CN109697815A - Anti-theft communication network alarming method, appliance arrangement and storage medium - Google Patents
Anti-theft communication network alarming method, appliance arrangement and storage medium Download PDFInfo
- Publication number
- CN109697815A CN109697815A CN201910065505.9A CN201910065505A CN109697815A CN 109697815 A CN109697815 A CN 109697815A CN 201910065505 A CN201910065505 A CN 201910065505A CN 109697815 A CN109697815 A CN 109697815A
- Authority
- CN
- China
- Prior art keywords
- sample
- video
- communication network
- characteristic value
- monitoring
- 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
Links
Classifications
-
- G—PHYSICS
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B13/00—Burglar, theft or intruder alarms
- G08B13/18—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength
- G08B13/189—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength using passive radiation detection systems
- G08B13/194—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength using passive radiation detection systems using image scanning and comparing systems
- G08B13/196—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength using passive radiation detection systems using image scanning and comparing systems using television cameras
- G08B13/19602—Image analysis to detect motion of the intruder, e.g. by frame subtraction
- G08B13/19613—Recognition of a predetermined image pattern or behaviour pattern indicating theft or intrusion
Landscapes
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Alarm Systems (AREA)
Abstract
The present invention relates to the technical fields of security protection, more particularly, to a kind of anti-theft communication network alarming method, appliance arrangement and storage medium, the anti-theft communication network alarming method is the following steps are included: step S10: obtaining the quantity of site alarm terminal, and establishes network communication protocol according to the quantity and monitor supervision platform of the alarm terminal；Step S20: the monitor supervision platform obtains monitoring video and is monitored in real time to site；Step S30: analysis identification is carried out to the monitoring video, and is detected according to the result of analysis identification, testing result is obtained；Step S40: it if reflecting there are suspicious actions from the testing result, according to the quantity of the alarm terminal, generates and sends alarm signal to the alarm terminal.The present invention has the effect of promoting the timeliness of alarm.
Description
Technical field
The present invention relates to the technical fields of security protection, more particularly, to a kind of anti-theft communication network alarming method, device
Equipment and storage medium.
Background technique
Currently, security protection system can protect the safety of important place, such as financial grid point, such as prevent personnel from invading,
Safety accident or suspicious actions etc. are detected, and is used for the retrospect of event.Security protection system specifically includes that video monitoring system
System, alarm control system, access control system etc., these systems connect each other again independently of one another.
When existing video monitoring, video recording of monitoring area can not when there are illegal incidents where only obtaining it yet
And alarm can not even if subsequent can find illegal incidents by video recording, but because lawbreaker has moved away from scene
Illegal incidents are handled in time.
Summary of the invention
The object of the present invention is to provide anti-theft communication network alarming method, the appliance arrangements of a kind of timeliness for promoting alarm
And storage medium.
Foregoing invention purpose of the invention has the technical scheme that
A kind of anti-theft communication network alarming method, the anti-theft communication network alarming method the following steps are included:
Step S10: the quantity of site alarm terminal is obtained, and network is established according to the quantity and monitor supervision platform of the alarm terminal
Communication protocol；
Step S20: the monitor supervision platform obtains monitoring video and is monitored in real time to site；
Step S30: analysis identification is carried out to the monitoring video, and is detected according to the result of analysis identification, is detected
As a result；
Step S40: it if reflecting there are suspicious actions from the testing result, according to the quantity of the alarm terminal, generates simultaneously
Alarm signal is sent to the alarm terminal.
By using above-mentioned technical proposal, by the way that network communication association is put up between monitor supervision platform and alarm terminal in advance
View, makes to be communicated by wireless network between monitor supervision platform and alarm terminal, it is ensured that detected in monitor supervision platform
When suspicious actions, alarm signal can be sent to the alarm terminal of site scene Security Personnel in time, make the security at scene
Personnel can be in time to can be handled with behavior；Meanwhile identifying monitor video by server-side, manpower can be saved
To the real time monitoring of monitor video, while guaranteeing control and monitoring, the efficiency of monitoring also can be further promoted.
The present invention is further arranged to: before the step S40, the anti-theft communication network alarming method further include with
Lower step:
Step S41: the performance indicator of detection service device cluster；
Step S42: if detecting, the performance indicator of the server cluster is more than capacity threshold, sends sample to the monitor supervision platform
This library splits request；
Step S43: if getting the sample database splitting condition information, the sample database is split, obtains sample word bank；
Step S44: characteristic value is extracted from the sample word bank, obtains the characteristic value library；
Step S45: if after extracting the characteristic value library, detection service device cluster resource loading condition, if detecting, resource is negative
Carrying situation is more than preset threshold value, then sends and increase server cluster request.
Further, the step S44 the following steps are included:
Step S441: according to the quantity of the sample word bank, the characteristic value library is established；
Step S442: extracting the characteristic value from the sample word bank, is stored in the list of feature values, and the list of feature values is stored in and is corresponded to
Characteristic value library；
Step S444: being monitored the sample size in the sample word bank, if detecting, the sample size is more than default
Threshold value, then the sample word bank and the corresponding characteristic value library are further split.
It is more than preset by the performance indicator of detection service end cluster, and to capacity by using above-mentioned technical proposal
Threshold value when, sample database and characteristic value library are split, to alleviate the pressure of server-side cluster used.
The present invention is further arranged to: the step S30 includes:
Step S301: identifying the face picture in the monitoring video, according to the face picture, using the monitoring video as
Detect video；
Step S302: the detection video input to video surveillance model is detected, the testing result is obtained.
