CN111444014A

CN111444014A - Ocean observation network system and method based on EC-REST system structure

Info

Publication number: CN111444014A
Application number: CN202010207716.4A
Authority: CN
Inventors: 胡一帆; 郑轶; 刘海林; 吕斌; 陈杰; 袁健; 吕成兴
Original assignee: Oceanographic Instrumentation Research Institute Shandong Academy of Sciences
Current assignee: Oceanographic Instrumentation Research Institute Shandong Academy of Sciences; Institute of Oceanographic Instrumentation Shandong Academy of Sciences
Priority date: 2020-03-23
Filing date: 2020-03-23
Publication date: 2020-07-24

Abstract

The disclosure discloses a marine observation network system and a method based on an EC-REST system structure, wherein an edge sensing device for sensing a data source of marine observation data performs primary processing on the acquired data and stores the data; the data acquisition layer acquires and transmits the ocean observation data sensed by the data source and transmits the acquired data to the data storage layer; the data storage layer pre-processes the data transmitted from the data acquisition layer by using the edge gateway and uploads the data to the data processing layer; the data processing layer realizes unified and coordinated migration and scheduling of resource loads such as computing tasks and the like among the sensing devices which are distributed and deployed and among the sensing devices and the edge gateway through an EC-REST (enhanced computer-response technology) architecture for data uploaded by the edge gateway; and the data application layer applies the result processed by the data processing layer. The interactivity and resource load balance of all sensing devices in the ocean observation network are kept, and the high-efficiency low-energy consumption real-time observation of ocean medium-sea heterogeneous data is realized.

Description

Ocean observation network system and method based on EC-REST system structure

Technical Field

The disclosure relates to the technical field of marine observation, in particular to a marine observation network system and method based on an EC-REST system structure.

Background

The statements in this section merely provide background information related to the present disclosure and may not constitute prior art.

The construction target of the marine observation network is to establish an air, sky, ground and sea integrated marine three-dimensional sensing network based on fixed platforms such as buoys, submerged buoys and seabed observation nodes and mobile platforms such as satellite remote sensing, unmanned aerial vehicles, unmanned boats, shore-based stations and ship-based platforms, establish a marine data communication network fused in various modes such as broadband network and wireless communication, and initially form a three-dimensional sensing and communication network of environmental parameters such as weather, hydrology, power and ecology; the comprehensive three-dimensional real-time monitoring, multi-time and space resolution and multi-dimensional marine observation data of the marine environment can be acquired in time. Meanwhile, the marine observation network needs a rapid prediction mechanism and information propagation so as to cope with various emergencies such as tsunami and take corresponding measures. Sensor devices deployed on the ocean generate large amounts of data, consuming large amounts of resources and transmission bandwidth. Processing such data on the cloud faces delays. Edge calculations, however, can prevent data loss or sensor data transmission delays by processing nearby near the data source, helping the system to respond quickly.

The prior art scheme of the marine observation network has the problems of large scale of observation data, wide range, limited communication bandwidth, difficult maintenance of observation equipment, strong heterogeneity, less interaction, low efficiency and the like. Aiming at the problem of low integration efficiency of the current marine observation data, the inventor finds that the Shandong province academy of sciences puts forward a marine observation element self-matching method based on edge calculation. The edge computing device is responsible for storing the obtained observation data, preprocessing the data according to the selected training model to obtain a characteristic value, and transmitting the characteristic value obtained in a specified time period to the cloud. In the scheme, the device interactivity is less, a flexible resource load scheduling method is lacked, the system energy consumption cannot be effectively reduced, and the load balance of each device is realized.

Disclosure of Invention

In order to solve the deficiencies of the prior art, the present disclosure provides a marine observation network system and method based on an EC-REST architecture; resource loads in domains are uniformly scheduled and managed by adopting an REST distributed architecture, and meanwhile, the edge sensing equipment and the edge gateway can reduce the overall energy consumption of the system as far as possible on the premise of meeting the delay requirement of application, keep the resource loads of all equipment balanced and respond to emergency observation tasks in time.

In a first aspect, the present disclosure provides a marine observation network system based on an EC-REST architecture;

ocean observation network system based on EC-REST architecture includes:

the data source is used for perceiving marine observation data, distributed sensing equipment is arranged in the data source, and the sensing equipment comprises: the edge sensing device is used for carrying out primary processing on the acquired data and storing the data;

the data acquisition layer is used for acquiring and transmitting the ocean observation data sensed by the data source, and various devices transmit the acquired data to the data storage layer;

the data storage layer is used for preprocessing the data transmitted by the data acquisition layer by utilizing the edge gateway for format unification and data fusion, and uploading the preprocessed data to the data processing layer;

the data processing layer is used for realizing migration and scheduling of resource loads among the sensing devices which are distributed and deployed and among the sensing devices and the edge gateway through an EC-REST (enhanced logical control-representational state transfer) architecture for data uploaded by the edge gateway;

and the data application layer is used for applying the result processed by the data processing layer.

In a second aspect, the present disclosure provides a working method of a marine observation network system based on an EC-REST architecture;

the working method of the ocean observation network system based on the EC-REST system structure comprises the following steps:

the data source perception ocean observation data is equipped with the perception equipment of distributed deployment in the data source, the perception equipment includes: the edge sensing device is used for carrying out primary processing on the acquired data and storing the data;

the data acquisition layer acquires and transmits ocean observation data sensed by the data source, and various devices transmit the acquired data to the data storage layer;

the data storage layer performs preprocessing of format unification and data fusion on the data transmitted from the data acquisition layer by using the edge gateway, and uploads the preprocessed data to the data processing layer;

the data processing layer realizes migration and scheduling of resource loads among the sensing devices which are distributed and deployed and between the sensing devices and the edge gateway through an EC-REST (enhanced computer-response technology) architecture for data uploaded by the edge gateway;

and the data application layer applies the result processed by the data processing layer.

In a third aspect, the present disclosure provides a marine observation network system for underwater video acquisition based on an EC-REST architecture;

ocean observation network system for underwater video acquisition based on EC-REST architecture comprises:

the video stream acquisition layer is used for acquiring underwater video streams in the ocean; a video stream acquisition layer comprising: a distributively deployed camera device, comprising: the device comprises edge camera equipment and common camera equipment, wherein the edge camera equipment is provided with a processor; the common camera is not provided with a processor;

the video analysis layer is used for processing the collected underwater video stream in the ocean;

the video control layer is used for scheduling resources of the video processing task among the distributed camera equipment and the edge gateway based on an EC-REST system structure;

and the video application layer is used for applying the video data after the analysis processing.

In a fourth aspect, the present disclosure provides a working method of a marine observation network system for underwater video acquisition based on an EC-REST architecture;

the working method of the ocean observation network system for underwater video acquisition based on the EC-REST system structure comprises the following steps:

the cloud server cluster reads observation historical data of marine biological characteristics of known labels, and characteristic parameter ranges extracted from the observation historical data are prestored in a database;

after the edge camera acquires new observation video data, directly performing video preprocessing, and uploading the preprocessed observation video data to an edge gateway; the common camera directly uploads the observed video data to an edge gateway for processing;

the edge gateway receives and stores new observation video data, and the edge gateway realizes migration and scheduling of resource loads among the distributed deployed camera equipment and between the distributed deployed camera equipment and the edge gateway through an EC-REST (enhanced traffic control-REST) architecture; the edge camera extracts characteristic parameters from the new observation video data and feeds the characteristic parameters back to the edge gateway;

the edge gateway matches the extracted characteristic parameters with a characteristic parameter range prestored in a database to obtain effective biological activity state characteristic parameters and biological activity video data, and transmits the effective biological activity state characteristic parameters and the biological activity video data to a cloud server;

the cloud server stores the effective video data and the biological characteristic parameters in a database and provides the effective video data and the biological characteristic parameters to the visual application module so as to perform visual playing demonstration on the user.

