CN115914373A - Video analysis system based on edge center multilateral cooperation - Google Patents
Video analysis system based on edge center multilateral cooperation Download PDFInfo
- Publication number
- CN115914373A CN115914373A CN202211525590.0A CN202211525590A CN115914373A CN 115914373 A CN115914373 A CN 115914373A CN 202211525590 A CN202211525590 A CN 202211525590A CN 115914373 A CN115914373 A CN 115914373A
- Authority
- CN
- China
- Prior art keywords
- data
- edge
- subsystem
- module
- network
- 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
- 238000004458 analytical method Methods 0.000 title claims abstract description 65
- 238000005259 measurement Methods 0.000 claims abstract description 32
- 238000012545 processing Methods 0.000 claims abstract description 14
- 238000012546 transfer Methods 0.000 claims abstract description 4
- 230000006870 function Effects 0.000 claims description 76
- 238000002372 labelling Methods 0.000 claims description 26
- 238000012549 training Methods 0.000 claims description 21
- 230000002159 abnormal effect Effects 0.000 claims description 11
- 238000000034 method Methods 0.000 claims description 10
- 230000008569 process Effects 0.000 claims description 10
- 238000003062 neural network model Methods 0.000 claims description 9
- 238000013135 deep learning Methods 0.000 claims description 6
- 230000005540 biological transmission Effects 0.000 abstract description 10
- 238000007405 data analysis Methods 0.000 abstract 1
- 238000004891 communication Methods 0.000 description 7
- 238000011161 development Methods 0.000 description 5
- 238000005457 optimization Methods 0.000 description 5
- 230000000694 effects Effects 0.000 description 4
- 206010063385 Intellectualisation Diseases 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 230000006872 improvement Effects 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 230000000295 complement effect Effects 0.000 description 1
- 230000006378 damage Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 230000006855 networking Effects 0.000 description 1
- 238000011112 process operation Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
Images
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Landscapes
- Information Transfer Between Computers (AREA)
Abstract
The invention discloses a video analysis system based on edge center multilateral cooperation, which comprises: the system comprises a plurality of edge measurement subsystems, a front network center subsystem and a data center subsystem; the edge measurement subsystem comprises edge equipment and a camera; the system comprises a data acquisition module, a data processing module, a data analysis module and a data processing module, wherein the data acquisition module is used for acquiring video data and processing and analyzing the video data; the preposed network center subsystem is used for carrying out network transfer on the data of the edge measurement subsystem and the data of the data center subsystem so as to enable the data of the first local area network to be communicated with the data of the second local area network; and the data center subsystem is used for recording and managing the edge equipment information in the edge measurement subsystem and marking personnel or tasks on the video data acquired by the edge measurement subsystem. The invention solves the technical problems of high network delay and low data transmission efficiency in the prior art.
Description
Technical Field
The invention relates to the technical field of Internet of things, in particular to a video analysis system based on edge center multilateral cooperation.
Background
Along with the digital transformation of the financial industry, bank outlets are continuously developed towards intellectualization, machine implementation and unmanned, and security monitoring is particularly important. The number is the basis, the networking is the support, and the intellectualization is the target; edge computing is a key requirement in the aspects of network edge side close to a data source, such as constructing a distributed open system fusing network, computing, storage and application core capabilities, data optimization, application intelligence, safety, privacy protection and the like. With the improvement of edge computing capability and the development of intellectualization, the industry finds that edge terminal nodes of the internet have a large amount of real-time complete data to be mined; for example, in various security cameras in China, tens of millions of cameras can be used indoors and outdoors, massive video data of billions of hours can be generated every week, and the edge is the most complete and latest position of data. The development direction of industrial intelligence is to apply new technologies such as artificial intelligence to the combination of centralized analysis of a data center and edge real-time calculation, and a new driving force is brought to the business process optimization, operation and maintenance automation and business innovation of the industrial internet, so that the advantages of remarkable efficiency improvement, privacy safety and cost are brought. The development of edge computing can be complementary with the development of data centers on the analysis application of industrial data and the industrial intelligence, and a new mode of edge-center cooperation is realized.