Further, before the step S302, the anti-theft communication network alarming method further include:
Step S3021: according to history monitoring video, it regard history theft as Sample video；
Step S3022: continuous N frame monitored picture is chosen from the Sample video one by one, and is divided according to the Sample video
Group obtains the training image collection, and wherein N is positive integer；
Step S3023: the training image collection is input in convolutional neural networks MODEL C NN, the training image collection
Carry out down-sampling, convolution sum pondization operates, obtain monitoring feature vector corresponding with all training image collections；
Step S3024: it is trained, is corresponded to using the monitoring feature vector of the LSTM network to the training image collection
The video detection model.
By using above-mentioned technical proposal, sub-frame processing is carried out to history monitoring video in advance, obtains corresponding training figure
Image set, and the training image collection is able to solve single frames picture in video and is known using being trained in CNN-LSTM network
When other, the case where monitoring in video not can accurately reflect, improve the accuracy to the detection for stealing situation in video.
Foregoing invention purpose two of the invention has the technical scheme that
A kind of anti-theft communication network alarming apparatus, the anti-theft communication network alarming apparatus include:
Communication building block, for obtaining the quantity of site alarm terminal, and it is flat according to the quantity of the alarm terminal and monitoring
Platform establishes network communication protocol；
Monitoring module obtains monitoring video for the monitor supervision platform and is monitored in real time to site；
Identification module for carrying out analysis identification to the monitoring video, and is detected according to the result of analysis identification, is obtained
Testing result；
Alarm module, if for reflecting there are suspicious actions from the testing result, it is raw according to the quantity of the alarm terminal
Alarm terminal described in Cheng Bingxiang sends alarm signal.
By using above-mentioned technical proposal, by the way that network communication association is put up between monitor supervision platform and alarm terminal in advance
View, makes to be communicated by wireless network between monitor supervision platform and alarm terminal, it is ensured that detected in monitor supervision platform
When suspicious actions, alarm signal can be sent to the alarm terminal of site scene Security Personnel in time, make the security at scene
Personnel can be in time to can be handled with behavior；Meanwhile identifying monitor video by server-side, manpower can be saved
To the real time monitoring of monitor video, while guaranteeing control and monitoring, the efficiency of monitoring also can be further promoted.
Foregoing invention purpose three of the invention has the technical scheme that
A kind of computer equipment, including memory, processor and storage are in the memory and can be on the processor
The computer program of operation, the processor realize above-mentioned anti-theft communication network alarming method when executing the computer program
Step.
Foregoing invention purpose four of the invention has the technical scheme that
A kind of computer readable storage medium, the computer-readable recording medium storage have computer program, the computer
The step of above-mentioned anti-theft communication network alarming method is realized when program is executed by processor.
In conclusion advantageous effects of the invention are as follows:
1. making monitor supervision platform and alarm terminal by putting up network communication protocol between monitor supervision platform and alarm terminal in advance
Between can be communicated by wireless network, it is ensured that when monitor supervision platform has detected suspicious actions, can in time by
Alarm signal is sent to the alarm terminal of site scene Security Personnel, enable scene Security Personnel in time to can with behavior into
Row processing；Meanwhile identifying monitor video by server-side, manpower can be saved to the real time monitoring of monitor video,
While guaranteeing control and monitoring, the efficiency of monitoring also can be further promoted；
2. by the performance indicator of detection service end cluster, and to capacity be more than preset threshold value when, to sample database and feature
Value library is split, to alleviate the pressure of server-side cluster used.
Detailed description of the invention
Fig. 1 is a flow chart of anti-theft communication network alarming method in one embodiment of the invention；
Fig. 2 is the flow chart that anti-theft communication network alarming method is another in one embodiment of the invention；
Fig. 3 is the implementation flow chart of step S44 in anti-theft communication network alarming method in one embodiment of the invention；
Fig. 4 is the implementation flow chart of step S30 in anti-theft communication network alarming method in one embodiment of the invention；
Fig. 5 is another flow chart of anti-theft communication network alarming method in one embodiment of the invention；
Fig. 6 is a functional block diagram of anti-theft communication network alarming apparatus in one embodiment of the invention；
Fig. 7 is a schematic diagram of computer equipment in one embodiment of the invention.
Specific embodiment
Below in conjunction with attached drawing, invention is further described in detail.
Embodiment one:
In one embodiment, it as shown in Figure 1, being a kind of anti-theft communication network alarming method disclosed by the invention, including walks as follows
It is rapid:
Step S10: the quantity of site alarm terminal is obtained, and network communication is established according to the quantity and monitor supervision platform of alarm terminal
Agreement.
In the present embodiment, site refers to each financial institution, such as the places such as business hall under the line of bank.People would generally
It carries various bank cards or cash goes to the site and carries out business handling, it is therefore desirable to which the site and the site are nearby carried out
Anti-thefting monitoring.Alarm terminal refers in Security Personnel's wearing by site, sends for receiving monitor supervision platform or monitoring center
The equipment such as the equipment, such as headset of instruction.Monitor supervision platform refers to the platform for obtaining monitoring video and being monitored.
Specifically, according to the quantity of the alarm terminal in site, pass through network communication protocol, such as wireless network skill
Art, distributing mailing address for each alarm terminal will be established by the mailing address between each alarm terminal and monitor supervision platform
Communication enables monitor supervision platform to send message to alarm terminal, and alarm terminal can receive and be disappeared by what monitor supervision platform was sent
Breath.
Step S20: monitor supervision platform obtains monitoring video and is monitored in real time to site.
Specifically, by arranging the photographic devices such as monitoring device, such as camera in site, around site and site
It is monitored, and by the established network communication protocol of step S10, it is flat that the monitoring video of monitoring is sent to monitoring in real time
Platform.
Step S30: analysis identification is carried out to monitoring video, and is detected according to the result of analysis identification, is detected
As a result.
Specifically, when monitor supervision platform gets the monitoring video, analysis identification is carried out to the situation in monitoring video.It is first
First identify the portrait in monitoring video, it specifically can be by first identifying the face in monitoring video, and then by identifying
Face identifies the portrait in monitoring video.
Further, after identifying the portrait in monitoring video, and then identify that the behavior of the personage in monitoring video is moved
Make and detected, obtains testing result.
Step S40: it if reflecting there are suspicious actions from testing result, according to the quantity of alarm terminal, generates and to report
Alert terminal sends alarm signal.
Specifically, if reflection summarizes that there are suspicious actions to monitoring video from testing result, such as detect that someone doubts
The knapsack or pocket that hand or tool are seemingly protruded into other people, then determine the behavior for suspicious actions.
Further, each alarm is sent to eventually using the suspicious actions as alarm signal according to the quantity of alarm terminal
In end, the Security Personnel at notice site scene is handled in time.
In the present embodiment, by putting up network communication protocol between monitor supervision platform and alarm terminal in advance, make to supervise
It can be communicated by wireless network between control platform and alarm terminal, it is ensured that detected suspicious actions in monitor supervision platform
When, alarm signal can be sent to the alarm terminal of site scene Security Personnel in time, enable the Security Personnel at scene
In time to can be handled with behavior；Meanwhile identifying monitor video by server-side, manpower can be saved, monitoring is regarded
The real time monitoring of frequency also can further promote the efficiency of monitoring while guaranteeing control and monitoring.
In one embodiment, as shown in Fig. 2, before step S40, anti-theft communication network alarming method further includes following step
It is rapid:
Step S41: the performance indicator of detection service device cluster.
In the present embodiment, server cluster refer to one for being handled in data of the backstage to monitor supervision platform or
The cluster of multiple server compositions.Performance indicator refers to the service condition of server cluster.
Specifically, when detecting the service condition of the server cluster, the detection service device cluster by way of heartbeat detection
Service condition, the index specifically detected may include the QPS(Qeury Per Second inquiry times per second of database),
TPS(Transaction Per Second number of transactions per second) and database connection number.In detection QPS, TPS and database
It, can be by installing Grafana tool, and to QPS, TPS and database under the tool under linux system when connection number
Connection number is detected.
Wherein, QPS refers to the number of database energy response application server inquiry in one second, which embodies database and gulp down
Spit ability；TPS refers to the number of operations of energy processing business submission affairs in database one second, which includes insertion, and modification is deleted
Except operation；Data connection number refers to that database in sometime all Thread Counts, including active threads and inactive Thread Count, is lived
It is Sleep state that jump Thread Count, which refers to thread state not, and inactive connection number refers to that thread state is the thread of Sleep state
Number.
Step S42: if detecting, the performance indicator of the server cluster is more than capacity threshold, sends sample to monitor supervision platform
This library splits request.
These parameters are detected by preset capacity threshold, which includes QPS threshold value, TPS threshold value, non-
Active threads number threshold value and active threads number threshold value, for example, the QPS threshold value can be set to 20,000 times per second for QPS index；
For TPS index, which is set as 2000 times per second；For database connection number, which can be set
50 are set to, which can be set to 2000.By detecting one section to the server set pocket transmission heartbeat packet
In time, for example, every 10 seconds, 15 seconds or 1 minute etc., in the server cluster, calculates separately the per second of indices and put down
Mean value.It by the average value of the indices, is compared with the capacity threshold, judges whether to need to split sample database
Dilatation is carried out with to the server cluster.
Further, if the result of heartbeat packet feedback is, in the capacity threshold, at least one exceeds capacity threshold, then
To client issue sample database split request, client receive the sample database split request after, according to actual use situation,
The mode and condition for splitting sample database are set, and the mode of setting and condition are sent to server cluster.
Step S43: if getting sample database splitting condition information, sample database is split, obtains sample word bank.
In the present embodiment, after sample database splitting condition information refers to that sample database fractionation request obtains response, to sample
The condition that library is split is configured the information of composition, and the server cluster is according to the setting as a result, splitting to sample database.
Step S44: characteristic value is extracted from sample word bank, obtains characteristic value library.
Specifically, in order to reduce in identification process, extraction sample, will to I/O mouthfuls of consumption and the speed of quickening extraction
Characteristic value in sample is extracted from sample word bank, is stored in preset characteristic value library, and corresponding in sample word bank.Wherein,
In recognition of face, sample is true collected face picture, and characteristic value is by the algorithm in face recognition technology, to people
The feature of face extracts, and the characteristic value calculated, and this feature value is the data that are compared in recognition of face.
Step S45: if after extracting characteristic value library, detection service device cluster resource loading condition, if detecting, resource is negative
Carrying situation is more than preset threshold value, then sends and increase server cluster request.
Specifically, resource load situation refers to that computer refers to the CPU of server cluster, memory, I/O, the resources such as network interface card
Utilization rate.Server cluster refers to the server-side identified using the sample word bank and characteristic value library to monitoring video.Increase
Add server cluster request to refer to that request increases server, the message of dilatation is carried out to the server cluster.
Preferably, it is split to sample database, and after obtaining characteristic value library, it can be negative with detection service device cluster resource
Carry situation.It can specifically be detected by linux system, for example, cpu can be looked by top, sar-u and sar-d
It sees；Memory can check that general memory utilization rate is no more than 90% by free-m and vmstat；Io can pass through iostat
- x and iotop checks that general utilization rate is no more than 90%.Network interface card is checked by netstat-i.
It is possible to further corresponding threshold value be arranged to each resource, if detecting according to above-mentioned resource load situation
The utilization rate for having resource is more than the threshold value, then issues to the server cluster and increase server cluster request.
In one embodiment, in step S44, i.e., characteristic value is extracted from sample word bank, obtains characteristic value library,
Specifically comprise the following steps:
Step S441: according to the quantity of sample word bank, characteristic value library is established.
Specifically, according to the quantity of sample word bank, the characteristic value library equal with the quantity is established, this feature value library is to be used for
Store the characteristic value in sample word bank.Characteristic value library and sample word bank correspond, and use the unique identification of sample word bank, with
The corresponding characteristic value library of the sample word bank is labeled.
Step S442: extracting characteristic value from sample word bank, is stored in the list of feature values, and the list of feature values is stored in corresponding spy
Value indicative library.
Specifically, with the data form in sample word bank for main table, characteristic value sublist is created, by personal information and characteristic value
It extracts into the sublist, and stores to characteristic value library；Finally, change extracts the data of data during identifying monitoring video
Data are extracted from original sample database in library, are changed to extract data from characteristic value library.
Step S443: being monitored the sample size in sample word bank, if detecting, sample size is more than preset threshold
Value, then further split sample word bank and corresponding characteristic value library.
In the present embodiment, sample size refers to the sample size in sample word bank.
Specifically, after being updated every time to sample word bank and corresponding characteristic value library, to the sample in the sample word bank
This capacity is monitored, if detecting, the sample size is more than preset threshold value, is further split to sample word bank.
The condition of the fractionation, which can be, carries out average mark to sample word bank and corresponding characteristic value library, for example, the sample size is more than pre-
If threshold value sample word bank in have 1,000,000 samples, this 1,000,000 samples and corresponding characteristic value are divided into 10 parts,
Every part has 100,000 samples, and every portion sample is stored to a database.
In one embodiment, as shown in figure 4, in step s 30, i.e., analysis identification carried out to monitoring video, and according to point
The result of analysis identification is detected, and is obtained testing result, is specifically comprised the following steps:
Step S301: identifying the face picture in the monitoring video, according to the face picture, using the monitoring video as
Detect video.
Specifically, it by identifying the face picture in monitoring video, and then the face picture by identifying, identifies
Portrait corresponding with the face picture.And there will be the monitoring video of portrait as detection screen.
Step S302: the detection video input to video surveillance model is detected, the testing result is obtained.
In the present embodiment, whether video detection model refers to trains in advance, go out in surveillance video for identification
The model for event of now breaking laws and commit crime.
Specifically, the surveillance video that will acquire is input in video detection model and carries out recognition detection.By this
Video detection model is monitored the behavior of the personnel in surveillance video, obtains the testing result.
In one embodiment, as shown in figure 5, in step s 302, i.e., by the detection video input to video surveillance mould
Type is detected, and is obtained the testing result, is specifically comprised the following steps:
Step S3021: according to history monitoring video, it regard history theft as Sample video.
History monitoring video refers to by the corresponding video for storing different larcenies.Sample video refers in history
It extracts and obtains in monitoring video, and the video of the sample for being trained, a Sample video correspond to a kind of larceny.It can
To understand that ground, the present embodiment are the history history monitoring videos using known larceny as Sample video, and then to this
A little Sample videos are trained, and obtain the model for larceny whether occur that can be used for detecting in one section of video.
Specifically, according to above-mentioned history larceny, several corresponding videos are obtained from history monitoring video, as
Sample video.It should be noted that in order to reach the precision for training the model come, the history of available different angle expires
The video of behavior, as Sample video.
Step S3022: choosing continuous N frame monitored picture from the Sample video one by one, and according to the Sample video into
Row grouping, obtains the training image collection, and wherein N is positive integer.
In order to enable the model trained preferably to detect whether larceny occur in video, regarded to sample
When frequency is trained, the continuous N frame monitored picture that larceny consecutive variations can be embodied in Sample video is extracted, as training
Image.For example, stealer in Sample video is close to stolen person, and hand or tool are taken out and be deeply stolen person pocket or
Whole process in person's knapsack, continuous N frame monitored picture (such as 150 frames).
Specifically, from Sample video, according to the sequence that the Sample video plays, the monitoring of continuous N frame picture is chosen
Picture (such as 150 frames), the monitored picture of each frame is extracted, as training image.For example, from the Sample video,
Extract the monitored picture of continuous 150 frame.
Further, to the corresponding all Sample videos of each larceny, the monitored picture of continuous N frame is all extracted
Afterwards, according to Sample video, the monitored picture for extracting continuous N frame from the same Sample video is divided into one group.After grouping,
Using the result of grouping as training image collection.
Step S3023: the training image collection is input in convolutional neural networks MODEL C NN, the training figure
Image set carries out down-sampling, convolution sum pondization operates, and obtains monitoring feature vector corresponding with all training image collections.
Specifically, training image collection is input to convolutional neural networks CNN(Convolutional Neural
Network, CNN) in model, and then convolution is carried out to training image collection by CNN model, obtain all training image collections pair
The monitoring characteristic value answered.
CNN model used by the application motion specifically includes that data input layer, convolutional calculation layer, pond layer
(Pooling Layer).
Wherein, data input layer refers to training image collection is pre-processed after obtained data.
Wherein, convolutional calculation layer refers in convolutional calculation, and the information for generating multichannel is concentrated from training image, is then existed
Each channel discretely carries out convolution operation.Finally the information in all channels is combined to obtain final monitoring mood
Feature description.
Wherein, pond layer is clipped among continuous convolutional calculation layer, for the amount of compressed data and parameter, is reduced quasi-
It closes, i.e., if input is image, the main effect of pond layer is exactly to compress image.The characteristic pattern of input is carried out
Compression, on the one hand makes characteristic pattern become smaller, and simplifies network query function complexity；On the one hand Feature Compression is carried out, main feature, pond are extracted
The main method for changing layer has maximum pond (Max pooling) and is averaged pond (Average pooling), it is preferable that originally mentions
Case uses Max pooling.
MAX-pooling refers to assuming there is N number of channel for each channel(), by the channel's
The pixel value of feature map chooses wherein representative of the maximum value as the channel, so that obtaining a N-dimensional vector indicates.
(channel) can be understood as visual angle, angle in channel.Such as be equally the convolution kernel for extracting boundary characteristic, it can be with
According to the angle extraction boundary of tri- kinds of elements of R, G, B, RGB has different expression in this angle of boundary.
Step S3024: it is trained, is obtained using the monitoring feature vector of the LSTM network to the training image collection
The corresponding video detection model.
Wherein, video detection model refers to is trained training image collection excessively, obtaining to steal for detecting whether existing
The surreptitiously model of behavior.LSTM network refers to shot and long term memory network (Long Short-Term Memory, abbreviation LSTM network),
It is a kind of implementation of Recognition with Recurrent Neural Network (Recurrent Neural Network, abbreviation RNN network), passes through Keras
Interface is connect with CNN model.LSTM model has the function of time memory, by the feature of each frame monitored picture in this present embodiment
There are close ties with the behavior of two frame of front and back sign, thus using long recurrent neural networks model in short-term to the feature extracted into
Row training, to embody the long-term memory ability of data, improves the accuracy rate of model.
It should be understood that the size of the serial number of each step is not meant that the order of the execution order in above-described embodiment, each process
Execution sequence should be determined by its function and internal logic, the implementation process without coping with the embodiment of the present invention constitutes any limit
It is fixed.
Embodiment two:
In one embodiment, a kind of anti-theft communication network alarming apparatus is provided, the anti-theft communication network alarming apparatus and above-mentioned reality
Anti-theft communication network alarming method in example is applied to correspond.As shown in fig. 6, the anti-theft communication network alarming apparatus includes that communication is built
Formwork erection block 10, monitoring module 20, identification module 30 and alarm module 40.Detailed description are as follows for each functional module:
Communication building block 10, for obtaining the quantity of site alarm terminal, and according to the quantity and monitor supervision platform of alarm terminal
Establish network communication protocol；
Monitoring module 20 obtains monitoring video for monitor supervision platform and is monitored in real time to site；
Identification module 30 for carrying out analysis identification to monitoring video, and is detected according to the result of analysis identification, is examined
Survey result；
Alarm module 40, if for reflecting there are suspicious actions from testing result, according to the quantity of alarm terminal, generate and to
Alarm terminal sends alarm signal.
Preferably, anti-theft communication network alarming apparatus further include:
Detection module 41, the performance indicator for detection service device cluster；
Request sending module 42 is split, if for detecting that the performance indicator of the server cluster is more than capacity threshold, to prison
It controls platform and sends sample database fractionation request；
It splits module 43 and obtains sample word bank if splitting for getting sample database splitting condition information to sample database；
Characteristic extracting module 44 obtains characteristic value library for extracting characteristic value from sample word bank；
Message transmission module 45, if for after extracting characteristic value library, detection service device cluster resource loading condition, if detecting
Resource load situation is more than preset threshold value, then sends and increase server cluster request.
Preferably, characteristic extracting module 44 includes:
Characteristic value library setting up submodule 441 establishes characteristic value library for the quantity according to sample word bank；
Sub-module stored 442 is stored in the list of feature values for extracting characteristic value from sample word bank, and by list of feature values deposit pair
The characteristic value library answered；
Sample monitoring submodule 443, for being monitored to the sample size in sample word bank, if detecting, sample size is more than
Preset threshold value then further splits sample word bank and corresponding characteristic value library.
Preferably, identification module 30 includes:
Recognition of face submodule 301, the face picture in monitoring video makees monitoring video according to face picture for identification
To detect video；
Face datection submodule 302 being detected for will test video input to video surveillance model, obtaining testing result.
Preferably, anti-theft communication network alarming apparatus further include:
Sample video obtains module 3021, for regarding history theft as Sample video according to history monitoring video；
Framing module 3022 for choosing continuous N frame monitored picture from Sample video one by one, and is divided according to Sample video
Group obtains training image collection, and wherein N is positive integer；
Image processing module 3023, for training image collection to be input in convolutional neural networks MODEL C NN, to the training image
Collection carries out down-sampling, the operation of convolution sum pondization, obtains monitoring feature vector corresponding with all training image collections；
Model is obtained module 3024 and obtained for being trained using monitoring feature vector of the LSTM network to training image collection
Corresponding video detection model.
Specific restriction about anti-theft communication network alarming apparatus may refer to above for anti-theft communication network alarming
The restriction of method, details are not described herein.Modules in above-mentioned anti-theft communication network alarming apparatus can be fully or partially through
Software, hardware and combinations thereof are realized.Above-mentioned each module can be embedded in the form of hardware or independently of the place in computer equipment
It manages in device, can also be stored in a software form in the memory in computer equipment, in order to which processor calls execution or more
The corresponding operation of modules.
Embodiment three:
In one embodiment, a kind of computer equipment is provided, which can be server, internal structure chart
It can be as shown in Figure 7.The computer equipment includes processor, memory, network interface and the data connected by system bus
Library.Wherein, the processor of the computer equipment is for providing calculating and control ability.The memory of the computer equipment includes non-
Volatile storage medium, built-in storage.The non-volatile memory medium is stored with operating system, computer program and database.
The built-in storage provides environment for the operation of operating system and computer program in non-volatile memory medium.The computer is set
Standby database is for storing history monitoring video.The network interface of the computer equipment is used to pass through network with external terminal
Connection communication.To realize a kind of anti-theft communication network alarming method when the computer program is executed by processor.
In one embodiment, a kind of computer equipment is provided, including memory, processor and storage are on a memory
And the computer program that can be run on a processor, processor perform the steps of when executing computer program
Step S10: the quantity of site alarm terminal is obtained, and network communication is established according to the quantity and monitor supervision platform of alarm terminal
Agreement；
Step S20: monitor supervision platform obtains monitoring video and is monitored in real time to site；
Step S30: analysis identification is carried out to monitoring video, and is detected according to the result of analysis identification, testing result is obtained；
Step S40: it if reflecting there are suspicious actions from testing result, according to the quantity of alarm terminal, generates and whole to alarm
End sends alarm signal.
Example IV:
In one embodiment, a kind of computer readable storage medium is provided, computer program, computer journey are stored thereon with
It is performed the steps of when sequence is executed by processor
Step S10: the quantity of site alarm terminal is obtained, and network communication is established according to the quantity and monitor supervision platform of alarm terminal
Agreement；
Step S20: monitor supervision platform obtains monitoring video and is monitored in real time to site；
Step S30: analysis identification is carried out to monitoring video, and is detected according to the result of analysis identification, testing result is obtained；
Step S40: it if reflecting there are suspicious actions from testing result, according to the quantity of alarm terminal, generates and whole to alarm
End sends alarm signal.
Those of ordinary skill in the art will appreciate that realizing all or part of the process in above-described embodiment method, being can be with
Relevant hardware is instructed to complete by computer program, the computer program can be stored in a non-volatile computer
In read/write memory medium, the computer program is when being executed, it may include such as the process of the embodiment of above-mentioned each method.Wherein,
To any reference of memory, storage, database or other media used in each embodiment provided herein,
Including non-volatile and/or volatile memory.Nonvolatile memory may include read-only memory (ROM), programming ROM
(PROM), electrically programmable ROM(EPROM), electrically erasable ROM(EEPROM) or flash memory.Volatile memory may include
Random-access memory (ram) or external cache.By way of illustration and not limitation, RAM is available in many forms,
Such as static state RAM(SRAM), dynamic ram (DRAM), synchronous dram (SDRAM), double data rate sdram (DDRSDRAM), enhancing
Type SDRAM(ESDRAM), synchronization link (Synchlink) DRAM(SLDRAM), memory bus (Rambus) direct RAM
(RDRAM), direct memory bus dynamic ram (DRDRAM) and memory bus dynamic ram (RDRAM) etc..
It is apparent to those skilled in the art that for convenience of description and succinctly, only with above-mentioned each function
Can unit, module division progress for example, in practical application, can according to need and by above-mentioned function distribution by different
Functional unit, module are completed, i.e., the internal structure of described device is divided into different functional unit or module, more than completing
The all or part of function of description.
Embodiment described above is merely illustrative of the technical solution of the present invention, rather than its limitations；Although referring to aforementioned reality
Applying example, invention is explained in detail, those skilled in the art should understand that: it still can be to aforementioned each
Technical solution documented by embodiment is modified or equivalent replacement of some of the technical features；And these are modified
Or replacement, the spirit and scope for technical solution of various embodiments of the present invention that it does not separate the essence of the corresponding technical solution should all
It is included within protection scope of the present invention.
Claims (10)
1. a kind of anti-theft communication network alarming method, which is characterized in that the anti-theft communication network alarming method includes following step
It is rapid:
Step S10: the quantity of site alarm terminal is obtained, and network is established according to the quantity and monitor supervision platform of the alarm terminal
Communication protocol；
Step S20: the monitor supervision platform obtains monitoring video and is monitored in real time to site；
Step S30: analysis identification is carried out to the monitoring video, and is detected according to the result of analysis identification, is detected
As a result；
Step S40: it if reflecting there are suspicious actions from the testing result, according to the quantity of the alarm terminal, generates simultaneously
Alarm signal is sent to the alarm terminal.
2. anti-theft communication network alarming method as described in claim 1, which is characterized in that described before the step S40
Anti-theft communication network alarming method is further comprising the steps of:
Step S41: the performance indicator of detection service device cluster；
Step S42: if detecting, the performance indicator of the server cluster is more than capacity threshold, sends sample to the monitor supervision platform
This library splits request；
Step S43: if getting the sample database splitting condition information, the sample database is split, obtains sample word bank；
Step S44: characteristic value is extracted from the sample word bank, obtains the characteristic value library；
Step S45: if after extracting the characteristic value library, detection service device cluster resource loading condition, if detecting, resource is negative
Carrying situation is more than preset threshold value, then sends and increase server cluster request.
3. anti-theft communication network alarming method as claimed in claim 2, which is characterized in that the step S44 includes following step
It is rapid:
Step S441: according to the quantity of the sample word bank, the characteristic value library is established；
Step S442: extracting the characteristic value from the sample word bank, is stored in the list of feature values, and the list of feature values is stored in and is corresponded to
Characteristic value library；
Step S443: being monitored the sample size in the sample word bank, if detecting, the sample size is more than default
Threshold value, then the sample word bank and the corresponding characteristic value library are further split.
4. anti-theft communication network alarming method as described in claim 1, which is characterized in that the step S30 includes:
Step S301: identifying the face picture in the monitoring video, according to the face picture, using the monitoring video as
Detect video；
Step S302: the detection video input to video surveillance model is detected, the testing result is obtained.
5. anti-theft communication network alarming method as claimed in claim 4, which is characterized in that before the step S302, institute
State anti-theft communication network alarming method further include:
Step S3021: according to history monitoring video, it regard history theft as Sample video；
Step S3022: continuous N frame monitored picture is chosen from the Sample video one by one, and is divided according to the Sample video
Group obtains the training image collection, and wherein N is positive integer；
Step S3023: the training image collection is input in convolutional neural networks MODEL C NN, the training image collection
Carry out down-sampling, convolution sum pondization operates, obtain monitoring feature vector corresponding with all training image collections；
Step S3024: it is trained, is corresponded to using the monitoring feature vector of the LSTM network to the training image collection
The video detection model.
6. a kind of anti-theft communication network alarming apparatus, which is characterized in that the anti-theft communication network alarming apparatus includes:
Communication building block, for obtaining the quantity of site alarm terminal, and it is flat according to the quantity of the alarm terminal and monitoring
Platform establishes network communication protocol；
Monitoring module obtains monitoring video for the monitor supervision platform and is monitored in real time to site；
Identification module for carrying out analysis identification to the monitoring video, and is detected according to the result of analysis identification, is obtained
Testing result；
Alarm module, if for reflecting there are suspicious actions from the testing result, it is raw according to the quantity of the alarm terminal
Alarm terminal described in Cheng Bingxiang sends alarm signal.
7. anti-theft communication network alarming apparatus as claimed in claim 6, which is characterized in that the anti-theft communication network alarming dress
It sets further include:
Detection module, the performance indicator for detection service device cluster；
Request sending module is split, if the performance indicator for detecting the server cluster is more than capacity threshold, Xiang Suoshu
Monitor supervision platform sends sample database and splits request；
Module is split, if splitting to the sample database for getting the sample database splitting condition information, obtaining sample
Word bank；
Characteristic extracting module obtains the characteristic value library for extracting characteristic value from the sample word bank；
Message transmission module, if for after extracting the characteristic value library, detection service device cluster resource loading condition, if detection
It is more than preset threshold value to resource load situation, then sends and increase server cluster request.
8. anti-theft communication network alarming apparatus as claimed in claim 7, which is characterized in that the characteristic extracting module includes:
Characteristic value library setting up submodule establishes the characteristic value library for the quantity according to the sample word bank；
Sub-module stored is stored in the list of feature values, and the list of feature values is deposited for extracting the characteristic value from the sample word bank
Enter corresponding characteristic value library；
Sample monitoring submodule, for being monitored to the sample size in the sample word bank, if detecting, the sample holds
Amount is more than preset threshold value, then is further split to the sample word bank and the corresponding characteristic value library.
9. a kind of computer equipment, including memory, processor and storage are in the memory and can be in the processor
The computer program of upper operation, which is characterized in that the processor realized when executing the computer program as claim 1 to
The step of any one of 5 anti-theft communication network alarming method.
10. a kind of computer readable storage medium, the computer-readable recording medium storage has computer program, and feature exists
In realization anti-theft communication network alarming side as described in any one of claim 1 to 5 when the computer program is executed by processor
The step of method.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910065505.9A CN109697815A (en) | 2019-01-24 | 2019-01-24 | Anti-theft communication network alarming method, appliance arrangement and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910065505.9A CN109697815A (en) | 2019-01-24 | 2019-01-24 | Anti-theft communication network alarming method, appliance arrangement and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109697815A true CN109697815A (en) | 2019-04-30 |
Family
ID=66234280
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910065505.9A Pending CN109697815A (en) | 2019-01-24 | 2019-01-24 | Anti-theft communication network alarming method, appliance arrangement and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109697815A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111582179A (en) * | 2020-05-09 | 2020-08-25 | 广州市炬盾科技发展有限公司 | Monitoring video analysis method and device, computer equipment and storage medium |
| CN111797972A (en) * | 2020-05-27 | 2020-10-20 | 北京迈格威科技有限公司 | Method, device and electronic system for processing data by using convolutional neural network |
| CN111950494A (en) * | 2020-08-20 | 2020-11-17 | 宁波广播电视集团 | Monitoring method, device and system based on convolutional neural network and readable storage medium |
| US11055518B2 (en) * | 2019-08-05 | 2021-07-06 | Sensormatic Electronics, LLC | Methods and systems for monitoring potential losses in a retail environment |
| CN113836604A (en) * | 2021-08-27 | 2021-12-24 | 太原市高远时代科技有限公司 | Case with anti-theft video recording function |
Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102881100A (en) * | 2012-08-24 | 2013-01-16 | 济南纳维信息技术有限公司 | Video-analysis-based antitheft monitoring method for physical store |
| CN103401705A (en) * | 2013-07-24 | 2013-11-20 | 佳都新太科技股份有限公司 | Centralized memory management method in video monitoring and control network |
| CN104111924A (en) * | 2013-04-16 | 2014-10-22 | 中国移动通信集团广东有限公司 | Database system |
| KR101675692B1 (en) * | 2015-06-26 | 2016-11-14 | 성균관대학교산학협력단 | Method and apparatus for crowd behavior recognition based on structure learning |
| CN206042227U (en) * | 2016-08-31 | 2017-03-22 | 深圳市超视科技有限公司 | Distributing type video monitoring platform |
| CN107644519A (en) * | 2017-10-09 | 2018-01-30 | 中电科新型智慧城市研究院有限公司 | A kind of intelligent alarm method and system based on video human Activity recognition |
| US20180173956A1 (en) * | 2016-12-21 | 2018-06-21 | Axis Ab | Method for identifying events in a motion video |
| CN108229338A (en) * | 2017-12-14 | 2018-06-29 | 华南理工大学 | A kind of video behavior recognition methods based on depth convolution feature |
| CN108351968A (en) * | 2017-12-28 | 2018-07-31 | 深圳市锐明技术股份有限公司 | It is a kind of for the alarm method of criminal activity, device, storage medium and server |
| CN108446649A (en) * | 2018-03-27 | 2018-08-24 | 百度在线网络技术（北京）有限公司 | Method and device for alarm |
| CN108600322A (en) * | 2018-03-27 | 2018-09-28 | 通号通信信息集团有限公司 | Integrated video monitoring cloud service system and method based on railway video technical specification |
| US20180357247A1 (en) * | 2017-06-07 | 2018-12-13 | Amazon Technologies, Inc. | Behavior-aware security systems and associated methods |
-
2019
- 2019-01-24 CN CN201910065505.9A patent/CN109697815A/en active Pending
Patent Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102881100A (en) * | 2012-08-24 | 2013-01-16 | 济南纳维信息技术有限公司 | Video-analysis-based antitheft monitoring method for physical store |
| CN104111924A (en) * | 2013-04-16 | 2014-10-22 | 中国移动通信集团广东有限公司 | Database system |
| CN103401705A (en) * | 2013-07-24 | 2013-11-20 | 佳都新太科技股份有限公司 | Centralized memory management method in video monitoring and control network |
| KR101675692B1 (en) * | 2015-06-26 | 2016-11-14 | 성균관대학교산학협력단 | Method and apparatus for crowd behavior recognition based on structure learning |
| CN206042227U (en) * | 2016-08-31 | 2017-03-22 | 深圳市超视科技有限公司 | Distributing type video monitoring platform |
| US20180173956A1 (en) * | 2016-12-21 | 2018-06-21 | Axis Ab | Method for identifying events in a motion video |
| US20180357247A1 (en) * | 2017-06-07 | 2018-12-13 | Amazon Technologies, Inc. | Behavior-aware security systems and associated methods |
| CN107644519A (en) * | 2017-10-09 | 2018-01-30 | 中电科新型智慧城市研究院有限公司 | A kind of intelligent alarm method and system based on video human Activity recognition |
| CN108229338A (en) * | 2017-12-14 | 2018-06-29 | 华南理工大学 | A kind of video behavior recognition methods based on depth convolution feature |
| CN108351968A (en) * | 2017-12-28 | 2018-07-31 | 深圳市锐明技术股份有限公司 | It is a kind of for the alarm method of criminal activity, device, storage medium and server |
| CN108446649A (en) * | 2018-03-27 | 2018-08-24 | 百度在线网络技术（北京）有限公司 | Method and device for alarm |
| CN108600322A (en) * | 2018-03-27 | 2018-09-28 | 通号通信信息集团有限公司 | Integrated video monitoring cloud service system and method based on railway video technical specification |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US11055518B2 (en) * | 2019-08-05 | 2021-07-06 | Sensormatic Electronics, LLC | Methods and systems for monitoring potential losses in a retail environment |
| CN111582179A (en) * | 2020-05-09 | 2020-08-25 | 广州市炬盾科技发展有限公司 | Monitoring video analysis method and device, computer equipment and storage medium |
| CN111582179B (en) * | 2020-05-09 | 2023-04-07 | 广州市炬盾科技发展有限公司 | Monitoring video analysis method and device, computer equipment and storage medium |
| CN111797972A (en) * | 2020-05-27 | 2020-10-20 | 北京迈格威科技有限公司 | Method, device and electronic system for processing data by using convolutional neural network |
| CN111950494A (en) * | 2020-08-20 | 2020-11-17 | 宁波广播电视集团 | Monitoring method, device and system based on convolutional neural network and readable storage medium |
| CN111950494B (en) * | 2020-08-20 | 2024-01-16 | 宁波广播电视集团 | Monitoring method, device and system based on convolutional neural network and readable storage medium |
| CN113836604A (en) * | 2021-08-27 | 2021-12-24 | 太原市高远时代科技有限公司 | Case with anti-theft video recording function |
| CN113836604B (en) * | 2021-08-27 | 2024-03-01 | 太原市高远时代科技有限公司 | Case with anti-theft video recording function |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109697815A (en) | Anti-theft communication network alarming method, appliance arrangement and storage medium | |
| JP5747116B1 (en) | Security system | |
| EP3407232B1 (en) | Spatiotemporal authentication | |
| WO2018194814A1 (en) | Object analysis in live video content | |
| US20150047006A1 (en) | Method for Authenticating Identity of Handset User | |
| CN108806153A (en) | Alert processing method, apparatus and system | |
| CN107222660A (en) | A kind of distributed network visual monitor system | |
| CN112464030B (en) | Suspicious person determination method and suspicious person determination device | |
| US11348415B2 (en) | Cognitive automation platform for providing enhanced automated teller machine (ATM) security | |
| CN111191507A (en) | Safety early warning analysis method and system for smart community | |
| CN110119730A (en) | A kind of monitor video processing method, system, terminal and storage medium | |
| CN115187439A (en) | Comprehensive operation management method and system for intelligent park | |
| CN111091047A (en) | Living body detection method and device, server and face recognition equipment | |
| KR20110104457A (en) | Remote security control system using cloud computing | |
| CN110659087A (en) | Face recognition algorithm engineering system applied to intelligent medical treatment | |
| CN113645439B (en) | Event detection method and system, storage medium and electronic device | |
| CN109815828A (en) | Realize the system and method for initiative alarming or help-seeking behavior detection control | |
| CN109120896A (en) | Security protection video monitors bodyguard's system | |
| CN111541864B (en) | Digital retina software defined camera method and system | |
| CN113988867A (en) | Fraud detection method and device, computer equipment and storage medium | |
| US11361630B1 (en) | Identifying and logging mobile devices posing security threats | |
| CN105959329B (en) | High-definition video superposition processing system | |
| CN111783594A (en) | Alarm method and device and electronic equipment | |
| US20190057589A1 (en) | Cloud based systems and methods for locating a peace breaker | |
| JP2004199468A (en) | Seal impression collation system and its controlling program |
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: 20190430 |
|
| RJ01 | Rejection of invention patent application after publication |