Compared with the prior art, the beneficial effect of this disclosure is:

1. by introducing an edge computing mechanism into the ocean observation network, partial operations of application programs, data materials and services processed in a data center background are moved down to an edge gateway for processing, and the edge gateway is closer to a data source, so that data transmission quantity can be reduced, transmission delay is reduced, and ocean observation efficiency is improved.

2. The specific design and implementation of the equipment end and the distributed system thereof are supported through a unified and coordinated EC-REST distributed system architecture abstraction, so that resource loads such as computing tasks and the like migrated by the cloud server end are borne; and sensing the heterogeneous diversity of the equipment end, and dynamically scheduling the system according to the resource type and the computing capability of the sensing equipment, so that the computing resources on the sensing equipment are scheduled for the application program, and the minimization of data transmission overhead and the maximization of the execution performance of the application program are realized.

3. The ocean observation data is large in scale, wide in range and limited in communication bandwidth, the observation data is difficult to upload to a data center rapidly in a large scale, the edge calculation is processed nearby in a place close to a data source, local preprocessing and storage can be conducted on massive observation data, only a small amount of data is transmitted to a cloud server to be analyzed, and therefore the bandwidth is effectively utilized. Meanwhile, mass data can be accumulated on the edge side, and the system can effectively reduce the data transmission amount of each level of edge side through the data fusion and preprocessing of two levels of edge sides such as edge sensing equipment and edge gateways, and share the data storage and processing pressure together.

4. The monitoring device is difficult to maintain and high in cost, the state of the device can be monitored locally in real time through the edge gateway, and the state information of the control device is updated regularly and synchronously with the cloud platform, so that remote maintenance management and fault diagnosis of the device and the sensor are realized.

5. The observation equipment has multiple types, strong isomerism, and different data acquisition mechanisms and communication modes. The EC-REST architecture can provide a uniform adaptive interface for heterogeneous equipment through a uniform and coordinated distributed system structure, and the influence of the diversity of a bottom sensor on upper-layer application is shielded. And analyzing the data resources, and realizing data format unification through abstract modeling.

6. The observation equipment has less interactivity and more unidirectional operations, and the resource loads such as access and interactive computing tasks among the sensing equipment and between the sensing equipment and the edge gateway can be controlled through a unified scheduling and management mechanism of EC-REST to the resources and the loads, so that the interactivity of the sensing equipment is improved.

Drawings

The accompanying drawings, which are included to provide a further understanding of the disclosure, illustrate embodiments of the disclosure and together with the description serve to explain the disclosure and are not to limit the disclosure.

FIG. 1 is a schematic diagram of the EC-REST architecture of the present disclosure;

fig. 2 is a schematic structural diagram of an underwater video intelligent acquisition system in an embodiment of the present disclosure.

Detailed Description

It should be noted that the following detailed description is exemplary and is intended to provide further explanation of the disclosure. Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this disclosure belongs.

It is noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of example embodiments according to the present disclosure. As used herein, the singular forms "a", "an" and "the" are intended to include the plural forms as well, and it should be understood that when the terms "comprises" and/or "comprising" are used in this specification, they specify the presence of stated features, steps, operations, devices, components, and/or combinations thereof, unless the context clearly indicates otherwise.

The embodiments and features of the embodiments in the present disclosure may be combined with each other without conflict.

Interpretation of terms:

edge Computing (EC)

Autonomous Underwater Vehicle (AUV for short)

EC-REST (Edge Computing-Representational State Transfer, EC-REST for short)

The embodiment I provides a marine observation network system based on an EC-REST architecture;

as shown in fig. 1, the ocean observation network system based on the EC-REST architecture includes:

Further, the data source refers to mass, multi-source and heterogeneous ocean data which are stored in a scattered mode.

Further, the resource load includes: and the tasks to be calculated, the data to be stored and the device access and device disconnection management tasks.

Further, the implementing migration and scheduling of resource load between sensing devices deployed in a distributed manner and between the sensing devices and the edge gateway through the EC-REST architecture includes:

the task to be calculated is realized through an EC-REST framework, and the task is migrated to sensing equipment from an edge gateway; alternatively, the first and second electrodes may be,

the task to be calculated is realized through an EC-REST framework, and the task is migrated from the sensing equipment to the edge gateway; alternatively, the first and second electrodes may be,

tasks to be calculated are achieved through an EC-REST framework, and the tasks are migrated from current sensing equipment to other sensing equipment; alternatively, the first and second electrodes may be,

data to be stored are migrated to the sensing equipment from the edge gateway through an EC-REST architecture; alternatively, the first and second electrodes may be,

data to be stored are migrated to the edge gateway from the sensing equipment through an EC-REST architecture; alternatively, the first and second electrodes may be,

data to be stored are migrated from the current sensing equipment to other sensing equipment through an EC-REST architecture; alternatively, the first and second electrodes may be,

and the management of the sensing equipment accessing to the network or the sensing equipment disconnecting from the network is realized through an EC-REST architecture.

Further, the non-edge sensing device includes: a meteorological sensor, a hydrological sensor, a power sensor or a water quality sensor; the non-edge sensing device has no data processing and storage functions.

Further, the edge perception device includes: the equipment has data processing and storing functions and is carried by fixed platforms such as buoys, submerged buoys and seabed observation nodes and mobile platforms such as satellite remote sensing, unmanned aerial vehicles, unmanned boats, underwater AUVs, shore base stations and ship base platforms.

It should be appreciated that the edge aware devices are used to pre-process and store the collected data in a distributed manner, thereby reducing the task burden on the edge gateway.

As one or more embodiments, the data acquisition layer comprises:

an underwater acoustic communication network, a marine communication network, a data collector, an optical fiber, a cable, a satellite communication device, a 4G or 5G mobile communication device.

Furthermore, the data acquisition layer acquires, transmits, migrates and distributes data of a data source through an underwater acoustic communication network, a marine communication network, a data acquisition device, an optical fiber, a cable, a satellite communication device or a mobile communication device, and acquires and uploads scattered marine information data to the edge gateway.

As one or more embodiments, the data store layer, comprises: an edge gateway and a cloud server;

the edge gateway is used for performing data preprocessing and data fusion on data stored in the edge sensing equipment and data uploaded by the non-edge sensing equipment locally to realize unification of data formats and uploading the data after data fusion to the cloud server;

the edge gateway is also used for moving down the application program, data information and partial operation processed in the cloud server to the distributed sensing equipment for processing;

and the cloud server is used for storing, analyzing, processing and applying the data uploaded by the edge gateway.

It should be understood that, since the edge gateway is closer to the data source, the data transmission amount can be reduced and the transmission delay can be reduced through data preprocessing.

It should be understood that, due to different sampling transmission rules of various sensors, the edge gateway performs uniform processing on heterogeneous original data streams, performs abstract modeling on sensing data, and unifies data formats.

It should be understood that the edge gateway is responsible for locally storing and analyzing mass data, and only transmits effective observation data to the cloud server for further analysis and long-term storage, thereby reducing data transmission to the cloud server.

It should be appreciated that the vast amount of video data is deposited on the local edge gateway, ensuring that private data is not leaked.

It should be understood that the linkage between the edge sensing devices and the sensor devices may also be coordinated in real time through the local edge gateway.

It should be understood that the sensor will be in an idle state in a non-working state, and the edge gateway can be fully utilized, thereby improving the utilization rate of resources.

It should be understood that the sensing devices and the edge gateways deployed in a distributed mode can periodically and synchronously update the state information of the control device with the cloud server so as to perform real-time diagnosis on the device fault.

And information interaction between different devices is realized between the edge sensing device which is distributed and deployed and other sensor devices through an EC-REST (enhanced emission control) architecture.

It should be understood that the above scheme can alleviate resource over consumption brought to upper-layer equipment by diversity and disorder of the sensors and various observation platforms.

Further, the migrating and scheduling of the resource load between the sensing devices deployed in a distributed manner and between the sensing devices and the edge gateway is realized through an EC-REST architecture for the data uploaded by the edge gateway, which specifically includes:

adding a calculation task in the edge gateway and the perception equipment according to the user requirement; for example: adding a picture identification task to an edge gateway, and adding a video frame capturing task to edge camera equipment;

according to the device performance and network bandwidth deployment computing tasks and other resource loads of different observation platforms and sensors in the non-edge sensing device, response delay is reduced;

and (3) configuring computing resources: when different sensors are selected to execute a certain task, the Uniform Resource Identifiers (URIs) of the sensors in the task are only needed to be reconfigured; the access to various sensing devices at any time is convenient, and the sensing devices with different communication protocols and modes are accessed to the edge gateway;

unified scheduling and management of resources and computational resource loads within the network: and the access and interaction resources and the computing resource load between the sensing devices and the edge gateway are supported.

through four operations of GET, PUT, POST and DE L ETE, sensing equipment and a calculation task specified by a user are subjected to increasing, deleting, modifying, checking and scheduling;

in the aspect of realizing resource scheduling deployment, when the task A needs to be scheduled from the sensor B to the sensor C, the task A is scheduled from the sensor B to the sensor C

The edge gateway sends GET to URI address of sensor B to obtain computing resource information of task A;

the edge gateway sends a POST to the URI address of sensor C to create task a resources and enters the task queue of sensor C, and at the same time,

the edge gateway sends a sensor URI corresponding to the PUT updating task A resource to the task A; finally, the process is carried out in a batch,

and the edge gateway sends a PUT (push update) calculation resource list to the data processing layer to finish the scheduling of the calculation resources.

Alternatively, the first and second electrodes may be,

when a new sensor D needs to be deployed, the sensor D sends POST to the edge gateway to create a URI address of a sensor D resource, the edge gateway sends a POST response to the URI of the sensor D and allocates a new task to the sensor D, and sends a PUT update resource list to the data processing layer to complete resource deployment.

Alternatively, the first and second electrodes may be,

when a task F needs to be added to the sensor E, the edge gateway sends POST to the URI of the sensor E to create F resources, the sensor E sends POST response, the edge gateway sends PUT to F to update the sensor URI corresponding to the F resources, and finally, the PUT update resource list is sent to the data processing layer to complete the deployment of the computing resources.

The resource calculation load can be freely scheduled and migrated, and the scheduling aims to reduce the total power consumption level of the system while meeting the time delay requirements of different resource calculation loads.

As one or more embodiments, the data processing layer includes:

the system comprises a marine data processing module, an equipment fault diagnosis module or an REST resource scheduling and deploying module;

the ocean data processing module is used for classifying or analyzing and processing ocean data so as to match different types of data and facilitate storage and application;

the equipment fault diagnosis module is used for carrying out fault diagnosis and processing on various sensor equipment and edge sensing equipment;

the REST resource scheduling and deploying module is used for carrying out unified and coordinated deployment, migration and scheduling on sensor equipment, edge sensing equipment, edge gateways and computing resources of the edge gateways.

As one or more embodiments, the data application layer includes:

a data management module;

the data management module comprises: a marine environment monitoring submodule, a marine ecological protection submodule, a marine disaster early warning submodule, a marine ranch submodule or a marine fishery submodule;

the data management module is background software used for data management and visual demonstration of users in marine environment monitoring, marine ecological protection, marine disaster early warning and marine ranch application.

It should be understood that the data application layer performs personalized processing on the marine data according to specific business requirements, realizes visualization of the data, and performs data management and application development through background data management software.

The data application fields comprise marine environment monitoring, marine ecological protection, marine disaster early warning, marine ranching, marine fishery, marine traffic and the like.

The EC-REST framework is mainly oriented to the field of marine environment monitoring, and can provide a data resource sharing interface for development of other various applications, so that new applications are expanded according to requirements.

The EC-REST architecture system includes a plurality of entities: users, devices, sensors, tasks, data types, to perform various observation operations with respect to a marine observation network. The relationship is as follows:

(1) each user has a device with data resource processing capability, and sensor data is accessed to the network through the device. The device comprises sensing devices such as edge sensing devices and other sensor devices. An edge gateway is also a device.

(2) Each user can add, delete, change and check any number of devices, sensors, sensor tasks and the like, and each device and sensor can be accessed by any user.

(3) Each device, sensor task, data and type corresponds to a unique URI address.

(4) Each device possesses multiple types of sensors, and the sensor data types include a numerical type, a switch type, an image type, and the like.

(5) Each type of sensor possesses data information, which is the data generated at a particular time.

(6) Each sensor has multiple tasks, each task performing a different designated task.

(7) Each device, sensor, and generated data can be viewed as a resource. A task is also a computing resource.

The model supports four resource request modes, namely GET, PUT, POST, DE L ETE, GET obtaining resource list, PUT updating resource, POST establishing resource, DE L ETE deleting resource.

If the value of a certain sensor is to be obtained, a GET request is sent to a URI corresponding to the sensor;

if a device is to be created, constructing a URI containing the device name, and then sending a PUT request to the URI;

if a sensor or a task of a sensor is to be deleted, a DE L ETE request is sent to its URI.

The beneficial effects of the above embodiment are:

(1) the EC-REST system structure designed by the disclosure can carry out local preprocessing and storage on mass observation data through the edge sensing equipment and the edge gateway, and only a small amount of data is transmitted to the cloud server for analysis and storage, so that the bandwidth is effectively utilized, and the data transmission quantity is reduced.

(2) According to the edge sensing equipment and the edge gateway, the state of the equipment is monitored in real time locally, and the state information of the control equipment is updated synchronously with the cloud server periodically, so that the maintenance management and the fault diagnosis of each sensing equipment are realized.

(3) The EC-REST system structure can provide a uniform adaptive interface for heterogeneous equipment through a uniform and coordinated distributed system, and shields the influence of the diversity of a bottom sensor on upper application. And analyzing the data resources, and realizing data format unification through abstract modeling.

(4) According to the method and the system, resource loads such as mutual access and interactive computing tasks among sensing devices and between the sensing devices and the edge gateway can be controlled through a unified scheduling and management mechanism of resources and loads, and the interactivity of the sensing devices is improved.

The second embodiment provides a working method of the ocean observation network system based on the EC-REST architecture;

The embodiment provides a marine observation network system for underwater video acquisition based on an EC-REST (engineering environmental protection and response) architecture;

as shown in fig. 2, the ocean observation network system for underwater video acquisition based on EC-REST architecture comprises:

And the video analysis layer and the video control layer form an edge layer.

As one or more embodiments, the video stream acquisition layer includes:

the edge camera equipment is used for acquiring a real-time video stream and analyzing the video stream into corresponding video frames; IP addresses of different edge cameras are different;

and the common camera is used for acquiring the real-time video stream and uploading the video stream to the edge gateway for processing.

The edge camera apparatus includes: and the processor is connected with the camera and the signal transmission unit. The processor is used for realizing the calculation task of video analysis processing.

The feature data can be actively identified and processed relative to the edge camera equipment, the common camera is not suitable for processing tasks of video analysis and identification due to the fact that no processor exists, and the common camera directly transfers resource loads such as computing tasks to the edge gateway for processing.

As one or more embodiments, the video analytics layer includes:

an edge gateway configured to: acquiring and processing images and videos from distributed camera equipment, preprocessing image and video data, connecting the images and the video data from the inside of a local area network to a public network through a Webserver module, and sending the video data to the public network;

a data pre-processing module configured to: preprocessing the data, removing redundant data, fusing and classifying the data, and then sending the data to a target detection feature extraction module or sending the data to a fault diagnosis module for fault diagnosis processing;

a target detection feature extraction module configured to: acquiring video data from an edge gateway, identifying and storing videos, starting to store the biological activity videos when biological characteristic targets such as sea cucumbers are identified, extracting characteristic values, and sending the videos and the characteristic parameters to a cloud server; stopping storing the continuous N frames of images if the images do not contain the biological characteristic target; n is a set value and is a positive integer;

a fault diagnosis module configured to: and obtaining data such as voltage and current of the sensing equipment from the data preprocessed by the edge gateway, and performing fault diagnosis and reporting to the cloud server.

An edge gateway configured to: and video data are acquired from the camera, preprocessed and sent to the target detection feature extraction module and the fault diagnosis module, and meanwhile, the local area network is connected to the public network from the inside, and the video data are sent to the public network.

In a common camera interaction scene, the edge gateway has a hardware facility for identifying and processing video classification. The video can be collected and preprocessed by a camera, and the obtained video data is transmitted to a video control layer and a public network for further processing.

The operation of the video analysis layer includes preprocessing of video frames, image processing of video frames, and the like. The edge gateway performs target detection and feature extraction and recognition on organisms in the video images according to the analysis results so as to track the activity states of the organisms, can also realize the calculation load migration of equipment fault diagnosis, and performs fault diagnosis and management on the camera by periodically and synchronously updating the state information of the camera equipment with the cloud server.

As one or more embodiments, the video control layer includes:

a resource monitoring module configured to: monitoring resources, monitoring the configurable resource conditions of each resource and computing task, such as the memory and CPU of an edge gateway, and transmitting the configurable resource conditions to an REST resource scheduling and deploying module;

a data storage module configured to: storing the data processed by the edge gateway;

a REST resource scheduling deployment module configured to: carrying out unified and coordinated migration and scheduling on the computing resources among the camera devices and in the camera devices and the edge gateway by utilizing the data reported by the resource monitoring module;

a Webserver module configured to: video visualization service is provided for users through a web protocol, and a multi-user video visualization task is dynamically supported.

In the video control layer, resource calculation load can be dynamically divided and scheduled during running, and calculation and transmission power consumption is reduced.

As one or more embodiments, the video application layer includes:

a cloud server configured to: based on strong computing power, massive video streams uploaded by the edge gateway are analyzed and processed, detection, identification and tracking of specific organisms are achieved, and effective video data are stored in a database or provided for a video visualization application module.

A video visualization application module configured to: and providing a visual interface, playing and demonstrating the video data after analysis and processing for a user, and simultaneously carrying out interactive operations such as video retrieval and the like for the user.

A database configured to: and storing and managing the video data analyzed and processed by the cloud server.

The video application layer is used for displaying and applying video analysis results, for example, the video analysis results are visually presented through a web visualization technology, so that real-time interaction with a user is facilitated. The system can store and manage the video content cached in the edge gateway based on the video analysis result, and the video real-time analysis result and the video retrieval interface can provide users for use through a visual interface.

The embodiment provides a working method of an ocean observation network system for underwater video acquisition based on an EC-REST (engineering environmental protection and response) architecture;

s1: the cloud server cluster reads observation historical data of marine biological characteristics of known labels, and characteristic parameter ranges extracted from the observation historical data are prestored in a database;

s2: after the edge camera acquires new observation video data, directly performing video preprocessing, and uploading the preprocessed observation video data to an edge gateway; the common camera directly uploads the observed video data to an edge gateway for processing;

s3: the edge gateway receives and stores new observation video data, and the edge gateway realizes migration and scheduling of resource loads among the distributed deployed camera equipment and between the distributed deployed camera equipment and the edge gateway through an EC-REST (enhanced traffic control-REST) architecture; the edge camera extracts characteristic parameters from the new observation video data and feeds the characteristic parameters back to the edge gateway;

s4: the edge gateway matches the extracted characteristic parameters with a characteristic parameter range prestored in a database to obtain effective biological activity state characteristic parameters and biological activity video data, and transmits the effective biological activity state characteristic parameters and the biological activity video data to a cloud server;

s5: the cloud server stores the effective video data and the biological characteristic parameters in a database and provides the effective video data and the biological characteristic parameters to the visual application module so as to perform visual playing demonstration on the user.

As one or more embodiments, the edge gateway realizes migration and scheduling of resource loads between the distributed deployed camera devices and the edge gateway through an EC-REST architecture, and the specific steps include:

firstly, assuming distributed deployed camera equipment, namely a camera A and a camera B;

when the task of the camera A is full and the designated high-definition video recording task cannot be completed, the edge gateway schedules a calculation task C for recording the high-definition video of the camera A to a neighbor camera B;

the edge gateway firstly sends GET to the URI address of the camera A to acquire the computing resource C of the camera A, sends POST to the URI address of the camera B to create the computing resource C for high-definition video recording and enters a task queue of the camera B, simultaneously sends PUT to the computing task C to update the URI of the camera B corresponding to the computing task C, and finally sends PUT to update a computing resource list of a cloud server to complete the scheduling of the computing resource.

In a marine observation network, a plurality of cameras are usually deployed to monitor a specified target, such as real-time monitoring of biological activity conditions of sea cucumbers and the like in a marine ranch.

The surveillance process entails first performing object detection and feature recognition on a given living being, and then recording a video of the living being's activity.

The underwater camera can transfer videos through the edge gateway, intelligently identifies and stores the videos, and meanwhile, low-delay visual online monitoring can be provided for users.

The method comprises the steps of firstly storing and analyzing mass video data by an edge gateway, only intercepting videos of biological activities such as sea cucumbers and the like, and then transmitting the videos to the edge gateway/cloud for further analysis and long-term storage, so that data transmission to the cloud is reduced.

The edge gateway deployed in the local area offloads part of the traffic, reduces the requirements on the transmission network and the core network bandwidth, directly processes and responds to the user, meets the requirements on high bandwidth and low time delay, and reduces the network load.

The above description is only a preferred embodiment of the present disclosure and is not intended to limit the present disclosure, and various modifications and changes may be made to the present disclosure by those skilled in the art. Any modification, equivalent replacement, improvement and the like made within the spirit and principle of the present disclosure should be included in the protection scope of the present disclosure.

Claims

1. Ocean observation network system based on EC-REST system structure, characterized by including:

2. The system of claim 1, wherein the data store layer comprises: an edge gateway and a cloud server;

3. The system according to claim 1, wherein the migration and scheduling of resource load between sensing devices in distributed deployment and between a sensing device and an edge gateway are realized through an EC-REST architecture for data uploaded by the edge gateway, specifically comprising:

adding a calculation task in the edge gateway and the perception equipment according to the user requirement;

according to the device performance and the network bandwidth of different observation platforms and sensors in the non-edge sensing device, resource load is deployed, and response delay is reduced;

unified scheduling and managing the resource load in the network: and the access and interaction resource load between sensing devices and between the sensing devices and the edge gateway is supported.

4. The system according to claim 1, wherein the migration and scheduling of resource load between sensing devices in distributed deployment and between a sensing device and an edge gateway are realized through an EC-REST architecture for data uploaded by the edge gateway, specifically comprising:

in the aspect of implementation of resource scheduling deployment, when a task A needs to be scheduled from a sensor B to a sensor C, an edge gateway sends a GET to a URI address of the sensor B to acquire computing resource information of the task A;

5. The system of claim 4, wherein when a new sensor D needs to be deployed, the sensor D sends POST to the edge gateway to create the URI address of the sensor D resource, the edge gateway sends POST response to the URI of the sensor D and allocates new task to the sensor D, and sends PUT update resource list to the data processing layer to complete resource deployment;

alternatively, the first and second electrodes may be,

6. The system of claim 1, wherein the data processing layer comprises:

the REST resource scheduling and deploying module is used for carrying out unified and coordinated deployment, migration and scheduling on sensor equipment, edge sensing equipment, edge gateways and computing resources thereof for ocean observation;

the data application layer comprises:

a data management module;

7. The working method of the ocean observation network system based on the EC-REST system structure is characterized by comprising the following steps:

8. Ocean observation network system for underwater video acquisition based on EC-REST system structure, which is characterized by comprising:

9. The working method of the ocean observation network system for underwater video acquisition based on the EC-REST system structure is characterized by comprising the following steps:

10. The method of claim 9, wherein the edge gateway implements migration and scheduling of resource loads between the distributively deployed camera devices and the edge gateway through an EC-REST architecture, and the specific steps include: