are two, the camera is communicated with the operation processing unit respectively, at least two institutes Camera is stated towards different, camera the acquisition different angle, the image data of different time points, while multiple angles can also To collect the human body image of multidimensional, more fully image and video data are provided for the operation processing unit, can detecte Entering and go out, personalization is reminded, such as: prompting of going out, " small owner, today refuel!" " small owner, welcomes home!″.
The microphone at least there are two, the microphone position is non-conterminous, at least two microphones acquisitions These data are sent to operation processing unit processing, by multiple by the audio data in multiple and different orientation of detected object The microphone acquires data, and the audio data for playing noise reduction, judging orientation, acquiring detected object sends these data Operation processing unit processing is given, speech recognition and corresponding control is realized, voice interface may be implemented after Fast Learning.
The local memory device is data storage device, and the local memory device and the operation processing unit are logical Letter, the local memory device storage setting parameter, local image recognition algorithm, legal or illegal face and sound ginseng Conveniently in the absence of a network calculating is quickly timely compared in number.
The reserve battery be Rechargeable battery, the reserve battery using lithium battery, nickel-metal hydride battery, lead-acid battery its One of, can power for a long time can also be used as it is main it is electrically disconnected after emergency service;Outdoors or some conditions permits Environment, can using thermal energy, wind energy or solar energy give reserve battery charging, for maintaining the normal operation of system.
As shown in Figure 7 and Figure 8, the dynamic encryption be dynamic dual encryption algorithm, the dynamic encryption to the operation at Reason unit is encrypted to the cloud, the cloud to the data between the multimedia terminal, i.e., the described cloud to identification Different Encryption Algorithm is respectively adopted to user for the multimedia terminal and the cloud;The cloud adding to user terminal simultaneously Close algorithm provides polyalgorithm and selects to user, and cipher key user is specified;In addition the Encryption Algorithm and changeable key of system, constitute Safely controllable dual encryption algorithm;Ensure multimedia described in the operation processing unit to server and the cloud to user Data safety between terminal device, dual encryption refer to that server uses a kind of Encryption Algorithm of fixation, Mei Geshi to identification terminal The key of other terminal is not identical;Server is selected to specify to the algorithm between user's intelligent terminal by user, and key is also user Dynamic generation when setting, each user are different;Ensure the independence and safety of links encryption.
As shown in fig. 6, the Cloud Server of the distal end is the center that entire big data calculates, it is service using distribution more The mode of device backup is run, it is ensured that data safety is not lost;The major function of Cloud Server is user management, terminal parameter, sample Notebook data saves, and the sample data includes face characteristic code, the data such as sound characteristic code, push alarm, real-time status, And the monitoring device that historical information can surf the Internet to user's intelligent terminal, mobile phone, plate, computer etc..
The utility model provides this anti-theft alarm system based on face and voice recognition, when user logs in for the first time, hand Generator terminal passes through app, webpage or wechat: as shown in Figures 3 to 5, registration, login, binding are put into terminal device, and set to terminal Standby parameter is configured, the typing including black and white lists information.
Entire parameter setting includes: setting verification mode, face, sound, face+sound;The prompting of theft-proof terminal is set Mode: indicator light is reminded, and sound prompting can be set text, can also record, high-pitched speaker alarm etc.;Theft-proof terminal is set Operating mode: greet a guest mode, gos out, enters, passes in and out, anti-theft modes, greets a guest+antitheft;Setting server push is sent to user's intelligence The information of energy terminal: inlet/outlet statistic record, inlet/outlet image information, remote speech intercommunication etc..
The utility model provides this anti-theft alarm system based on face and voice recognition, operation environment adaptability skill Art detects the environment of recognition of face using temperature, light sensor, starts the infrared of different wave length according to detecting result Light source, after obtaining image data, corresponding environment learning algorithm carries out calculating identification before calling;With multi-environment recognizer Study, carry out actual survey for a variety of environment such as high temperature, low temperature, high humidity, drying, daytime, night, rainy day, outdoor interiors Mathematics is practised, it is ensured that after algorithm study, can carry out calculating identification to different ambient image and voice data;Perfection processing network The black and white sample data of unstable situation, Cloud Server timing more new algorithm and user setting is to operation processing unit, at operation Managing unit can be in the case where being not connected to Cloud Server, independent operating, realizes that sensor, image, the local of sound save, meter It calculates and compares, while according to parameter preset, carrying out light, sound and output control;It is handled with anti-cheating, for using photograph Piece makes up, blocks the cheatings such as face, and using multi-cam, the data of multi-microphone and human body physical sign sensor carry out special Which kind of cheating different algorithm process, quickly judgement positioning belong to, are alarmed and exported accordingly control according to default;Using Dual encryption algorithm: to a kind of Encryption Algorithm is used between server, the key of each equipment is different from operation processing unit, Ensure the secret and safe between individual;Server uses fixed key to the Encryption Algorithm between user's intelligent terminal, using use Family id and random code composition, all users are different, and multiple Encryption Algorithm are selected for user, it is ensured that communication security;Can remotely it lead to Words: by the microphone of theft-proof terminal, user's intelligent terminal may be implemented and identified personnel carry out voice communication.
It, cannot the above content is specific preferred embodiment further detailed description of the utility model is combined Assert that the utility model specific implementation is only limited to these instructions.For the ordinary skill people of the utility model technical field For member, without departing from the concept of the premise utility, a number of simple deductions or replacements can also be made, all should be considered as Belong to the protection scope of utility model.

Claims (8)

1. a kind of anti-theft alarm system based on face and voice recognition, it is characterised in that: the system includes multimedia terminal, cloud End, operation processing unit, at least four sensors, at least two cameras, at least two microphones, is locally stored and sets network Standby, reserve battery, the operation processing unit and the cloud pass through network dynamic coded communication, the multimedia terminal and institute State cloud by network carry out dynamic encryption communication, the operation processing unit respectively with the sensor, the camera, institute Microphone, the local memory device, reserve battery connection are stated, the camera is recognition of face camera.
2. the anti-theft alarm system according to claim 1 based on face and voice recognition, it is characterised in that: the operation Processing unit includes CPU, at least two data transmission interfaces, at least two output control interfaces, and the operation processing unit is logical It crosses the data transmission interface and is separately connected the sensor, the camera, the microphone, the local memory device, The operation processing unit is also connected with the reserve battery by the network connection cloud, the operation processing unit It connects.
3. the anti-theft alarm system according to claim 1 based on face and voice recognition, it is characterised in that: the sensing Device includes infrared sensor, temperature check sensor, human body sensor, light sensor, the infrared sensor, body temperature Detection sensor, human body sensor, light sensor are communicated with the operation processing unit respectively, the sensor acquisition Setting data value is simultaneously exported setting data value to the operation processing unit, and the setting data value includes the infrared, body Temperature detector survey, human body sensing, light.
4. the anti-theft alarm system according to claim 1 based on face and voice recognition, it is characterised in that: the camera shooting Head at least there are two, the camera communicate with the operation processing unit respectively, at least two cameras towards differences, The camera provides more fully image and video data for the operation processing unit.
5. the anti-theft alarm system according to claim 1 based on face and voice recognition, it is characterised in that: the Mike Wind at least there are two, the microphone position is non-conterminous, at least two microphones acquisition detected objects it is multiple These data are sent to operation processing unit processing by the audio data of different direction.
6. the anti-theft alarm system according to claim 1 based on face and voice recognition, it is characterised in that: the local Storage equipment is data storage device, and the local memory device is communicated with the operation processing unit, and described be locally stored sets Standby storage setting parameter, local image recognition algorithm, legal or illegal face and audio parameter.
7. the anti-theft alarm system according to claim 1 based on face and voice recognition, it is characterised in that: described spare Battery is Rechargeable battery, and the reserve battery is one such using lithium battery, nickel-metal hydride battery, lead-acid battery.
8. the anti-theft alarm system according to claim 1 based on face and voice recognition, it is characterised in that: the dynamic It is encrypted as dynamic dual encryption algorithm, the dynamic encryption is to the operation processing unit to the cloud, the cloud described in Data between multimedia terminal are encrypted.
CN201821331193.9U 2018-08-17 2018-08-17 A kind of anti-theft alarm system based on face and voice recognition Active CN208903379U (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201821331193.9U CN208903379U (en) 2018-08-17 2018-08-17 A kind of anti-theft alarm system based on face and voice recognition

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201821331193.9U CN208903379U (en) 2018-08-17 2018-08-17 A kind of anti-theft alarm system based on face and voice recognition

Publications (1)

Publication Number Publication Date
CN208903379U true CN208903379U (en) 2019-05-24

Family

ID=66569613

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201821331193.9U Active CN208903379U (en) 2018-08-17 2018-08-17 A kind of anti-theft alarm system based on face and voice recognition

Country Status (1)

Country Link
CN (1) CN208903379U (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110415724A (en) * 2019-08-08 2019-11-05 中南大学湘雅二医院 Transmission method, device, system and the computer readable storage medium of alert data
CN111768580A (en) * 2020-06-30 2020-10-13 上海上实龙创智能科技股份有限公司 Indoor anti-theft system and anti-theft method based on edge gateway

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110415724A (en) * 2019-08-08 2019-11-05 中南大学湘雅二医院 Transmission method, device, system and the computer readable storage medium of alert data
CN111768580A (en) * 2020-06-30 2020-10-13 上海上实龙创智能科技股份有限公司 Indoor anti-theft system and anti-theft method based on edge gateway

Similar Documents

Publication Publication Date Title
CN105469484B (en) A kind of APP smart homes lock
CN203588406U (en) Intelligent security system based on infrared detection and video monitoring
CN103810806B (en) Dormitory security alarm device and dormitory security alarm system
CN106760671B (en) A kind of multifunctional police silver pavilion
CN104881914A (en) Building access control security management system
CN205990117U (en) Intelligent security guard apparatus for controlling elevator
CN209625308U (en) A kind of Household security system
CN208903379U (en) A kind of anti-theft alarm system based on face and voice recognition
CN208875459U (en) A kind of intelligent safety helmet
CN108961497A (en) A kind of door-locking system with warning function
CN205566533U (en) Intelligent doorbell
CN108648316A (en) A kind of method for unlocking and system based on infrared detection and recognition of face
CN104809776A (en) Intelligent automatic door as well as graph recognition unlocking method and automatic control method of intelligent automatic door
CN107064588A (en) A kind of electricity anti-theft system and its application process
CN109544755A (en) A kind of intelligent door lock control method and system
CN109035657A (en) A kind of anti-theft alarm system based on face and voice recognition
CN104157055A (en) Safe-guard door lock with remote warning function
CN206961284U (en) A kind of intelligent door
CN207571817U (en) A kind of automatic vending machine alarm system of anti-violence damage and technology unlocking
CN118097869A (en) Intelligent park security management system based on digital twinning
CN106530563A (en) Antitheft system of computer information
CN104952232B (en) A kind of intelligence gate control method
CN209103457U (en) Project alarm system
CN206340123U (en) It is a kind of to set up the intelligent door monitored automatically in electronic peephole viewer
CN206339586U (en) A kind of electricity anti-theft system

Legal Events

Date Code Title Description
GR01 Patent grant
GR01 Patent grant