CN111225280B - Lightweight video analysis system based on embedded platform - Google Patents
Lightweight video analysis system based on embedded platform Download PDFInfo
- Publication number
- CN111225280B CN111225280B CN202010074352.7A CN202010074352A CN111225280B CN 111225280 B CN111225280 B CN 111225280B CN 202010074352 A CN202010074352 A CN 202010074352A CN 111225280 B CN111225280 B CN 111225280B
- Authority
- CN
- China
- Prior art keywords
- module
- data
- video analysis
- video
- streaming media
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
- H04N21/43—Processing of content or additional data, e.g. demultiplexing additional data from a digital video stream; Elementary client operations, e.g. monitoring of home network or synchronising decoder's clock; Client middleware
- H04N21/443—OS processes, e.g. booting an STB, implementing a Java virtual machine in an STB or power management in an STB
- H04N21/4431—OS processes, e.g. booting an STB, implementing a Java virtual machine in an STB or power management in an STB characterized by the use of Application Program Interface [API] libraries
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/20—Servers specifically adapted for the distribution of content, e.g. VOD servers; Operations thereof
- H04N21/23—Processing of content or additional data; Elementary server operations; Server middleware
- H04N21/234—Processing of video elementary streams, e.g. splicing of video streams, manipulating MPEG-4 scene graphs
- H04N21/23418—Processing of video elementary streams, e.g. splicing of video streams, manipulating MPEG-4 scene graphs involving operations for analysing video streams, e.g. detecting features or characteristics
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/20—Servers specifically adapted for the distribution of content, e.g. VOD servers; Operations thereof
- H04N21/25—Management operations performed by the server for facilitating the content distribution or administrating data related to end-users or client devices, e.g. end-user or client device authentication, learning user preferences for recommending movies
- H04N21/254—Management at additional data server, e.g. shopping server, rights management server
- H04N21/2541—Rights Management
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
- H04N21/43—Processing of content or additional data, e.g. demultiplexing additional data from a digital video stream; Elementary client operations, e.g. monitoring of home network or synchronising decoder's clock; Client middleware
- H04N21/44—Processing of video elementary streams, e.g. splicing a video clip retrieved from local storage with an incoming video stream, rendering scenes according to MPEG-4 scene graphs
- H04N21/44008—Processing of video elementary streams, e.g. splicing a video clip retrieved from local storage with an incoming video stream, rendering scenes according to MPEG-4 scene graphs involving operations for analysing video streams, e.g. detecting features or characteristics in the video stream
Abstract
The invention provides a lightweight video analysis system based on an embedded platform, which is used for carrying out lightweight processing on streaming media data so as to obtain and output a video analysis result, and is characterized by comprising the following components: the streaming media module is used for acquiring streaming media data and preprocessing the streaming media data to obtain preprocessed data, wherein the streaming media data at least comprises video streaming data; the video analysis module is used for continuously receiving the preprocessed data and analyzing according to a preset video analysis process to obtain a video analysis result; and the data persistence module is used for performing persistence processing on the video analysis result and storing the video analysis result as dynamic data, wherein the streaming media module, the video analysis module and the data persistence module are deployed on the edge node through a container virtualization technology, so that the streaming media module, the video analysis module and the data persistence module complete video analysis processing by utilizing the computing resources of the edge node.
Description
Technical Field
The invention belongs to the field of computer vision, relates to a video analysis system, and particularly relates to a lightweight video analysis system based on an embedded platform.
