CN108012121A - A kind of edge calculations and the real-time video monitoring method and system of cloud computing fusion - Google Patents
A kind of edge calculations and the real-time video monitoring method and system of cloud computing fusion Download PDFInfo
- Publication number
- CN108012121A CN108012121A CN201711340896.8A CN201711340896A CN108012121A CN 108012121 A CN108012121 A CN 108012121A CN 201711340896 A CN201711340896 A CN 201711340896A CN 108012121 A CN108012121 A CN 108012121A
- Authority
- CN
- China
- Prior art keywords
- image
- real
- training
- neural networks
- layer
- 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
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N7/00—Television systems
- H04N7/18—Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/40—Scenes; Scene-specific elements in video content
- G06V20/41—Higher-level, semantic clustering, classification or understanding of video scenes, e.g. detection, labelling or Markovian modelling of sport events or news items
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/40—Scenes; Scene-specific elements in video content
- G06V20/46—Extracting features or characteristics from the video content, e.g. video fingerprints, representative shots or key frames
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Computational Linguistics (AREA)
- Software Systems (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses the real-time video monitoring method that a kind of edge calculations and cloud computing merge, including：The training storehouse of sample is built, trains CNN convolutional neural networks using the image information in sample training storehouse, training chip is embedded in front edge equipment by structure CNN convolutional neural networks model structure as training chip content；Extract real-time edge device monitor video, multiple image is decomposed into by original video；Chip is trained to carry out feature and Similarity measures to multiple image by CNN convolutional neural networks；By the similitude picture and information that extract storage to long-range cloud device, long term backup management, there is provided calculating is used and analyzed to supervision department, message push relevant departments unit is implemented to alarm to relevant departments.The invention also discloses a kind of real-time video monitoring system.The advantage of the invention is that：Solve Cloud Server mass memory in real time to picture processing using CNN convolutional neural networks image recognitions retrieval technique at edge device end and calculate pressure.
Description
Technical field
The present invention relates to the real-time prison in a kind of edge calculations field, field of cloud calculation and Image Processing and Pattern Recognition field
Control method and system.
Background technology
In nearest decades, video monitoring system has become a kind of even national essential management in city
Means, administrative staff, which need not be in the action, can grasp live information, reduce the expense of man power and material.Pass through monitoring
System can improve the effect of supervision and oversight, while reduce the probability of the generation of major accident.But traditional monitoring is still
Need a large amount of manpowers be compared, classify to image, identify monitor in real time, cause monitoring result accuracy and precision compared with
Difference, while huge pressure is brought to storage device.
As machine learning is risen, the application of neutral net is also more and more extensive.Some neutral nets that we recognize
Algorithm, it is intended to which the computation model of simulation biology study, successfully application also allows people to recognize its unique evil spirit in every field for it
Power.And the one kind of convolutional neural networks as neutral net, it is the neutral net for being specifically used to handle similar grid data, especially
It is to have significant effect in terms of the classification in image, identification, similitude and feature calculation.
Reach its maturity however as cloud computing development, the space limitation of conventional store mode is broken.Cloud computing is to use
The functions such as Clustering, distributed file system, pass through application software by a large amount of various types of storage devices in network
Collaborative work is gathered, the common system that data storage and Operational Visit function are externally provided.
The content of the invention
The technical problems to be solved by the invention are that video monitoring system edge device end in real time cannot carry out picture
The problem of processing, Cloud Server mass memory and calculating pressure.
The technical solution adopted by the present invention to solve the technical problems is：Provide a kind of edge calculations and cloud computing fusion
Real-time video monitoring method, include the following steps：
Step S1, builds the training storehouse of sample, and CNN convolutional Neural nets are trained using the image information in sample training storehouse
Training chip is embedded in front edge equipment by network, structure CNN convolutional neural networks model structure as training chip content；
Step S2, extract real-time front edge monitoring of tools video, multiple image is decomposed into by original video；
Step S3, trains chip to test multiple image, extracts the feature of image by CNN convolutional neural networks
And similitude；
Step S4, by the similitude picture and information that extract storage to long-range cloud device, long term backup management, there is provided
Calculating is used and analyzed to supervision department, and message push relevant departments unit is implemented to alarm to relevant departments.
Further as above-mentioned technical proposal is improved, and supervision department provides corresponding monitoring image information in the present invention,
As the sample storehouse of CNN convolutional neural networks, CNN convolutional neural networks model structures are built.
Further as above-mentioned technical proposal is improved, and convolutional neural networks model structure includes：
(1) input layer：Input layer is the image of multiple training samples obtained；
(2) convolutional layer：By the image in input layer, convolution behaviour is carried out to image by the filter filter in convolutional layer
Make, the characteristics of using local sensing and parameter sharing, extract the various features of image；
(3) excitation layer：Excitation function Relu is added after each convolutional layer, adds non-linear factor processing；
(4) pond layer：Processing is compressed to the characteristic pattern of input, pond layer is inserted among continuous convolutional layer；
(5) full articulamentum：All features are connected, give output valve to softmax graders；
(6) output layer：The characteristic of training sample is extracted by classification and similitude is used as the result of output.
Further as above-mentioned technical proposal is improved, and step S2 is specifically included：
Video in extract real-time monitoring device, extraction wherein have facial image part, the video extracted by
Decomposed according to 5 frame per second, choose the wherein highest image of clarity, the test set figure as CNN convolutional neural networks
Picture.
Further as above-mentioned technical proposal is improved, and step S4 is specifically included：
Step S41, the video that front edge equipment obtains, extracts image, voice and HD video as desired, and right
These information are encrypted, and are transmitted to long-range cloud server；
The data received are carried out polytype calculating by step S42, long-range high in the clouds；
Step S43, safeguards data in Cloud Server；
Step S44, the information data that analysis calculates is transmitted directly to regulator unit by high in the clouds, and is alarmed.
Present invention also offers the real-time video monitoring system that a kind of edge calculations and cloud computing merge, including：
CNN convolutional neural networks training modules：The training storehouse of sample is built, is instructed using the image information in sample training storehouse
Practice CNN convolutional neural networks, training chip is embedded in by structure CNN convolutional neural networks model structure as training chip content
In front edge equipment；
Front edge computing module：Extract real-time front edge monitoring of tools video, multiframe figure is resolved into by original video
Picture, by CNN convolutional neural networks intelligent chips, tests multiple image, extracts the feature and similitude of image；
Rear end cloud storage module：By the similitude picture and information that extract storage to long-range cloud device, long term backup
Management, carries out complicated calculations, push alarm, supervision department is transmitted to by the valuable information of tool for analyzing next.
Further as above-mentioned technical proposal is improved, and in CNN convolutional neural networks training modules, supervision department provides phase
The monitoring image information answered, as the sample storehouse of CNN convolutional neural networks, builds CNN convolutional neural networks model structures.
Further as above-mentioned technical proposal is improved, and convolutional neural networks model structure includes：
(1) input layer：Input layer is the image of multiple training samples obtained；
(2) convolutional layer：By the image in input layer, convolution behaviour is carried out to image by the filter filter in convolutional layer
Make, the characteristics of using local sensing and parameter sharing, extract the various features of image；
(3) excitation layer：Excitation function Relu is added after each convolutional layer, adds non-linear factor processing；
(4) pond layer：Processing is compressed to the characteristic pattern of input, pond layer is inserted among continuous convolutional layer；
(5) full articulamentum：All features are connected, give output valve to softmax graders；
(6) output layer：The characteristic of training sample is extracted by classification and similitude is used as the result of output.
Further as above-mentioned technical proposal is improved, and in front edge computing module, is regarded in extract real-time monitoring device
Frequently, extraction wherein has the part of facial image, and the video extracted is decomposed according to 5 frame per second, chooses wherein clear
The clear highest image of degree, the test set image as CNN convolutional neural networks.
Further as above-mentioned technical proposal is improved, in the cloud storage module of rear end, including following units：
Data storage cell, for the video for obtaining front edge equipment, extracts image, voice and height as desired
Clear video, and these information are encrypted, and it is transmitted to long-range cloud server；
Distributed Calculation unit, polytype calculating is carried out for long-distance cloud to be terminated received data；
Data maintenance unit, for being safeguarded to data in Cloud Server；
Alarm unit, for being transmitted directly to regulator unit by the information data calculated is analyzed by high in the clouds, and
Alarm.
Compared with prior art, beneficial effect possessed by the present invention is：
Efficiently solve the intelligent processing capacity of video monitoring system front end camera, so realize major criminal cases and
Attack of terrorism activity early warning system and handling device, improve the energy for taking precautions against criminal offense and the attack of terrorism of video monitoring system
Power so that video monitoring system has stronger Practical significance, has broad application prospects in public safety industry, subtracts in addition
The workload of Shao Liao supervision departments, improves work efficiency, meets demand of the supervision department to real-time monitoring system higher.
Brief description of the drawings
Fig. 1 is the flow signal of a kind of edge calculations of the embodiment of the present invention and the real-time video monitoring method of cloud computing fusion
Figure；
Fig. 2 is the module signal of the real-time video monitoring system of a kind of edge calculations of the embodiment of the present invention and cloud computing fusion
Figure.
Embodiment
The embodiment of the present invention is described in detail below in conjunction with attached drawing.It should be appreciated that this place is retouched
The embodiment stated is merely to illustrate and explain the present invention, and is not intended to limit the invention.
As shown in Figure 1, there is provided a kind of edge calculations and the real-time video monitoring method of cloud computing fusion, including：
Step S1, builds the training storehouse of sample, and CNN convolutional Neural nets are trained using the image information in sample training storehouse
Network, training chip is embedded in front edge equipment.Supervision department provides corresponding monitoring image information, as CNN convolution god
Sample storehouse through network, builds convolutional neural networks model structure, including：
(1) input layer：Input layer is the image for multiple training samples that we obtain.
(2) convolutional layer：By the image in input layer, convolution behaviour is carried out to image by the filter filter in convolutional layer
Make, the characteristics of using local sensing and parameter sharing, extract the various features of image.
(3) pond layer：Processing is compressed to the characteristic pattern of input.Pond layer is inserted among continuous convolutional layer.Pond
Change operation and reduce data volume, reduce parameter, reducing calculating prevents over-fitting.
(4) full articulamentum：All features are connected, give output valve to softmax graders.
(5) output layer：The characteristic of training sample is extracted by classification and similitude is used as the result of output.
In specific scheme, training before to image carry out pretreatment operation, such as using gray proces, go average,
Coloured image is switched to gray level image by the processing of PCA dimensionality reductions degree, ash operation:I=W1 × R+W2 × G+W3 × B, wherein I are ashing
Image afterwards, R be coloured image red component, G be coloured image green component, B be coloured image blue component, W1
For the weight of the red component of coloured image, W2 is the weight of the green component of coloured image, and W3 is the blueness point of coloured image
The weight of amount.The averaging operation that goes of image is that ashed image is carried out to subtract average value processing, makes sample image for center
Position, easy to similitude and feature calculation below.After training is completed, neural network model is write into artificial intelligence chip
In, and carry out relevant test.Finally artificial intelligence chip is embedded in front edge equipment, screening and mistake to similar image
Suspect image information, is transmitted to long-range high in the clouds and stored by filter.
Step S2, extract real-time front edge monitoring of tools video, multiple image is resolved into by original video.Use video
Decomposition technique.Detailed process is：
Video in extract real-time monitoring device, extraction wherein have facial image part, the video extracted by
Decomposed according to 5 frame per second, choose the wherein highest image of clarity, the test set figure as CNN convolutional neural networks
Picture.
Step S3, by CNN convolutional neural networks intelligent chips, tests multiple image, extracts the spy of image
Seek peace similitude.Detailed process is：
Convolution operation is carried out to image by the filter filter in convolutional layer, utilizes local sensing and parameter sharing
Feature, extracts the various features of image；Processing is compressed to the characteristic pattern of input, pond is inserted among continuous convolutional layer
Change layer, pondization operation reduces data volume, reduces parameter, prevents from calculating over-fitting；All features are connected, output valve is given
Softmax graders；The feature and similitude of image are extracted by classification.
Step S4, by the similitude picture and information that extract storage to long-range cloud device, long term backup management, carries out
Complicated calculations, push alarm, supervision department is transmitted to by the valuable information of tool for analyzing next.Specifically include：
(1) video that front edge equipment obtains, extracts image, voice and HD video etc. as desired, and to this
A little information are encrypted, and ensure safe and reliable in message transmitting procedure.Pass through the nets such as HTTP, HTTPS, IPv4 and IPv6
Road transport protocol, is transmitted to long-range cloud server.
(2) data received are carried out polytype calculating by long-range high in the clouds.Including the screening to warning message and mistake
Filter and the mode of push alarm, the selection in the place of alarm and the prediction of similar case etc..
(3) data in Cloud Server are safeguarded, ensures the high availability of information.Main contents are guarantee server
Securely and reliably, fire wall is set, fire wall is configured, periodically server is backed up, monitors syslog file, is led to
Cross and report is analyzed, judge abnormal generation.
(4) information data that analysis calculates is transmitted directly to regulator unit by high in the clouds, and is alarmed.Specific side
Case is connection alarm system, and alarm content is shown in alarm system, by supervision unit recipient warning message, downloads relevant regard
Frequently, image and voice information.
As shown in Fig. 2, the present invention provides the real-time video monitoring system that a kind of edge calculations and cloud computing merge, including
Neural metwork training module, front edge computing module, rear end cloud storage module.
Neural metwork training module action is：Step S1, front edge computing module action are：Step S2 and
Step S3, rear end cloud storage module action are：Step S4.
Neural metwork training module, builds the training storehouse of sample, and CNN volumes is trained using the image information in sample training storehouse
Product neutral net, training chip is embedded in front edge equipment.Specifically include：Supervision department provides corresponding monitoring image letter
Breath, as the sample storehouse of CNN convolutional neural networks, builds convolutional neural networks model structure.Including：
(1) input layer：Input layer is the image for multiple training samples that we obtain.
(2) convolutional layer：By the image in input layer, convolution behaviour is carried out to image by the filter filter in convolutional layer
Make, the characteristics of using local sensing and parameter sharing, extract the various features of image.
(3) pond layer：Processing is compressed to the characteristic pattern of input.Pond layer is inserted among continuous convolutional layer.Pond
Change operation and reduce data volume, reduce parameter, reducing calculating prevents over-fitting.
(4) full articulamentum：All features are connected, give output valve to softmax graders.
(5) output layer：The characteristic of training sample is extracted by classification and similitude is used as the result of output.
Front edge computing module, extract real-time front edge monitoring of tools video, multiframe figure is resolved into by original video
Picture, by CNN convolutional neural networks intelligent chips, tests multiple image, extracts the feature and similitude of image.Tool
Body includes：
(1) video decomposition technique is used, video in extract real-time monitoring device, extraction wherein has the portion of facial image
Point, the video extracted is decomposed according to 5 frame per second, the wherein highest image of clarity is chosen, as CNN volumes
The test set image of product neutral net.
(2) convolution operation is carried out to image by the filter filter in convolutional layer, is total to using local sensing and parameter
The characteristics of enjoying, extracts the various features of image；Processing is compressed to the characteristic pattern of input, is interleave in continuous convolutional layer
Enter pond layer, pondization operation reduces data volume, reduces parameter, prevents from calculating over-fitting；All features are connected, by output valve
Give softmax graders；The feature and similitude of image are extracted by classification.
Rear end cloud storage module, including：Data storage cell, Distributed Calculation unit, data maintenance unit, data storage
Unit stores the similitude picture and information that extract to long-range cloud device, long term backup management；Distributed Calculation unit
Carry out complicated calculations；Data maintenance unit safeguards data in Cloud Server, ensures the high availability of information.Alarm is single
Member, supervision department's push alarm is transmitted to for will analyze the valuable information of tool come.
Specifically include：
(1) data storage cell, the video that front edge equipment obtains, extracts image, voice and high definition as desired
Video etc..And these information are encrypted, ensure safe and reliable in message transmitting procedure.By HTTP, HTTPS,
The network transmission agreement such as IPv4 and IPv6, is transmitted to long-range cloud server.
(2) data received are carried out complicated calculations by Distributed Calculation unit, long-range high in the clouds.Including to warning message
Screening and filtering and the mode of push alarm, the selection in the place of alarm and the prediction of similar case etc..
(3) data maintenance unit, safeguards data in Cloud Server, ensures the high availability of information.Including
Ensure the safe and reliable of server, fire wall is set, fire wall is configured, periodically server is backed up, monitoring system
System journal file, by analyzing report, judges abnormal generation.
(4) alarm unit, the information data that analysis calculates is transmitted directly to regulator unit by high in the clouds, and is reported
It is alert.Concrete scheme is connection alarm system, and alarm content is shown in alarm system, supervises unit recipient warning message.Download
Relevant video, image and voice information.
The foregoing is merely illustrative of the preferred embodiments of the present invention, is not intended to limit the invention, all essences in the present invention
All any modification, equivalent and improvement made within refreshing and principle etc., should be included in protection scope of the present invention.
Claims (10)
1. a kind of edge calculations and the real-time video monitoring method of cloud computing fusion, include the following steps：
Step S1, builds the training storehouse of sample, and CNN convolutional neural networks, structure are trained using the image information in sample training storehouse
CNN convolutional neural networks model structure is built as training chip content, training chip is embedded in front edge equipment；
Step S2, extract real-time front edge monitoring of tools video, multiple image is decomposed into by original video；
Step S3, trains chip to test multiple image, extracts the feature and phase of image by CNN convolutional neural networks
Like property；
Step S4, by the similitude picture and information that extract storage to long-range cloud device, long term backup management, there is provided to prison
Pipe portion door uses and analyzes calculating, and message push relevant departments unit is implemented to alarm to relevant departments.
2. a kind of edge calculations according to claim 1 and the real-time video monitoring method of cloud computing fusion, its feature exist
In：In the step S1, supervision department provides corresponding monitoring image information, as the sample storehouse of CNN convolutional neural networks, structure
Build CNN convolutional neural networks model structures.
3. a kind of edge calculations according to claim 2 and the real-time video monitoring method of cloud computing fusion, its feature exist
In：Convolutional neural networks model structure includes：
(1) input layer：Input layer is the image of multiple training samples obtained；
(2) convolutional layer：By the image in input layer, convolution operation is carried out to image by the filter filter in convolutional layer,
The characteristics of using local sensing and parameter sharing, extract the various features of image；
(3) excitation layer：Excitation function Relu is added after each convolutional layer, adds non-linear factor processing；
(4) pond layer：Processing is compressed to the characteristic pattern of input, pond layer is inserted among continuous convolutional layer；
(5) full articulamentum：All features are connected, give output valve to softmax graders；
(6) output layer：The characteristic of training sample is extracted by classification and similitude is used as the result of output.
4. a kind of edge calculations according to claim 1 and the real-time video monitoring method of cloud computing fusion, its feature exist
In：Step S2 is specifically included：Video in extract real-time monitoring device, extraction wherein has the part of facial image, extracting
The video come is decomposed according to 5 frame per second, the wherein highest image of clarity is chosen, as CNN convolutional neural networks
Test set image.
5. a kind of edge calculations according to claim 1 and the real-time video monitoring method of cloud computing fusion, its feature exist
In：Step S4 is specifically included：
Step S41, the video that front edge equipment obtains, extracts image, voice and HD video as desired, and to these
Information is encrypted, and is transmitted to long-range cloud server；
The data received are carried out polytype calculating by step S42, long-range high in the clouds；
Step S43, safeguards data in Cloud Server；
Step S44, the information data that analysis calculates is transmitted directly to regulator unit by high in the clouds, and is alarmed.
6. a kind of edge calculations and the real-time video monitoring system of cloud computing fusion, it is characterised in that：Including：
Neural metwork training module：The training storehouse of sample is built, uses the image information training CNN convolution god in sample training storehouse
Through network, training chip is embedded in front edge equipment by structure CNN convolutional neural networks model structure as training chip content
In；
Front edge computing module：Extract real-time front edge monitoring of tools video, multiple image is resolved into by original video, is led to
CNN convolutional neural networks intelligent chips are crossed, multiple image is tested, extract the feature and similitude of image；
Rear end cloud storage module：The similitude picture and information that extract are stored to long-range cloud device, long term backup management,
Complicated calculations are carried out, push alarm, supervision department is transmitted to by the valuable information of tool for analyzing next.
7. a kind of edge calculations according to claim 6 and the real-time video monitoring system of cloud computing fusion, its feature exist
In：In neural metwork training module, supervision department provides corresponding monitoring image information, the sample as CNN convolutional neural networks
This storehouse, builds CNN convolutional neural networks model structures.
8. a kind of edge calculations according to claim 7 and the real-time video monitoring system of cloud computing fusion, its feature exist
In：
Convolutional neural networks model structure includes：
(1) input layer：Input layer is the image of multiple training samples obtained；
(2) convolutional layer：By the image in input layer, convolution operation is carried out to image by the filter filter in convolutional layer,
The characteristics of using local sensing and parameter sharing, extract the various features of image；
(3) excitation layer：Excitation function Relu is added after each convolutional layer, adds non-linear factor processing；
(4) pond layer：Processing is compressed to the characteristic pattern of input, pond layer is inserted among continuous convolutional layer；
(5) full articulamentum：All features are connected, give output valve to softmax graders；
(6) output layer：The characteristic of training sample is extracted by classification and similitude is used as the result of output.
9. a kind of edge calculations according to claim 6 and the real-time video monitoring system of cloud computing fusion, its feature exist
In：In front edge computing module, video in extract real-time monitoring device, extraction wherein has the part of facial image, carrying
The video taken out is decomposed according to 5 frame per second, the wherein highest image of clarity is chosen, as CNN convolutional Neurals
The test set image of network.
10. a kind of edge calculations according to claim 6 and the real-time video monitoring system of cloud computing fusion, its feature exist
In：In the cloud storage module of rear end, including following units：
Data storage cell, for the video for obtaining front edge equipment, extracts image, voice and high definition and regards as desired
Frequently, and to these information it is encrypted, and is transmitted to long-range cloud server；
Distributed Calculation unit, polytype calculating is carried out for long-distance cloud to be terminated received data；
Data maintenance unit, for being safeguarded to data in Cloud Server；
Alarm unit, for being transmitted directly to regulator unit by the information data calculated is analyzed by high in the clouds, and carries out
Alarm.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711340896.8A CN108012121A (en) | 2017-12-14 | 2017-12-14 | A kind of edge calculations and the real-time video monitoring method and system of cloud computing fusion |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711340896.8A CN108012121A (en) | 2017-12-14 | 2017-12-14 | A kind of edge calculations and the real-time video monitoring method and system of cloud computing fusion |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108012121A true CN108012121A (en) | 2018-05-08 |
Family
ID=62059006
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711340896.8A Pending CN108012121A (en) | 2017-12-14 | 2017-12-14 | A kind of edge calculations and the real-time video monitoring method and system of cloud computing fusion |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108012121A (en) |
Cited By (36)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108416715A (en) * | 2018-05-11 | 2018-08-17 | 长江大学 | A kind of Campus security management system based on intelligent video camera head |
| CN108828409A (en) * | 2018-08-03 | 2018-11-16 | 南方电网科学研究院有限责任公司 | Fault detection system based on edge calculation |
| CN108873712A (en) * | 2018-06-15 | 2018-11-23 | 北京王府科技有限公司 | ECS edge calculations system |
| CN108921866A (en) * | 2018-07-24 | 2018-11-30 | 北京深瞐科技有限公司 | A kind of image processing method and system |
| CN109005390A (en) * | 2018-08-31 | 2018-12-14 | 山东建筑大学 | Personnel's distributed model method for building up and system based on signal strength and video |
| CN109194926A (en) * | 2018-10-19 | 2019-01-11 | 济南浪潮高新科技投资发展有限公司 | A kind of city security system and its detection method based on edge calculations |
| CN109347644A (en) * | 2018-12-06 | 2019-02-15 | 杭州环普数据技术有限公司 | A kind of block chain edge calculations device |
| CN109670544A (en) * | 2018-12-13 | 2019-04-23 | 广州小狗机器人技术有限公司 | A kind of object detecting apparatus and its acquisition methods, object detecting system |
| CN109685202A (en) * | 2018-12-17 | 2019-04-26 | 腾讯科技（深圳）有限公司 | Data processing method and device, storage medium and electronic device |
| CN109767594A (en) * | 2019-01-30 | 2019-05-17 | 长视科技股份有限公司 | A kind of household monitor video intelligent linkage early warning system |
| CN110062208A (en) * | 2019-04-23 | 2019-07-26 | 上海赫煊自动化系统工程有限公司 | A kind of security protection intelligence real-time analyzer and method |
| CN110197128A (en) * | 2019-05-08 | 2019-09-03 | 华南理工大学 | The recognition of face architecture design method planned as a whole based on edge calculations and cloud |
| CN110209140A (en) * | 2019-07-09 | 2019-09-06 | 深圳市华腾物联科技有限公司 | A kind of application method and equipment of edge calculations |
| CN110536299A (en) * | 2019-09-04 | 2019-12-03 | 电子科技大学 | Data packet validity confirmation method based on edge calculation and discrete random convolution |
| CN110581898A (en) * | 2019-10-09 | 2019-12-17 | 成都康胜思科技有限公司 | internet of things data terminal system based on 5G and edge calculation |
| CN110738309A (en) * | 2019-09-27 | 2020-01-31 | 华中科技大学 | DDNN training method and DDNN-based multi-view target identification method and system |
| CN111160321A (en) * | 2020-02-10 | 2020-05-15 | 杭州大数云智科技有限公司 | Storage battery car goes up terraced detection and early warning system |
| CN111199279A (en) * | 2019-10-30 | 2020-05-26 | 山东浪潮人工智能研究院有限公司 | Cloud edge calculation and artificial intelligence fusion method and device for police service industry |
| CN111273071A (en) * | 2020-02-17 | 2020-06-12 | 深圳供电局有限公司 | AI monitoring system and method for edge calculation |
| CN111333010A (en) * | 2020-05-19 | 2020-06-26 | 八维通科技有限公司 | Method and system for preventing oil adding error of gas station based on Bluetooth positioning |
| CN111382836A (en) * | 2018-12-30 | 2020-07-07 | 罗伯特·博世有限公司 | Distributed neural network for edge devices |
| CN111479048A (en) * | 2020-04-22 | 2020-07-31 | 安徽大学 | Intelligent video image processing equipment based on edge calculation |
| CN111565303A (en) * | 2020-05-29 | 2020-08-21 | 深圳市易链信息技术有限公司 | Video monitoring method, system and readable storage medium based on fog calculation and deep learning |
| CN111753597A (en) * | 2019-03-29 | 2020-10-09 | 中国安全生产科学研究院 | Splash early warning system based on image recognition |
| 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 |
| CN111951449A (en) * | 2020-09-09 | 2020-11-17 | 湛江港（集团）股份有限公司 | Unattended intelligent sentry card and intelligent sentry card system |
| CN112182193A (en) * | 2020-10-19 | 2021-01-05 | 山东旗帜信息有限公司 | Log obtaining method, device and medium in traffic industry |
| CN112263224A (en) * | 2020-12-24 | 2021-01-26 | 南京信息工程大学 | Medical information processing method based on FPGA edge calculation |
| CN112580481A (en) * | 2020-12-14 | 2021-03-30 | 康佳集团股份有限公司 | Edge node and cloud cooperative video processing method, device and server |
| CN112992119A (en) * | 2021-01-14 | 2021-06-18 | 安徽大学 | Deep neural network-based accent classification method and model thereof |
| CN113159005A (en) * | 2021-05-28 | 2021-07-23 | 青海中水数易信息科技有限责任公司 | Machine learning-based water level and foreign matter identification integrated monitoring system and method |
| CN113222637A (en) * | 2021-02-26 | 2021-08-06 | 深圳前海微众银行股份有限公司 | Architecture method, device, equipment, medium and program product of store visitor information |
| CN113642403A (en) * | 2021-07-13 | 2021-11-12 | 重庆科技学院 | Crowd abnormal intelligent safety detection system based on edge calculation |
| CN114244975A (en) * | 2021-11-15 | 2022-03-25 | 国能大渡河革什扎水电开发有限公司 | Mobile operation process control method and system based on video edge calculation |
| WO2022105090A1 (en) * | 2020-11-17 | 2022-05-27 | 重庆邮电大学 | Fine-grained data stream reliable offloading method based on callback mechanism |
| CN118627748A (en) * | 2024-03-22 | 2024-09-10 | 南京泽捷辰科技有限公司 | Intelligent supervision method for electric power facilities |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102184602A (en) * | 2011-03-02 | 2011-09-14 | 北京博雅英杰科技股份有限公司 | Bank automatic teller machine (ATM) and field facility narrow-band monitoring and warning system |
| CN105654047A (en) * | 2015-12-21 | 2016-06-08 | 中国石油大学(华东) | Online video intelligent processing system based on deep learning in cloud environment |
| CN106156765A (en) * | 2016-08-30 | 2016-11-23 | 南京邮电大学 | safety detection method based on computer vision |
| CN106981063A (en) * | 2017-03-14 | 2017-07-25 | 东北大学 | A kind of grid equipment state monitoring apparatus based on deep learning |
| CN107134144A (en) * | 2017-04-27 | 2017-09-05 | 武汉理工大学 | A kind of vehicle checking method for traffic monitoring |
| CN107426542A (en) * | 2017-08-08 | 2017-12-01 | 深圳市安软慧视科技有限公司 | The hardware structure of Real-Time Arltificial Intelligence video system based on edge calculations |
-
2017
- 2017-12-14 CN CN201711340896.8A patent/CN108012121A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102184602A (en) * | 2011-03-02 | 2011-09-14 | 北京博雅英杰科技股份有限公司 | Bank automatic teller machine (ATM) and field facility narrow-band monitoring and warning system |
| CN105654047A (en) * | 2015-12-21 | 2016-06-08 | 中国石油大学(华东) | Online video intelligent processing system based on deep learning in cloud environment |
| CN106156765A (en) * | 2016-08-30 | 2016-11-23 | 南京邮电大学 | safety detection method based on computer vision |
| CN106981063A (en) * | 2017-03-14 | 2017-07-25 | 东北大学 | A kind of grid equipment state monitoring apparatus based on deep learning |
| CN107134144A (en) * | 2017-04-27 | 2017-09-05 | 武汉理工大学 | A kind of vehicle checking method for traffic monitoring |
| CN107426542A (en) * | 2017-08-08 | 2017-12-01 | 深圳市安软慧视科技有限公司 | The hardware structure of Real-Time Arltificial Intelligence video system based on edge calculations |
Cited By (47)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108416715A (en) * | 2018-05-11 | 2018-08-17 | 长江大学 | A kind of Campus security management system based on intelligent video camera head |
| CN108873712A (en) * | 2018-06-15 | 2018-11-23 | 北京王府科技有限公司 | ECS edge calculations system |
| CN108921866A (en) * | 2018-07-24 | 2018-11-30 | 北京深瞐科技有限公司 | A kind of image processing method and system |
| CN108828409A (en) * | 2018-08-03 | 2018-11-16 | 南方电网科学研究院有限责任公司 | Fault detection system based on edge calculation |
| CN109005390B (en) * | 2018-08-31 | 2020-12-04 | 山东建筑大学 | Method and system for establishing personnel distribution model based on signal intensity and video |
| CN109005390A (en) * | 2018-08-31 | 2018-12-14 | 山东建筑大学 | Personnel's distributed model method for building up and system based on signal strength and video |
| CN109194926A (en) * | 2018-10-19 | 2019-01-11 | 济南浪潮高新科技投资发展有限公司 | A kind of city security system and its detection method based on edge calculations |
| CN109347644A (en) * | 2018-12-06 | 2019-02-15 | 杭州环普数据技术有限公司 | A kind of block chain edge calculations device |
| CN109670544A (en) * | 2018-12-13 | 2019-04-23 | 广州小狗机器人技术有限公司 | A kind of object detecting apparatus and its acquisition methods, object detecting system |
| CN109685202A (en) * | 2018-12-17 | 2019-04-26 | 腾讯科技（深圳）有限公司 | Data processing method and device, storage medium and electronic device |
| CN109685202B (en) * | 2018-12-17 | 2023-03-21 | 腾讯科技（深圳）有限公司 | Data processing method and device, storage medium and electronic device |
| CN111382836A (en) * | 2018-12-30 | 2020-07-07 | 罗伯特·博世有限公司 | Distributed neural network for edge devices |
| CN109767594A (en) * | 2019-01-30 | 2019-05-17 | 长视科技股份有限公司 | A kind of household monitor video intelligent linkage early warning system |
| CN111753597A (en) * | 2019-03-29 | 2020-10-09 | 中国安全生产科学研究院 | Splash early warning system based on image recognition |
| CN110062208A (en) * | 2019-04-23 | 2019-07-26 | 上海赫煊自动化系统工程有限公司 | A kind of security protection intelligence real-time analyzer and method |
| CN110197128A (en) * | 2019-05-08 | 2019-09-03 | 华南理工大学 | The recognition of face architecture design method planned as a whole based on edge calculations and cloud |
| CN110209140A (en) * | 2019-07-09 | 2019-09-06 | 深圳市华腾物联科技有限公司 | A kind of application method and equipment of edge calculations |
| CN110536299A (en) * | 2019-09-04 | 2019-12-03 | 电子科技大学 | Data packet validity confirmation method based on edge calculation and discrete random convolution |
| CN110738309A (en) * | 2019-09-27 | 2020-01-31 | 华中科技大学 | DDNN training method and DDNN-based multi-view target identification method and system |
| CN110738309B (en) * | 2019-09-27 | 2022-07-12 | 华中科技大学 | DDNN training method and DDNN-based multi-view target identification method and system |
| CN110581898A (en) * | 2019-10-09 | 2019-12-17 | 成都康胜思科技有限公司 | internet of things data terminal system based on 5G and edge calculation |
| CN110581898B (en) * | 2019-10-09 | 2022-04-08 | 成都康胜思科技有限公司 | Internet of things data terminal system based on 5G and edge calculation |
| CN111199279A (en) * | 2019-10-30 | 2020-05-26 | 山东浪潮人工智能研究院有限公司 | Cloud edge calculation and artificial intelligence fusion method and device for police service industry |
| CN111160321A (en) * | 2020-02-10 | 2020-05-15 | 杭州大数云智科技有限公司 | Storage battery car goes up terraced detection and early warning system |
| CN111273071A (en) * | 2020-02-17 | 2020-06-12 | 深圳供电局有限公司 | AI monitoring system and method for edge calculation |
| CN111479048A (en) * | 2020-04-22 | 2020-07-31 | 安徽大学 | Intelligent video image processing equipment based on edge calculation |
| CN111333010A (en) * | 2020-05-19 | 2020-06-26 | 八维通科技有限公司 | Method and system for preventing oil adding error of gas station based on Bluetooth positioning |
| CN111565303A (en) * | 2020-05-29 | 2020-08-21 | 深圳市易链信息技术有限公司 | Video monitoring method, system and readable storage medium based on fog calculation and deep learning |
| CN111565303B (en) * | 2020-05-29 | 2021-12-14 | 广东省电子口岸管理有限公司 | Video monitoring method, system and readable storage medium based on fog calculation and deep learning |
| 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 |
| CN111951449A (en) * | 2020-09-09 | 2020-11-17 | 湛江港（集团）股份有限公司 | Unattended intelligent sentry card and intelligent sentry card system |
| CN112182193B (en) * | 2020-10-19 | 2023-01-13 | 山东旗帜信息有限公司 | Log obtaining method, device and medium in traffic industry |
| CN112182193A (en) * | 2020-10-19 | 2021-01-05 | 山东旗帜信息有限公司 | Log obtaining method, device and medium in traffic industry |
| WO2022105090A1 (en) * | 2020-11-17 | 2022-05-27 | 重庆邮电大学 | Fine-grained data stream reliable offloading method based on callback mechanism |
| CN112580481B (en) * | 2020-12-14 | 2024-05-28 | 康佳集团股份有限公司 | Edge node and cloud collaborative video processing method, device and server |
| CN112580481A (en) * | 2020-12-14 | 2021-03-30 | 康佳集团股份有限公司 | Edge node and cloud cooperative video processing method, device and server |
| CN112263224A (en) * | 2020-12-24 | 2021-01-26 | 南京信息工程大学 | Medical information processing method based on FPGA edge calculation |
| CN112263224B (en) * | 2020-12-24 | 2021-03-23 | 南京信息工程大学 | Medical information processing method based on FPGA edge calculation |
| CN112992119B (en) * | 2021-01-14 | 2024-05-03 | 安徽大学 | Accent classification method based on deep neural network and model thereof |
| CN112992119A (en) * | 2021-01-14 | 2021-06-18 | 安徽大学 | Deep neural network-based accent classification method and model thereof |
| CN113222637A (en) * | 2021-02-26 | 2021-08-06 | 深圳前海微众银行股份有限公司 | Architecture method, device, equipment, medium and program product of store visitor information |
| CN113159005A (en) * | 2021-05-28 | 2021-07-23 | 青海中水数易信息科技有限责任公司 | Machine learning-based water level and foreign matter identification integrated monitoring system and method |
| CN113642403B (en) * | 2021-07-13 | 2023-07-18 | 重庆科技学院 | Crowd abnormal intelligent safety detection system based on edge calculation |
| CN113642403A (en) * | 2021-07-13 | 2021-11-12 | 重庆科技学院 | Crowd abnormal intelligent safety detection system based on edge calculation |
| CN114244975A (en) * | 2021-11-15 | 2022-03-25 | 国能大渡河革什扎水电开发有限公司 | Mobile operation process control method and system based on video edge calculation |
| CN118627748A (en) * | 2024-03-22 | 2024-09-10 | 南京泽捷辰科技有限公司 | Intelligent supervision method for electric power facilities |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108012121A (en) | A kind of edge calculations and the real-time video monitoring method and system of cloud computing fusion | |
| CN110674772B (en) | Intelligent safety control auxiliary system and method for electric power operation site | |
| WO2023142424A1 (en) | Power financial service risk control method and system based on gru-lstm neural network | |
| CN107992836A (en) | A kind of recognition methods of miner's unsafe acts and system | |
| CN105471882A (en) | Behavior characteristics-based network attack detection method and device | |
| CN107332698A (en) | A kind of Security Situation Awareness Systems and method towards bright Great Wall intelligent perception system | |
| US11934536B2 (en) | Dynamic network risk predicting method based on a graph neural network | |
| CN110674790B (en) | Abnormal scene processing method and system in video monitoring | |
| CN108470138A (en) | Method for target detection and device | |
| Idrissi et al. | An unsupervised generative adversarial network based-host intrusion detection system for internet of things devices | |
| CN111738044B (en) | Campus violence assessment method based on deep learning behavior recognition | |
| CN114241522B (en) | Site operation safety wearing identification method, system, equipment and storage medium | |
| CN111126820A (en) | Electricity stealing prevention method and system | |
| CN111723656B (en) | Smog detection method and device based on YOLO v3 and self-optimization | |
| CN115037559B (en) | Data safety monitoring system based on flow, electronic equipment and storage medium | |
| CN112183643A (en) | Hard rock tension-shear fracture identification method and device based on acoustic emission | |
| CN112367273A (en) | Knowledge distillation-based flow classification method and device for deep neural network model | |
| CN110445766A (en) | Ddos attack method for situation assessment and device | |
| CN116311081B (en) | Medical laboratory monitoring image analysis method and system based on image recognition | |
| CN112202718B (en) | XGboost algorithm-based operating system identification method, storage medium and device | |
| CN117749409A (en) | Large-scale network security event analysis system | |
| CN202120402U (en) | Mass video analysis based on cloud computing and rapid retrieval system | |
| CN116112283A (en) | CNN-LSTM-based power system network security situation prediction method and system | |
| CN111143835A (en) | Non-invasive protection method for business logic of electric power metering system based on machine learning | |
| CN115062686A (en) | Multi-KPI (Key performance indicator) time sequence abnormity detection method and system based on multi-angle features |
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: 20180508 |
|
| RJ01 | Rejection of invention patent application after publication |