Similar edge-center collaborative video analysis systems exist in the market at present, and the systems have two collaborative modes, namely off-line collaboration and on-line collaboration; the system supporting offline collaboration uses the edge device to connect the terminal network camera to analyze data on the edge side and stores the analysis result in the local, then people can periodically go to the edge device to take away the data and place the data in the data center for subsequent analysis and display, and meanwhile, when the application in the system needs to be updated, people need to go to the edge side to update; the system supporting online cooperation can be directly connected with a data center through a network on the edge side, and data transmission and updating of components in the system do not need manual participation and can be directly operated through the network. However, in a scene with data protection, such as security protection, finance and the like, the environment where the edge device is located and the environment where the data center is located are isolated from each other on a network, the edge device and the data center cannot be directly connected through the network, the similar systems want to analyze data and communicate with each other in the two networks, and the offline cooperative system needs to manually transmit and update the data in the two networks, so that the efficiency is low and the delay is high; online collaborative systems simply cannot support such scenarios.
Therefore, a video analysis system capable of reducing network delay and improving data transmission efficiency is needed.
Disclosure of Invention
The invention provides a video analysis system based on edge center multilateral cooperation, which aims to solve the technical problems of high network delay and low data transmission efficiency in the prior art.
In order to solve the above technical problem, an embodiment of the present invention provides a video analysis system based on edge center multilateral cooperation, including: the system comprises a plurality of edge measurement subsystems, a preposed network center subsystem and a data center subsystem; each edge measurement subsystem and the preposed network center subsystem are in a first local area network, and the data center subsystem is in a second local area network; the first local area network and the second local area network are isolated from each other;
the edge measurement subsystem comprises edge equipment and a camera; the system is used for acquiring video data and processing and analyzing the video data;
the preposed network center subsystem is used for carrying out network transfer on data of the edge measurement subsystem and data of the data center subsystem so as to enable the data of the first local area network to be communicated with the data of the second local area network;
and the data center subsystem is used for recording and managing the edge equipment information in the edge measurement subsystem and marking personnel or tasks on the video data acquired by the edge measurement subsystem.
Compared with the prior art, the invention manages and deploys the functions of the edge side in an online mode in the first local area network and the second local area network which are isolated from each other through the preposed network center subsystem, realizes the normal circulation of data, ensures multi-edge communication, and avoids the problem that the prior scheme needs to realize the communication between the edge side and the data center in an offline/manual mode, thereby consuming a large amount of labor cost, causing lower efficiency, higher delay and incapability of practical application in scenes with requirements on delay.
As a preferred scheme, the edge measurement subsystem further comprises a daemon process module;
the daemon process module is used for acquiring the information of the edge equipment, uploading the detailed information of the current edge equipment to the front-end network central subsystem in a file form through a network, and inquiring whether the front-end network central subsystem has an unread configuration file in turn; if yes, downloading all unread configuration files, downloading and operating the corresponding program package according to the configuration files, and marking the unread configuration files as read configuration files.
It can be understood that the daemon module ensures that the edge device can read a new unread configuration file and download a corresponding program package by watching the edge device, so as to start and watch a corresponding function, ensure that the state of the edge device can be updated, maintain the edge device to the latest configuration, and facilitate the adjustment and call of the function of the edge device by a user or an administrator.
As a preferred scheme, the edge measurement subsystem further comprises a video analysis module;
the video analysis module is used for processing and analyzing the video data and comprises a model function sub-module and a business function sub-module;
the model function sub-module is used for quantizing the video data into structured information according to a preset deep learning algorithm and then sending out the structured information;
the business function sub-module is used for subscribing the structural information sent by the model function sub-module, performing business logic operation on the structural information to obtain an analysis result file, and uploading the analysis result file to the preposed network center sub-system through the daemon process module so that the preposed network center sub-system forwards the analysis result file to the data center sub-system.
It can be understood that the video data acquired by the edge device is processed and analyzed through the model function submodule and the service function submodule in the video analysis module, and an analysis result file is obtained through service logic operation after the video data is structured, so that the analysis result file is sent to the preposed network center subsystem and then sent to the data center subsystem, and the accuracy of video data processing is ensured.
Preferably, the preposed network center subsystem comprises a server; the server is used for providing file storage service and Redis service, and the Redis service is used for sending heartbeat information to the data center subsystem.
It can be understood that the front hub subsystem only serves as a relay for the file data and information, so as to ensure that the information in the two isolated first local area networks and the second local area network is normally circulated.
Preferably, the data center subsystem comprises a function management module;
the function management module is used for managing the functions of the data center subsystem and responding to the operation of an administrator to create and register new functions.
It can be understood that, through the function management module, the function to be implemented in the data center subsystem can be managed, so that a user or an administrator can conveniently perform a full life cycle of creating, deploying, starting, stopping and destroying the corresponding function in the system, so that the user or the administrator can better control the edge device, and meanwhile, the video data acquired by the edge device is guaranteed to be transmitted across the network, and the efficiency of data transmission is improved.
As a preferred scheme, the data center subsystem further comprises a data display module;
and the data display module is used for receiving the analysis result file sent by the preposed network center subsystem and displaying the result at the front end.
It can be understood that the analysis result file sent from the front-end network center subsystem can be displayed through the data display module, so that a user or an administrator can accurately and quickly set and manage the equipment, the efficiency of data transmission management is improved, and the use experience of the user is improved.
Preferably, the data center subsystem further includes: a device management module;
the equipment management module is used for managing all the edge equipment, registering all the edge equipment information and displaying the abnormal condition when the abnormal edge equipment appears.
It can be understood that, the device management module manages and registers all the edge device information, and displays the abnormal edge device when the abnormal edge device appears, so that the user or the administrator can manage and control the system of the invention conveniently, and the efficiency of data transmission management is improved.
Preferably, the data center subsystem further includes: a data labeling module;
and the data annotation module is used for re-annotating tasks or personnel of the video data in the received analysis result file according to preset annotation requirements and annotation types, so that an annotation data set is generated and stored.
It can be understood that through the data labeling module, tasks or personnel in the video data can be re-labeled on the basis of the analysis result file, so that the accuracy of the result of video analysis processing is ensured, and the accuracy and efficiency of model training in a subsequent model training module can be improved.
Preferably, the data center subsystem further includes: a model training module;
and the model training module is used for retraining the neural network model corresponding to the corresponding labeling task or labeling personnel by taking the labeling data set as training data so as to obtain the neural network model for labeling different tasks or personnel.
It can be understood that the neural network model corresponding to the corresponding labeling task or labeling personnel is retrained by using the labeling data set as training data, so that the problem that the existing scheme needs to realize communication between the edge side and the data center in an off-line/manual mode, which needs to consume a large amount of labor cost and causes low efficiency and high delay is avoided.
Drawings
FIG. 1: the embodiment of the invention provides a structural schematic diagram of a video analysis system based on edge center multilateral cooperation;
FIG. 2: a flowchart of the daemon module operation provided by the embodiment of the invention;
FIG. 3: a flow chart of the operation of the video analysis module provided by the embodiment of the invention;
FIG. 4: the structure diagram of the preposed network center subsystem provided by the embodiment of the invention;
FIG. 5: a flow chart of the operation of the function management module provided by the embodiment of the invention;
FIG. 6: a flow chart of the operation of the device management module provided by the embodiment of the invention;
FIG. 7: a flow chart of the operation of the data annotation module provided by the embodiment of the invention;
FIG. 8: a flow chart of the operation of the model training module provided by the embodiment of the invention;
wherein the reference numbers of the drawings in the specification are as follows:
the system comprises an edge measurement subsystem 01, a preposed network center subsystem 02 and a data center subsystem 03.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Example one
Referring to fig. 1, a video analysis system based on edge-center multi-edge cooperation according to an embodiment of the present invention includes: the system comprises a plurality of edge measurement subsystems 01, a preposed network center subsystem 02 and a data center subsystem 03; each edge measurement subsystem 01 and the preposed network center subsystem 02 are in a first local area network, and the data center subsystem 03 is in a second local area network; the first local area network is isolated from the second local area network.
It should be noted that, the edge measurement subsystems 01 are all connected to the front-end network center subsystem 02, and the front-end network center subsystem is connected to the data center subsystem 03 to communicate the first local area network with the second local area network.
The edge measurement subsystem 01 comprises edge equipment and a camera; the video processing system is used for acquiring video data and processing and analyzing the video data.
As a preferred scheme of this embodiment, the edge measurement subsystem 01 further includes a daemon module; the daemon process module is used for acquiring the information of the edge equipment, uploading the detailed information of the current edge equipment to the front-end network center subsystem 02 in a file form through a network, and inquiring whether the front-end network center subsystem 02 has an unread configuration file in turn; if yes, downloading all unread configuration files, downloading and operating the corresponding program package according to the configuration files, and marking the unread configuration files as read configuration files.
It should be noted that, the camera in the edge measurement subsystem 01 serves as a data terminal, and can acquire and generate video data, and the edge side device can use the video data to perform intelligent video analysis. Referring to fig. 2, a daemon module watches edge-side equipment, and uploads detailed information of current edge equipment to a server of a front-end network center subsystem 02 in a file form through a network, a data center subsystem 03 equipment management module acquires a detailed information file of the edge equipment from the server of the front-end network center subsystem 02 and registers the detailed information file in the equipment management module, then the daemon module enters a watching state and continuously queries whether configuration information is newly issued in the front-end network center subsystem 02, and if the configuration information is newly issued, a corresponding program package needs to be downloaded, a corresponding function is started, and the corresponding function is watched; and simultaneously, performing state judgment on all started programs, and restarting if abnormal programs exist.
It can be understood that the daemon module ensures that the edge device can read a new unread configuration file and download a corresponding package by watching the edge device, so as to start and watch the corresponding function, thereby ensuring that the state of the edge device can be updated, maintaining the edge device to the most new configuration, and facilitating the adjustment and call of the function of the edge device by a user or an administrator.
As a preferred solution of this embodiment, the edge measurement subsystem 01 further includes a video analysis module; the video analysis module is used for processing and analyzing the video data and comprises a model function sub-module and a business function sub-module.
And the model function sub-module is used for quantizing the video data into structured information according to a preset deep learning algorithm and then sending the structured information.
The service function sub-module is used for subscribing the structural information sent by the model function sub-module, performing service logic operation on the structural information to obtain an analysis result file, and uploading the analysis result file to the front network center sub-system 02 through the daemon process module, so that the front network center sub-system 02 forwards the analysis result file to the data center sub-system 03.
Please refer to fig. 3, the flow of the video analysis module at the edge side analyzes the video of the camera in real time according to the functional requirements, the model function module draws the video stream of the network camera, quantizes the video content into structured information through the calculation and analysis of the deep learning algorithm, and then releases the information as a publisher; the service function module is used as a subscriber to subscribe the information sent by the receiving model function module, performs service logic operation on the information, then obtains an analysis result and generates an analysis result file, and the analysis result file is uniformly responsible for uploading to a server of the preposed network center subsystem 02 by the daemon process module.
It can be understood that the video data acquired by the edge device is processed and analyzed through the model function sub-module and the service function sub-module in the video analysis module, and an analysis result file is obtained through service logic operation after the video data is structured, so that the analysis result file is sent to the front-end network center subsystem 02 and then to the data center subsystem 03, and the accuracy of video data processing is ensured.
The front-end network center subsystem 02 is configured to perform network forwarding on data of the edge measurement subsystem 01 and data of the data center subsystem 03, so that data of the first local area network can be intercommunicated with data of the second local area network.
As a preferred solution of this embodiment, the front-end network center subsystem 02 includes a server; the server is configured to provide a file storage service and a Redis service, and the Redis service is configured to send heartbeat information to the data center subsystem 03.
It should be noted that, referring to fig. 4, the structure diagram of the front-end network center subsystem 02 is a storage server, which does not perform logical operations and is only responsible for transferring files and information, and the subsystem is used for normally transferring information of two isolated networks; therefore, a corresponding directory is arranged on the server in the preposed network center subsystem 02 to complete corresponding functions; secondly, in order to transmit information data such as the state of the edge device to the data center subsystem 03 in real time, a Redis service is installed in the front-end network center subsystem 02 and is used for sending heartbeat information sent by the transit edge side subsystem to the data center server.
It can be understood that the front hub subsystem 02 only serves as a relay for the document data and information, so as to ensure the normal flow of information in the two isolated first and second lans.
The data center subsystem 03 is configured to record and manage edge device information in the edge measurement subsystem 01, and label video data acquired by the edge measurement subsystem 01 with a person or a task.
As a preferred solution of this embodiment, the data center subsystem 03 includes a function management module; the function management module is configured to manage functions of the data center subsystem 03, and create and register new functions in response to an operation by an administrator.
Please refer to fig. 5, a flow of a function management module on the data center subsystem 03, where the function management module manages all functions to be implemented, and new function requirements are registered in the module, and generally, the flow of the module includes selecting a function, selecting a device, selecting a camera, selecting a model function package, selecting a service function package, configuring parameter information, generating a function configuration file, and placing the function configuration file in the front-end network center subsystem 02. While managing the full lifecycle of all functions from creation, deployment, start, stop, and destruction.
It can be understood that, through the function management module, the function to be implemented in the data center subsystem 03 can be managed, so that a user or an administrator can conveniently perform a full life cycle of creating, deploying, starting, stopping and destroying the corresponding function in the system, so that the user or the administrator can better control the edge device, and meanwhile, the video data acquired by the edge device is guaranteed to be transmitted across the network, and the efficiency of data transmission is improved.
As a preferred solution of this embodiment, the data center subsystem 03 further includes a data display module; the data display module is used for receiving the analysis result file sent from the preposed network center subsystem 02 and displaying the result at the front end.
It should be noted that the data display module on the data center subsystem 03 acquires the analysis result file of the video uploaded at the edge side from the front-end network center subsystem 02, and then displays the result at the web page side, so that an operator can view the video analysis result through the display of the data display module.
It can be understood that the analysis result file sent from the front-end network center subsystem 02 can be displayed through the data display module, so that a user or an administrator can accurately and quickly set and manage the equipment, the efficiency of data transmission management is improved, and the use experience of the user is improved.
As a preferred solution of this embodiment, the data center subsystem 03 further includes: a device management module; the equipment management module is used for managing all the edge equipment, registering all the edge equipment information and displaying the abnormal condition when the abnormal edge equipment appears.
Please refer to fig. 6, a flow of the device management module on the data center subsystem 03, where the device management module manages all edge devices, registers detailed information of all edge devices, including detailed information of cameras that can be connected to the edge devices, and is responsible for querying and displaying states of all edge devices to the front-end hub subsystem 02.
It can be understood that, the device management module manages and registers all the edge device information, and displays the abnormal edge device when the abnormal edge device appears, so that the user or the administrator can manage and control the system of the invention conveniently, and the efficiency of data transmission management is improved.
As a preferred solution of this embodiment, the data center subsystem 03 further includes: a data annotation module; and the data annotation module is used for re-annotating tasks or personnel of the video data in the received analysis result file according to preset annotation requirements and annotation types, so that an annotation data set is generated and stored.
Please refer to fig. 7, in the process of the data annotation module on the data center subsystem 03, the data annotation module may re-annotate the uploaded analysis result video data from the analysis result file of the acquired video data, in combination with the actual demand and the existing video analysis effect, and the annotation data may automatically generate a data set according to the task.
It can be understood that through the data labeling module, tasks or personnel in the video data can be re-labeled on the basis of the analysis result file, so that the accuracy of the result of video analysis processing is ensured, and the accuracy and efficiency of model training in a subsequent model training module can be improved.
As a preferred solution of this embodiment, the data center subsystem 03 further includes: a model training module; and the model training module is used for retraining the neural network model corresponding to the corresponding labeling task or labeling personnel by taking the labeling data set as training data so as to obtain the neural network model for labeling different tasks or personnel.
Please refer to fig. 8, in the process of the model training module on the data center subsystem 03, the model training module takes the labeled data set as training data, and performs retraining of the deep learning neural network on the model with the function to be optimized, so as to improve the accuracy of the neural network model and optimize the effect of the model function; in addition to optimizing existing functions, the model training module can also utilize the labeled data set to train functional models that need to be implemented in the future.
It can be understood that the neural network model corresponding to the corresponding labeling task or labeling personnel is retrained by using the labeling data set as training data, so that the problem that the existing scheme needs to realize communication between the edge side and the data center in an off-line/manual mode, which needs to consume a large amount of labor cost and causes low efficiency and high delay is avoided.
Further, by the arrangement of the front-end network center subsystem 02, the embodiments of the present invention can implement the functions of managing and deploying the edge sides in an online manner in two isolated networks, and implement normal data circulation. Compared with the existing scheme in the market, the communication between the edge side and the data center needs to be realized in an off-line/manual mode, a large amount of labor cost needs to be consumed, the efficiency is low, the delay is high, and the situation with a requirement on delay cannot be realized; the system is characterized in that a data transfer station, namely a preposed network center subsystem 02, is used for communicating an edge side subsystem with a data center subsystem 03, so that the normal circulation of data is realized; meanwhile, the embodiment of the invention can realize the next-step strategy optimization according to the actual analysis effect while managing the edge side equipment by arranging the data center, and continuously adds new functions and converts old functions. Compared with the existing scheme on the market, the data center cannot label and retrain actual data for optimizing functions, can only display, store and manage the data by one, has single function, cannot develop sustainably and has weak expansibility; the data center management system has the advantages that the data center function is rearranged, multiple novel functions are added, one-stop management, deployment and optimization can be realized, the edge-center multi-edge cooperation characteristic is fully utilized, the functions are rich, the sustainability development is strong, and the expansion capability is strong.
The embodiment of the invention has the following effects:
compared with the prior art, the embodiment of the invention manages and deploys the function of the edge side in an online mode in the first local area network and the second local area network which are isolated from each other through the preposed network center subsystem, realizes the normal circulation of data, ensures the multi-edge communication, and avoids the problem that the prior scheme needs to realize the communication between the edge side and the data center in an offline/manual mode, thereby consuming a large amount of labor cost, causing lower efficiency and higher delay, and being incapable of being practically applied in a scene with a requirement on delay.
Furthermore, the reason why the next-step optimization strategy, the continuous addition of new functions and the conversion of old functions are performed according to the actual analysis result file while the edge-side equipment is managed is that resources are rearranged in the data center system, and the functions of data labeling, deep learning network model training, function updating and updating, and old and new functions are realized in the data center.
The above-mentioned embodiments are provided to further explain the objects, technical solutions and advantages of the present invention in detail, and it should be understood that the above-mentioned embodiments are only examples of the present invention and are not intended to limit the scope of the present invention. It should be understood that any modifications, equivalents, improvements and the like, which come within the spirit and principle of the invention, may occur to those skilled in the art and are intended to be included within the scope of the invention.
Claims (9)
1. A video analysis system based on edge-centric multilateral collaboration, comprising: the system comprises a plurality of edge measurement subsystems, a front network center subsystem and a data center subsystem; each edge measurement subsystem and the preposed network center subsystem are in a first local area network, and the data center subsystem is in a second local area network; the first local area network and the second local area network are isolated from each other;
the edge measurement subsystem comprises edge equipment and a camera; the system is used for acquiring video data and processing and analyzing the video data;
the preposed network center subsystem is used for carrying out network transfer on data of the edge measurement subsystem and data of the data center subsystem so as to enable the data of the first local area network to be communicated with the data of the second local area network;
and the data center subsystem is used for recording and managing the edge equipment information in the edge measurement subsystem and marking personnel or tasks on the video data acquired by the edge measurement subsystem.
2. The video analysis system based on edge-centric multi-edge collaborative sharing, wherein the edge measurement subsystem further comprises a daemon module;
the daemon process module is used for acquiring the information of the edge equipment, uploading the detailed information of the current edge equipment to the front-end network central subsystem in a file form through a network, and inquiring whether the front-end network central subsystem has an unread configuration file in turn; if yes, downloading all unread configuration files, downloading and operating the corresponding program package according to the configuration files, and marking the unread configuration files as read configuration files.
3. The video analysis system based on edge-centric multi-edge collaborative sharing as claimed in claim 2, wherein said edge measurement subsystem further comprises a video analysis module;
the video analysis module is used for processing and analyzing the video data and comprises a model function sub-module and a service function sub-module;
the model function sub-module is used for quantizing the video data into structured information according to a preset deep learning algorithm and then sending the structured information;
the business function sub-module is used for subscribing the structural information sent by the model function sub-module, performing business logic operation on the structural information to obtain an analysis result file, and uploading the analysis result file to the preposed network center sub-system through the daemon process module so that the preposed network center sub-system forwards the analysis result file to the data center sub-system.
4. The video analytics system based on edge-centric multi-edge collaborative sharing as claimed in claim 1, wherein the front-end network-centric subsystem comprises a server; the server is used for providing file storage service and Redis service, and the Redis service is used for sending heartbeat information to the data center subsystem.
5. The video analytics system based on edge-centric multi-edge collaborative sharing as claimed in claim 1, wherein the data center subsystem comprises a function management module;
the function management module is used for managing the functions of the data center subsystem and responding to the operation of an administrator to create and register new functions.
6. The video analysis system based on edge-centric multi-edge collaborative sharing according to claim 3, wherein the data center subsystem further comprises a data presentation module;
and the data display module is used for receiving the analysis result file sent by the preposed network center subsystem and displaying the result at the front end.
7. The video analytics system based on edge-centric multi-edge collaborative sharing as claimed in claim 1, wherein the data center subsystem further comprises: a device management module;
the equipment management module is used for managing all the edge equipment, registering all the edge equipment information and displaying the abnormal condition when the abnormal edge equipment appears.
8. The video analytics system based on edge-centric multi-edge collaborative sharing as claimed in claim 3, wherein the data center subsystem further comprises: a data annotation module;
and the data labeling module is used for re-labeling the tasks or personnel of the video data in the received analysis result file according to the preset labeling requirement and the labeling type, so as to generate and store a labeled data set.
9. The video analytics system based on edge-centric multi-edge collaborative sharing as claimed in claim 8, wherein the data center subsystem further comprises: a model training module;
and the model training module is used for retraining the neural network model corresponding to the corresponding labeling task or labeling personnel by taking the labeling data set as training data so as to obtain the neural network model for labeling different tasks or personnel.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202211525590.0A CN115914373A (en) | 2022-11-29 | 2022-11-29 | Video analysis system based on edge center multilateral cooperation |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202211525590.0A CN115914373A (en) | 2022-11-29 | 2022-11-29 | Video analysis system based on edge center multilateral cooperation |
Publications (1)
Publication Number | Publication Date |
---|---|
CN115914373A true CN115914373A (en) | 2023-04-04 |
Family
ID=86475926
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202211525590.0A Pending CN115914373A (en) | 2022-11-29 | 2022-11-29 | Video analysis system based on edge center multilateral cooperation |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN115914373A (en) |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20190044918A1 (en) * | 2018-03-30 | 2019-02-07 | Intel Corporation | Ai model and data camouflaging techniques for cloud edge |
CN114666576A (en) * | 2022-05-23 | 2022-06-24 | 创意信息技术股份有限公司 | Public safety edge gateway system |
CN114666554A (en) * | 2022-05-23 | 2022-06-24 | 创意信息技术股份有限公司 | Edge gateway cloud service management system |
-
2022
- 2022-11-29 CN CN202211525590.0A patent/CN115914373A/en active Pending
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20190044918A1 (en) * | 2018-03-30 | 2019-02-07 | Intel Corporation | Ai model and data camouflaging techniques for cloud edge |
CN114666576A (en) * | 2022-05-23 | 2022-06-24 | 创意信息技术股份有限公司 | Public safety edge gateway system |
CN114666554A (en) * | 2022-05-23 | 2022-06-24 | 创意信息技术股份有限公司 | Edge gateway cloud service management system |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Hejazi et al. | Survey of platforms for massive IoT | |
CN102693108B (en) | Method and system for centralized printing management based on printer server | |
US20090157835A1 (en) | Presence Enabled Instance Messaging for Distributed Energy Management Solutions | |
CN100547545C (en) | The method and system that is used for the application fractionation of network edge calculating | |
CN110995859A (en) | Intelligent transformer substation supporting platform system based on ubiquitous Internet of things | |
CN108306804A (en) | A kind of Ethercat main station controllers and its communication means and system | |
CN103413210A (en) | Emergency material management system based on internet of things technology | |
US11606272B1 (en) | Techniques for cross platform communication process flow anomaly detection and display | |
CN113127307A (en) | Method for processing tracing request, related device, system and storage medium | |
WO2019042432A1 (en) | Information interaction method and system | |
CN104270432B (en) | Based on drilling well industry Real-time Data Service system and data interactive method | |
CN114205641A (en) | Video data processing method and device | |
CN116915827A (en) | Data transmission method and device of internet of things edge gateway, electronic equipment and medium | |
CN116506434B (en) | Multi-terminal offline-operation intelligent warehouse management method | |
CN105205735A (en) | Power dispatching data cloud service system and implementation method | |
CN112163708A (en) | Monitoring management system and method based on intelligent display terminal | |
CN112051816A (en) | Data acquisition system and method | |
CN112650653A (en) | Plug-and-play and visual operation and maintenance system and method for equipment | |
CN112434972A (en) | Bar code management system and method based on WEB | |
CN109995782B (en) | Information processing method, device, system and computer storage medium | |
CN115914373A (en) | Video analysis system based on edge center multilateral cooperation | |
CN115190147B (en) | Intelligent device control method, device and system | |
CN107222392B (en) | Communication method, device, system and computer storage medium | |
US20230090607A1 (en) | Techniques for cross platform communication process flow metric generation and display | |
CN112307092A (en) | Data architecture system of energy system and energy system |
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: 20230404 |
|
RJ01 | Rejection of invention patent application after publication |