Background
In the era of mobile internet, mobile terminal devices carrying operating systems such as Android, IOS and Windows phone have been rapidly increasing, and the ecology of mobile internet has been shaped. With the help of the mobile internet, the communication mode of people is greatly changed. At the beginning of the rise of the mobile internet, a user holding mobile terminal equipment can only transmit text information and picture information, and under the background of continuous evolution of the mobile internet, the mobile terminal user can transmit video information in real time, so that the network communication form of people is enriched. Entering the age of 5G, some emerging technologies based on the internet are emerging, such as: ar (augmented reality), vr (virtual reality), iot (internet of things), big data technology, etc. are applied in succession and continue to be of social interest. With such a trend, the video stream is expected to grow explosively in the future. According to recent Visual Network Index (VNI) reports promulgated by cisco, IP traffic generated in the year 2022 will be enormous, expected to be the sum of all traffic generated during the year 2016 since the birth of the internet. Meanwhile, by 2022, the total of the global netizens will reach 60\ percent of the global population, at that time, more than 280 hundred million devices will be connected to the internet, and the traffic generated by the devices will occupy 82 percent of all the IP traffic.
However, with the increasing number of devices accessing the mobile internet, video information streams will dominate the IP traffic, and this trend is irreversible. While the video stream brings rich information to the outside, it also has certain negative effects, such as: communication network congestion and time delay in video stream transmission are caused, and seamless connection is difficult to achieve through analysis aiming at videos. These problems are caused by the size of the video stream itself, and the data size of each video stream is relatively large, so when there is a need to analyze and process the video stream, if the video stream is still analyzed by the conventional video processing method, the analysis rate will be affected by the video stream, and the transmission efficiency of the video stream will be further delayed.
Disclosure of Invention
In order to solve the above problems, the present invention provides a video analysis system which can realize the deployment of a lightweight video analysis process in different resource environments of a mobile edge terminal, thereby weakening the influence of the data scale of a video stream on the video analysis process, and adopts the following technical scheme:
the invention provides a lightweight video analysis system based on an embedded platform, which is used for carrying out lightweight processing on streaming media data so as to obtain and output a video analysis result, and is characterized by comprising the following steps: the streaming media module is used for acquiring streaming media data and preprocessing the streaming media data to obtain preprocessed data, wherein the streaming media data at least comprises video streaming data; the video analysis module is used for continuously receiving the preprocessed data and analyzing according to a preset video analysis process to obtain a video analysis result; and the data persistence module is used for performing persistence processing on the video analysis result and storing the video analysis result as dynamic data, wherein the streaming media module, the video analysis module and the data persistence module are deployed on the edge node through a container virtualization technology, so that the streaming media module, the video analysis module and the data persistence module complete video analysis processing by utilizing the computing resources of the edge node.
The embedded platform-based lightweight video analysis system provided by the invention can also have the technical characteristics that: the API access module is used for monitoring an external request of a user and provides a uniform API interface for the outside; the API access module analyzes the external request into an internal message and identifies the legality of the external request once monitoring the external request, and sends the corresponding internal message to the routing module and enables the routing module to distribute according to the message header of the internal message if the external request is a legal request; and if the external request is an illegal request, the API access module responds to the external request according to a preset program.
The embedded platform-based lightweight video analysis system provided by the invention can also have the technical characteristics that the external request is a request for creating a video stream for processing the video stream, the routing module distributes internal information corresponding to the request for creating the video stream to the streaming media module, the streaming media module creates a corresponding working thread according to the received internal information, and the working thread is used for performing stream pulling, stream fetching and decoding preprocessing of the streaming media module and packaging the obtained preprocessed data into internal information corresponding to image analysis.
The embedded platform-based lightweight video analysis system provided by the invention can also have the technical characteristics that the routing module distributes internal information corresponding to image analysis to the video analysis module, and once the video analysis module completes analysis and obtains a video analysis result, the API access module outputs the video analysis result according to a video stream creation request.
The embedded platform-based lightweight video analysis system provided by the invention can also have the technical characteristics that an external request is a data input request for inputting static data by a user, the routing module distributes internal information corresponding to the data input request to the data persistence module, and the data persistence module carries out persistence processing according to the static data in the received internal information and stores the data as the static data.
The embedded platform-based lightweight video analysis system provided by the invention can also have the technical characteristics that static data comprises user information, equipment information and base information.
The embedded platform-based lightweight video analysis system provided by the invention can also have the technical characteristics that: and the result display part is used for displaying the video analysis result through a display page.
The embedded platform-based lightweight video analysis system provided by the invention can also have the technical characteristics that the video analysis process is a face recognition process, a behavior detection process or a vehicle detection process.
Action and Effect of the invention
According to the embedded platform-based lightweight video analysis system, the streaming media module, the video analysis module and the data persistence module are arranged, so that media stream data can be acquired, analyzed and persistently stored, and analysis on videos is completed. In addition, the streaming media module, the video analysis module and the data persistence module are deployed on the edge node through a container virtual technology, so that the analysis process of the video can be processed by utilizing the computing resources of the edge node, and the effects of lightweight video analysis and computing localization are achieved, so that data transmission with a cloud end is reduced, the bandwidth pressure of a network is further reduced, and the data security is improved. The lightweight video analysis system can be applied to the video analysis fields of non-sensory traffic, behavior detection, face detection, video acceleration and the like, so that the video analysis processes in the fields are light in weight.
Drawings
Fig. 1 is a block diagram of a video analysis system according to an embodiment of the present invention.
Detailed Description
In 2014, in the early stage that the 5G technology is about to enter into commercial use, the European Telecommunications Standards Institute formally standardizes the Mobile Edge Computing (MEC) technology, which is coming out along with the emergence of the 5G technology, and mainly aims to solve the problems of high delay, low throughput and poor network scalability existing in the conventional communication, so as to realize high integration of a communication base station and internet services. The MEC technology can put external application at the edge of the network, realize seamless butt joint of content and service, improve service experience, and effectively improve the speed of a mobile network and reduce time delay. MEC technology has been proposed to date and has found application in various fields, typically multi-access edge application orchestration. By means of the MEC technology, technicians can efficiently collect information in wireless communication and automatically optimize mobile services.
The invention is based on embedded platform (such as Jetson TX2) mobile terminal platform, using mobile edge calculation in network transmission and virtual technology, through container to design a set of lightweight video analysis system, to realize high efficiency processing of video data and intelligent monitoring. The video analysis system can realize automation of an analysis process, labor force of manual analysis in traditional video analysis is liberated, external key information can be captured in real time, big data analysis is carried out on video stream information through an analysis model of the construction system, and a final analysis result can provide basis for intelligent monitoring.
For technicians in the software development industry, the invention can provide new software development ideas for the technicians and apply the latest communication technology to actual scenes.
In order to make the technical means, the creation features, the achievement purposes and the effects of the invention easy to understand, the following describes the embedded platform-based lightweight video analysis system specifically with reference to the embodiments and the accompanying drawings.
< example >
The system in this embodiment is implemented on a Linux platform, which has at least one GPU card support.
Fig. 1 is a system architecture diagram of a video analysis system in an embodiment of the invention.
As shown in fig. 1, the video analysis system 100 includes an API access module 1, a routing module 2, a streaming media module 3, a video analysis module 4, a data persistence module 5, and a result display module 6.
In this embodiment, the video analysis system 100 is implemented based on an edge computing framework, which provides the video analysis system 100 with a container technology and a service management function, and the communication between containers has virtualized a network transmission technology, and network resource virtualization can be implemented through the containers, so as to achieve an effect of communication between different edge nodes by modules in the system.
Specifically, the edge nodes are virtualized by a container virtualization technology, so that the API access module 1, the routing module 2, the streaming media module 3, the video analysis module 4, and the data persistence module 5 operate in each virtual node, and an external unified interface is provided through the API access module 1, thereby providing greater flexibility for the deployment of different modules in different edge nodes.
The API access module 1 is provided with a unified API interface for a user to access the video analysis system 100 through the API interface. The user may be a user or other system.
In this embodiment, after monitoring the external request received by the video analysis system 100, the API access module 1 analyzes the external request and generates a corresponding internal message, and identifies the legitimacy of the external request while analyzing the external request, and if the external request is an illegal request, directly responds to the illegal request (for example, rejecting, alarming, or performing a conventional response measure such as a countermeasure); otherwise, the corresponding internal message is sent to the routing module 2. The internal message includes a message header and a message body.
In this embodiment, the main interaction between the user and the video analysis system 100 is to send a request through the API module. For example, the user sends a request for creating a video stream to the video analysis system 100 through the user terminal held by the user terminal based on the RESTful API and the ZeroMQ protocol, and once the API access module 1 listens to the request for creating a video stream, which is an external request, is parsed into a corresponding internal message (i.e., a message header corresponds to the request for creating a video stream), and at the same time, the legitimacy is authenticated.
The routing module 2 is configured to identify the received internal message, so as to perform next hop module selection according to a message header of the internal message.
In this embodiment, when the routing module 2 receives the internal message sent by the API access module 1, the routing module 2 allocates the internal message to the streaming media module 3 because the header of the internal message indicates a request for creating a video stream.
The streaming media module 3 is used for creating a working thread so as to acquire streaming media and perform preprocessing. The streaming media module 3 mainly functions to support multiple streaming media transmission protocols and complete video acquisition of multiple video streams; the support to various stream media coding and decoding is realized, and the data preprocessing of various coding and decoding formats is completed. In this embodiment, in order to support parallel processing of multiple media streams, the streaming media module 3 adopts a programming model of a master and workers.
In this embodiment, streaming media (streaming media) is a part of multimedia, including streaming data of video and audio, and since the video analysis system 100 is a lightweight system for processing video, the streaming media module 3 in this embodiment only acquires and processes video streaming data.
In this embodiment, when receiving the internal message corresponding to the request for creating the video stream, the streaming media module 3 creates a working thread, thereby completing the pre-processing of stream pulling, stream fetching, and encoding and decoding of the video stream. Specifically, the streaming media module 3 includes a flow determination unit 31, a work thread creation unit 32, a decoding preprocessing unit 33, and a media message packetizing unit 34.
The flow determination unit 31 is configured to determine the video stream acquisition flow according to the video source type and the transmission protocol indicated in the internal message.
The work thread creating unit 32 is configured to create a corresponding work thread according to the video stream acquisition flow, so as to pull the video stream data. In this embodiment, the worker thread uses the gstreamer library to pull streams of different streaming media transmission protocols.
The decoding preprocessing unit 33 is configured to perform decoding preprocessing on the video stream data continuously acquired by the working thread to obtain image data.
The media message packing unit 34 is configured to pack the image data processed by the decoding preprocessing unit 33, assemble the image data into an internal message with a message header corresponding to the image analysis, and send the internal message to the routing module 2. In this embodiment, when receiving an internal message of an image analysis corresponding to a message header, the routing module 2 sends the internal message to the video analysis module 4.
The video analysis module 4 is used for extracting information from the image data and further analyzing the image data.
In this embodiment, since the image data processed by the streaming media module 3 can be directly used, the video analysis module 4 directly analyzes and processes the image data.
In this embodiment, the video analysis module adopts a programming model of a master and workers. The flow of video analysis is numerous and complex and the system will abstract it. Defining a high-level analysis task as a process pipeline, such as face recognition, behavior detection, vehicle detection and the like; a component used in one flow is defined as one task.
In this embodiment, the video analysis module 4 includes an analysis event parsing unit 41, an analysis flow distribution unit 42, and an analysis message packing unit 43, and when the video analysis module 4 receives image data (i.e., an internal message) continuously pushed by the streaming media module 3, the video analysis module completes processing of each image data through these units.
The analysis event parsing unit 41 is configured to, when the internal message of the streaming media module 3 is acquired, parse an event ID in the internal message.
The analysis flow allocation unit 42 distributes data in the message body of the corresponding internal message to the video analysis flow corresponding to the event according to the event ID acquired by the analysis event parsing unit 41, thereby obtaining a video analysis result.
For example, in this embodiment, video analysis flows such as a face recognition analysis flow, a vehicle detection analysis flow, and a behavior detection flow may be implemented, and when the video analysis module 4 receives a preprocessed image message sent by a video stream, the video analysis module 4 determines a corresponding flow for distributing the message to the system according to a unique identifier of the message. Each analysis process is a complete algorithm process, and an algorithm result, namely a video analysis result, is obtained after the process is ended.
The analysis message packing unit 43 is configured to pack the video analysis result obtained by the analysis flow allocation unit 42 into an internal message, and send the internal message to the routing module 2.
In this embodiment, when receiving the internal message corresponding to the video analysis result, the routing module 2 may simultaneously send the internal message to the API access module 1 and the data persistence module 5 for response, for example, send the internal message to the data persistence module 5 for persistent storage of the analysis result and send the internal message to the API access module 1 for real-time pushing of the analysis result.
The data persistence module 5 is used for persisting and storing the video analysis result into the storage module.
In this embodiment, the data persistence module 5 includes a data event analysis unit 51, a persistence flow assignment unit 52, and a storage message packing unit 53.
The data event parsing unit 51 is configured to, when the internal message of the video analysis module 4 is acquired, parse an event ID in the internal message.
The persistent flow allocating unit 52 allocates the corresponding internal message to the corresponding persistent process according to the event ID acquired by the data event parsing unit 51.
In this embodiment, different storage back ends, such as NAS, local file system, relational database, time-series database, and the like, are obtained in different persistence processes, so that the analysis result in the internal message is stored in the corresponding storage back end. In addition, if the analysis result has stronger time sequence, the time sequence database Cassandra can be used for persistence. By utilizing different characteristics of different storage back-ends, different types of data storage of the embodiment is satisfied. For example, for some user-inserted data of the present embodiment, the underlying database data of the video analytics may be stored using a MySQL relational database; data such as a video analysis flow model can be stored by using an NAS file system; analytical data generated on the fly may be stored using a Cassandra time series database.
The storage message packaging unit 53 is configured to generate a corresponding result message body according to the result written in the persistence process, and send the result message body to a corresponding storage back end, such as a MySQL relational database, a Cassandra time-series database, an NAS file system, and the like, for persistent storage.
In addition, in this embodiment, the user may also write the metadata and the static data into the video analysis system 100 by sending a data input request to the API access module 1, and the data persistence module 5 persists the metadata and the static data through the relational database MySQL when receiving the internal message sent by the API access module 1. The static data is user information, device information, base information, and the like.
The result display module 6 is used for displaying the video analysis result analyzed by the video analysis module 4.
In this embodiment, the result display module 6 is a display module (e.g., a display device) of the video analysis system 100, and when the API access module 1 receives an internal message corresponding to a video analysis result, the internal message is sent to the result display module 6, so that the result display module 6 displays the video analysis result and allows a user to view the video analysis result.
Meanwhile, in other embodiments, the API access module 1 may directly convert the internal message corresponding to the video analysis result into an external request and send the external request to another terminal (e.g., a user terminal) or a system, so as to perform corresponding processing (e.g., display and view or perform further processing according to the video analysis data).
The following specifically describes the flow of the video analysis system in the embodiment of the present invention:
in this embodiment, an entrance of the video analysis system 100 is an API access module 1, and after the API access module 1 monitors an external request sent by a user or another system, the external request is analyzed by the API access module as an internal message, and the legitimacy of the external request is identified, and if the external request is an illegal request, the external request is directly responded to the illegal request; if the request is a legal request, the corresponding internal message is sent to the routing module 2. The routing module 2 makes the next hop module selection by means of the header of the internal message, which is sent to the streaming module 3 if the header indicates a request to create a video stream. After receiving the internal message corresponding to the request for creating the video stream, the streaming media module 3 creates a working thread to perform the operations of pulling stream, fetching stream, encoding and decoding, packages the video stream data which is continuously obtained and is subjected to decoding preprocessing into the internal message, and sends the internal message to the routing module 2. The routing module 2 sends the header of the message to the video analysis module 4 as an internal message for image analysis. After the video analysis module 4 obtains the data, the image data is distributed, a corresponding video analysis flow is selected for analysis processing, and finally, the analysis result is packaged and sent to the routing module 2. The routing module 2 sends the internal message with the message header as the video analysis result to the API access module 1 and the data persistence module 5 for response and storage.
In practical application, a third-party application may deploy the video analysis system 100 of this embodiment on an edge node with certain computing resources, and obtain data storage services, video analysis services, and video stream management services through an API interface (i.e., the API access module 1) provided by the video analysis system 100, so that the third-party application can obtain video analysis functions conveniently and quickly. For example, the third-party check-in system may use the face recognition video analysis process implemented by the system 100 through the RESTful API and the ZeroMQ interface to add a face recognition function to the third-party check-in system.
Examples effects and effects
According to the embedded platform-based lightweight video analysis system provided by the embodiment, the streaming media module, the video analysis module and the data persistence module are provided, so that media stream data can be acquired, analyzed and persistently stored, and the analysis of videos is completed. In addition, the streaming media module, the video analysis module and the data persistence module are deployed on the edge node through a container virtual technology, so that the analysis process of the video can be processed by utilizing the computing resources of the edge node, and the effects of lightweight video analysis and computing localization are achieved, so that data transmission with a cloud end is reduced, the bandwidth pressure of a network is further reduced, and the data security is improved. The lightweight video analysis system can be applied to the video analysis fields of non-sensory traffic, behavior detection, face detection, video acceleration and the like, so that the video analysis processes in the fields are light in weight.
In addition, in the embodiment, because the streaming media module, the video analysis module and the data persistence module are encapsulated by using a container virtualization technology, the virtualization of the edge node resources is realized, and the unified interface is exposed to the outside through the API access module, so that the system of the embodiment can be conveniently accessed with other systems or terminals of users, and the complexity of realizing the combination with other systems is reduced.
The above-described embodiments are merely illustrative of specific embodiments of the present invention, and the present invention is not limited to the description of the above-described embodiments.
Claims (8)
1. A lightweight video analysis system based on an embedded platform is used for carrying out lightweight processing on streaming media data so as to obtain and output a video analysis result, and is characterized by comprising:
the streaming media module is used for acquiring the streaming media data and preprocessing the streaming media data to obtain preprocessed data, wherein the streaming media data at least comprises video streaming data;
the video analysis module is used for continuously receiving the preprocessed data and analyzing according to a preset video analysis process to obtain a video analysis result;
the data persistence module is used for performing persistence processing on the video analysis result and storing the video analysis result as dynamic data;
the API access module is used for monitoring an external request of a user; and
a routing module for routing the data packets to the network,
the streaming media module, the video analysis module, the data persistence module, the API access module, and the routing module are deployed on a plurality of different edge nodes through a container virtualization technology, and each edge node is virtualized by the container virtualization technology to form a virtual node, so that the streaming media module, the video analysis module, and the data persistence module run in each virtual node, and the API access module provides a uniform API interface to the outside and distributes messages through the routing module, so that the streaming media module, the video analysis module, the data persistence module, the API access module, and the routing module communicate among the plurality of different edge nodes, and complete video analysis processing by using computing resources of the edge nodes.
2. The embedded platform based lightweight video analytics system of claim 1, wherein:
wherein, once the API access module hears the external request, the external request is parsed into an internal message and the legitimacy of the external request is authenticated,
if the external request is a legal request, the API access module sends the corresponding internal message to the routing module and enables the routing module to distribute according to the message header of the internal message;
and if the external request is an illegal request, the API access module responds to the external request according to a preset program.
3. The embedded platform based lightweight video analytics system of claim 2, wherein:
wherein the external request is a request for creating a video stream for processing a video stream,
the routing module distributes the internal message corresponding to the create video stream request to the streaming media module,
the streaming media module creates a corresponding work thread according to the received internal message,
the working thread is used for carrying out stream pulling, stream taking and decoding preprocessing on the streaming media module and packaging the obtained preprocessing data into an internal message corresponding to image analysis.
4. The embedded platform based lightweight video analytics system of claim 3, wherein:
wherein the routing module distributes the internal message of the corresponding image analysis to the video analysis module,
and once the video analysis module completes analysis and obtains the video analysis result, the API access module outputs the video analysis result according to the video stream creation request.
5. The embedded platform based lightweight video analytics system of claim 2, wherein:
wherein the external request is a data input request for inputting static data by a user,
the routing module distributes an internal message corresponding to the data entry request to the data persistence module,
and the data persistence module performs persistence processing according to the static data in the received internal message and stores the data as the static data.
6. The embedded platform based lightweight video analytics system of claim 5, wherein:
the static data comprises user information, equipment information and base information.
7. The embedded platform based lightweight video analytics system of claim 1, further comprising:
and the result display part is used for displaying the video analysis result through a display page.
8. The embedded platform based lightweight video analytics system of claim 1, wherein:
the video analysis process is a face recognition process, a behavior detection process or a vehicle detection process.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010074352.7A CN111225280B (en) | 2020-01-22 | 2020-01-22 | Lightweight video analysis system based on embedded platform |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010074352.7A CN111225280B (en) | 2020-01-22 | 2020-01-22 | Lightweight video analysis system based on embedded platform |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111225280A CN111225280A (en) | 2020-06-02 |
| CN111225280B true CN111225280B (en) | 2021-10-01 |
Family
ID=70829642
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010074352.7A Active CN111225280B (en) | 2020-01-22 | 2020-01-22 | Lightweight video analysis system based on embedded platform |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111225280B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112309550A (en) * | 2020-11-11 | 2021-02-02 | 复旦大学 | Multidimensional medical image data storage management system |
| US20220165304A1 (en) * | 2020-11-24 | 2022-05-26 | Nvidia Corporation | Trigger-responsive clip extraction based on remote analysis |
| CN114286127B (en) * | 2022-03-08 | 2022-05-27 | 浙江微能科技有限公司 | Distributed artificial intelligence analysis method and device |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101494655A (en) * | 2009-03-12 | 2009-07-29 | 中国电信股份有限公司 | RTP distributed stream media service system and method |
| CN107819704A (en) * | 2017-11-24 | 2018-03-20 | 量子云未来（北京）信息科技有限公司 | A kind of wireless extendible media application edge calculations device, system and method |
| CN108600777A (en) * | 2018-04-09 | 2018-09-28 | 北京大米科技有限公司 | A kind of stream media playback server, client |
| CN109002744A (en) * | 2017-06-06 | 2018-12-14 | 中兴通讯股份有限公司 | Image-recognizing method, device and video monitoring equipment |
| CN109815371A (en) * | 2019-01-30 | 2019-05-28 | 无锡华云数据技术服务有限公司 | Video monitoring data stores access system |
| JP6561632B2 (en) * | 2015-07-06 | 2019-08-21 | 日本電気株式会社 | Edge server and method thereof |
| CN110175546A (en) * | 2019-05-15 | 2019-08-27 | 深圳市商汤科技有限公司 | Image processing method and device, electronic equipment and storage medium |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10110495B1 (en) * | 2017-11-22 | 2018-10-23 | Intel Corporation | Multi-access edge computing (MEC) service provision based on local cost measurements |
| CN109657588A (en) * | 2018-12-11 | 2019-04-19 | 上海工业自动化仪表研究院有限公司 | Intelligent edge calculations built-in terminal based on video identification |
| CN110290140B (en) * | 2019-06-28 | 2021-09-24 | 腾讯科技（深圳）有限公司 | Multimedia data processing method and device, storage medium and electronic equipment |
-
2020
- 2020-01-22 CN CN202010074352.7A patent/CN111225280B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101494655A (en) * | 2009-03-12 | 2009-07-29 | 中国电信股份有限公司 | RTP distributed stream media service system and method |
| JP6561632B2 (en) * | 2015-07-06 | 2019-08-21 | 日本電気株式会社 | Edge server and method thereof |
| CN109002744A (en) * | 2017-06-06 | 2018-12-14 | 中兴通讯股份有限公司 | Image-recognizing method, device and video monitoring equipment |
| CN107819704A (en) * | 2017-11-24 | 2018-03-20 | 量子云未来（北京）信息科技有限公司 | A kind of wireless extendible media application edge calculations device, system and method |
| CN108600777A (en) * | 2018-04-09 | 2018-09-28 | 北京大米科技有限公司 | A kind of stream media playback server, client |
| CN109815371A (en) * | 2019-01-30 | 2019-05-28 | 无锡华云数据技术服务有限公司 | Video monitoring data stores access system |
| CN110175546A (en) * | 2019-05-15 | 2019-08-27 | 深圳市商汤科技有限公司 | Image processing method and device, electronic equipment and storage medium |
Non-Patent Citations (2)
| Title |
|---|
| 基于MEC的智能视频管理系统研究;李素游;《中国优秀硕士学位论文-信息科技辑》;20181015;12-61 * |
| 饶着译.移动边缘计算及视频分析应用研究.《中国优秀硕士学位论文-信息科技辑》.2018, * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111225280A (en) | 2020-06-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111225280B (en) | Lightweight video analysis system based on embedded platform | |
| CN111818136B (en) | Data processing method, device, electronic equipment and computer readable medium | |
| CN108200190A (en) | Internet of Things data service system and method based on cloud computing | |
| CN110347694B (en) | Equipment monitoring method, device and system based on Internet of things | |
| CN110719215B (en) | Flow information acquisition method and device of virtual network | |
| CN112291724A (en) | 5G signaling visualization method and device | |
| WO2022062860A1 (en) | Data processing method, apparatus and device for point cloud media, and storage medium | |
| CN112560724A (en) | Vehicle monitoring method and device and cloud control platform | |
| CN112671697A (en) | Data processing method, device and system of comprehensive monitoring system | |
| CN114372084A (en) | Real-time processing system for sensing stream data | |
| CN110611937B (en) | Data distribution method and device, edge data center and readable storage medium | |
| CN107493299A (en) | A kind of user behavior source tracing method based on three-tier architecture | |
| CN114490280A (en) | Log processing method, device, equipment and medium | |
| CN108134746B (en) | Method and device for processing rail transit data | |
| US10693736B2 (en) | Real time simulation monitoring | |
| CN115567563B (en) | Comprehensive transportation hub monitoring and early warning system based on end edge cloud and control method thereof | |
| CN113852841A (en) | Visual scene establishing method, device, equipment, medium and system | |
| CN112152879A (en) | Network quality determination method and device, electronic equipment and readable storage medium | |
| CN111506769A (en) | Video file processing method and device, storage medium and electronic device | |
| CN110830822A (en) | Resource updating method, device and computer readable storage medium | |
| CN115994156A (en) | Method and system for real-time analysis of data streams | |
| CN113687958A (en) | Data processing method, system, computer device and storage medium | |
| KR20220071544A (en) | System and method of processing image data for batch processing of big data and computer program for the same | |
| CN112035316A (en) | Method for monitoring and analyzing data by cloud monitoring system | |
| CN112100414A (en) | Data processing method, device, system and computer readable storage medium |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |