Method and system for real scene monitoring and minute-level meteorological data superposition display of meteorological station

Abstract

The invention discloses a meteorological station live-action monitoring and minute-level meteorological data superposition display method and system. The core comprises a data acquisition module, a transmission module, a fusion display module, an optional storage module, an authority management module and the like. A Haikangwei vision 4K camera (24h full-color night vision) and a Yunhua terminal (minute-level meteorological data acquisition) are used for data acquisition; uDP + TCP dual protocols are adopted for transmission, and the optical fiber and the category 6 + cable guarantee stability; a video is decoded by a provincial server, meteorological data is analyzed, timestamps are aligned through an algorithm, a semitransparent label is superposed, and user-defined configuration is supported. Local / provincial storage, multi-terminal access, four-level authority management and alarm linkage can be selected. Through Qinghai trial verification, the end-to-end video delay is smaller than or equal to 500 ms, the data synchronization error is smaller than or equal to 1 ms, the system availability is larger than or equal to 99.9%, the problems that traditional station data is separated from real scenes and the like are solved, and meteorological monitoring early warning and disaster recovery are supported.

Description

TECHNICAL FIELD

[0001] The application relates to a meteorological station information technology visualization technology, in particular to a meteorological station live monitoring and minute-level meteorological data superimposed display method and system. BACKGROUND

[0002] With the deepening of the meteorological modernization construction, the real-time, accuracy and visualization requirements of meteorological observation data are increasing. The traditional meteorological station monitoring lacks intuitive visual monitoring means for superimposing the station site environment, equipment operation state and real-time meteorological observation data, and it is difficult to comprehensively grasp the weather situation, equipment abnormalities and changes in the surrounding environment, which restricts the further improvement of meteorological monitoring and early warning and service efficiency.

[0003] The automatic observation equipment of the traditional meteorological station can only collect meteorological data such as temperature and air pressure, and the video monitoring can only provide the live picture, and the two are independently stored and not associated. Business personnel need to switch between different systems to view, and cannot intuitively obtain the superimposed display information of the picture and data linkage, resulting in low efficiency of judging the weather situation (such as the matching of precipitation intensity and rain intensity in the picture) and equipment abnormalities (such as the correspondence between sensor failure and equipment state in the picture).

[0004] The existing station video monitoring mostly uses a single protocol, and has problems of high transmission delay and poor signaling synchronization. Moreover, the night vision mode is mostly black and white picture with insufficient resolution (mostly 1080P), and under low light (such as at night) or complex weather (such as fog and sandstorm) conditions, the details such as visibility and cloud shape cannot be clearly captured, which restricts the real-time monitoring capability under extreme weather.

[0005] Research on meteorological station live monitoring and meteorological data superimposition technology will further improve the real-time monitoring capability of station weather situation, provide intuitive video data support for meteorological forecasting, disaster warning and public service, provide visual basis for meteorological disaster review and observation data calibration, help the meteorological business to change from “data-driven” to “scene-driven”, and promote the meteorological observation business to move towards digitization and intelligentization. SUMMARY

[0006] The technical problem to be solved by the application is to provide a meteorological station live monitoring and minute-level meteorological data superimposed display method and system which can realize “data-picture-alarm” full-link cooperation and promote the upgrade of meteorological monitoring from “decentralization and passivity” to “integration and initiative” for the case of separation of data and live picture.

[0007] In order to solve the above technical problems, the application provides the following technical solutions: The application discloses a weather station real scene monitoring and minute-level meteorological data superimposed display method, which comprises the following steps: a. data acquisition: collecting real scene video stream through a video camera arranged at the weather station, and transmitting the real scene video stream by using a UDP protocol; reading at least one kind of meteorological element data of the weather station in real time through an open API interface of a meteorological big data cloud platform, wherein the meteorological element data comprises one or more of temperature, air pressure, precipitation, humidity, wind speed, snow depth and visibility, the data update frequency is minute-level and matches the frame rate of the real scene video stream to realize timestamp alignment; b. data transmission and storage: connecting the real scene video stream collected by the video camera to a local video recorder for real-time storage; converging multiple video signals through a convergence switch, transmitting the converged video signals to a provincial video management system in a mode that the UDP protocol is used as a main transmission channel and the TCP protocol is used as a signaling check, and realizing remote backup and centralized storage; c. data fusion and display: aligning the meteorological element data obtained through the open API interface with the parsed video stream through a special data fusion algorithm, mapping the meteorological element data labels to any area of the video picture through pixel coordinates, and presetting alarm thresholds of each meteorological element, so that the system automatically superimposes a red alarm mark on the video picture when the meteorological element data exceeds the alarm threshold.

[0008] The application further provides a weather station real scene monitoring and minute-level meteorological data superimposed display system, which comprises a data acquisition module, a data transmission and storage module and a data fusion and display module; the data acquisition module comprises a video camera and a meteorological data acquisition unit, the video camera is used for collecting real scene video stream and transmitting the real scene video stream by using the UDP protocol, and the meteorological data acquisition unit is used for reading at least one kind of meteorological element data of the weather station in real time through an open API interface of a meteorological big data cloud platform, wherein the meteorological element data comprises one or more of temperature, air pressure, precipitation, humidity, wind speed, snow depth and visibility, the data update frequency is minute-level and matches the frame rate of the real scene video stream to realize timestamp alignment; the data transmission and storage module comprises a local video recorder, a convergence switch and a provincial video management system, the local video recorder is used for storing the real scene video stream in real time, the convergence switch is used for converging multiple video signals and transmitting the converged video signals to the provincial video management system in a mode that the UDP protocol is used as a main transmission channel and the TCP protocol is used as a signaling check, and remote backup and centralized storage are realized; the data fusion and display module is internally provided with a special data fusion algorithm, which is used for realizing timestamp alignment of the meteorological element data and the video stream and data label superimposition, and a preset alarm threshold is used for automatically superimposing a red alarm mark on the video picture when the data exceeds the threshold.

[0009] Preferably, the video camera supports 24-hour full-color display with a resolution of not less than 4K, and is compatible with GB / T28181 national standard signaling to collect device control signaling; the video camera is connected to the station's local area network via a gigabit network port and supports PoE power supply.

[0010] Preferably, the local video recorder supports the H.265 encoding format, has a storage capacity of ≥2TB, and is compatible with H.264, Smart265, and Smart264 video decoding formats; the aggregation switch has Gigabit Ethernet interfaces, including at least 48 Gigabit Ethernet ports and 4 Gigabit optical ports, supports PoE+ power supply, and has a packet forwarding rate of ≥78Mpps / 132Mpps.

[0011] Preferably, the meteorological big data cloud platform is the Qinghai Provincial Meteorological Big Data Cloud Platform (“Tianqing”), the meteorological element data request frequency of the open API interface is once per minute, and the returned data format is JSON; the parsing format of the video stream is RTSP.

[0012] Preferably, the system supports user-defined overlay positions of meteorological element data tags and types of meteorological elements to be displayed, with the meteorological element data tags being semi-transparent. When meteorological element data exceeds a threshold, a local device's audible and visual alarm is triggered simultaneously, along with a message push to the relevant user's client. The alarm response time is ≤1 minute, and the alarm thresholds include wind speed ≥10m / s and visibility ≤1km.

[0013] Preferably, the video camera is a Hikvision DS-2DY54QAZ-HJ model with a frame rate of 25fps; the local recorder is a Hikvision DS-7604N-E1-V3 model; and the aggregation switch is a Hikvision DS-3E2552-H(B) model. The video signal is transmitted to the provincial video management system via optical fiber or twisted pair cable.

[0014] Preferably, it also includes a multi-terminal access unit and a permission management module. The multi-terminal access unit includes a B / S web server, a C / S client server, and a mobile APP server, all of which are connected to the provincial video management system to enable multi-terminal access. The permission management module divides permissions into four levels based on a role-based access control model: system administrator permissions, provincial account permissions, municipal / prefectural bureau account permissions, and county / station-level account permissions.

[0015] Preferably, the alarm linkage module is used to trigger local audible and visual alarms and client message push when meteorological element data exceeds the threshold, with an alarm response time of ≤1 minute. The alarm thresholds include wind speed ≥10m / s and visibility ≤1km. The provincial video management system communicates bidirectionally with the meteorological integrated business real-time monitoring platform, which is used to monitor the status of all-link devices and event logs.

[0016] Preferably, the B / S web server supports historical data retrieval and analysis functions. Users can filter recordings by time range and event type. The event types include equipment alarms and excessive wind speed. The retrieval results can be downloaded and tagged for archiving.

[0017] An invention discloses a method for real-time monitoring of meteorological stations and overlay display of minute-level meteorological data, wherein... Compared with existing technologies, the invention of a real-time monitoring technology for meteorological stations and a minute-level meteorological data overlay display technology has at least the following beneficial effects: 1. Improve the intuitiveness and efficiency of monitoring: By overlaying the real-time monitoring screen of the meteorological station with minute meteorological data elements, the problem of "separation between data and reality" in the traditional way is solved. Business personnel can monitor the real-time weather data of the station in the real-time monitoring screen at the same time. For example, they can quickly judge the intensity of precipitation by "rain intensity screen + minute precipitation".

[0018] 2. Enhanced video acquisition and transmission stability: Adopting dual-track acquisition (data acquisition layer) of UDP protocol + national standard signaling and a transmission scheme of "wired network + protocol optimization" (transmission layer), the video transmission latency is reduced to ≤500ms, which improves the transmission stability by 40% compared with the traditional HTTP protocol; 24-hour full-color 4K resolution output (key technology point 1) solves the shortcomings of blurry and black-and-white night vision images, and improves the detail recognition rate by 80% in low light environment.

[0019] 3. Enables flexible customization and intelligent alarms: Supports custom overlay of meteorological elements (data fusion layer) and threshold alarm linkage to meet the needs of different seasons and different business scenarios (such as focusing on monitoring wind speed and visibility during disaster warnings). The alarm response time is shortened to within 1 minute, which is a significant improvement over the lag of manual inspection. Attached Figure Description

[0020] Figure 1 This is a schematic diagram of the technical roadmap for inventing a method for real-time monitoring of meteorological stations and overlaying minute-level meteorological data.

[0021] Figure 2 This is a flowchart of a method for real-time monitoring of meteorological stations and overlay display of minute-level meteorological data. Detailed Implementation

[0022] This invention discloses a method and system for real-time monitoring of meteorological stations and overlay display of minute-level meteorological data. The following describes the implementation process of this solution in detail with actual deployment cases of multiple pilot stations, supporting drawings and core technical details, to ensure that those skilled in the art can accurately reproduce this technology.

[0023] I. System Composition like Figure 1 The image shows a real-time monitoring and minute-level meteorological data overlay display system for meteorological stations, as described in this invention. The system consists of three main functional modules: data acquisition, data transmission, and data fusion and display. Depending on the station's scale and operational needs, optional modules can be added, including data storage, aggregation switches, multi-terminal access, access control, and alarm linkage. Each component selection is tailored to the on-site environment and operational characteristics of meteorological monitoring. The specific configuration is as follows: 1. Data Acquisition Module The data acquisition module comprises two core devices: a video camera and a meteorological data acquisition unit. The video camera is a Hikvision DS-2DY54QAZ-HJ high-definition device, supporting 4K resolution output of 3840×2160 pixels. Equipped with a starlight-level sensor and supplementary lighting optimization technology, it achieves a minimum illumination of 0.001 Lux in 24-hour full-color night vision mode, clearly capturing cloud formations, visibility, and the operational status of station equipment under complex weather conditions such as nighttime, fog, and sandstorms. The camera is compatible with UDP protocol and GB / T28181 national standard signaling. UDP protocol is specifically used for low-latency video stream transmission, while GB / T28181 signaling enables device control command interaction. The default frame rate is set to 25fps to ensure smooth dynamic images. It connects to the station's local area network via a gigabit Ethernet port and supports PoE power supply, allowing for simultaneous power supply and data transmission via network cable, significantly simplifying on-site wiring.

[0024] The meteorological data acquisition unit adopts an embedded data acquisition terminal with a built-in industrial-grade processor. It establishes stable communication with the meteorological big data cloud platform via an Ethernet interface and supports HTTP / HTTPS protocol calls to open API interfaces. The terminal integrates an NTP clock synchronization module, with time synchronization accuracy controlled within 1ms, ensuring strict alignment between the acquired meteorological data and the video stream timestamps.

[0025] 2. Data transmission module The transmission employs a dual-protocol collaborative mechanism of "UDP as the primary protocol and TCP as the secondary protocol." UDP is dedicated to carrying 4K video stream transmission, leveraging its connectionless and low-latency characteristics to ensure real-time video delivery. TCP handles signaling verification and data retransmission; when data packets are lost during UDP transmission, a retransmission request is initiated via TCP to ensure data transmission integrity. Regarding the transmission medium, the backbone link uses single-mode fiber optic cable, and the internal cabling of the station uses Category 6e twisted-pair cable, fully adapting to the transmission distance and environmental requirements of the meteorological station, ensuring stable and reliable transmission of video streams and meteorological data.

[0026] For the transmission medium, single-mode fiber (core diameter 9 / 125μm, transmission bandwidth ≥10Gbps) is used for the backbone transmission, and Category 6A twisted-pair cable (supporting Gigabit Ethernet transmission, attenuation ≤0.5dB / 100m) is used inside the station to ensure the stability of video stream and data transmission.

[0027] 3. Data Fusion and Display Module The provincial-level video surveillance management system is built on rack-mount servers, equipped with high-performance processors, large-capacity memory, and high-speed storage disks. It features a self-developed video decoding and data fusion engine capable of simultaneously decoding over 32 channels of 4K video streams, compatible with both H.264 and H.265 encoding formats, and efficiently completing real-time fusion processing of video streams and meteorological data. The system runs on a Linux operating system and employs a distributed architecture design, allowing for flexible cluster expansion to meet the concurrent access needs of multiple stations across the province, adapting to different scales of business scenarios.

[0028] 4. Data storage module For local storage, the Hikvision DS-7604N-E1-V3 hard disk recorder can be selected. It supports multiple video decoding formats including H.264, H.265, Smart264, and Smart265, with a total storage capacity of at least 24TB. Based on the storage requirements of 4K video streaming (25fps, 16Mbps bitrate), a single recorder can continuously store more than 30 days of video data. The recorder supports RAID array configuration, effectively mitigating the risk of data loss due to hard drive failure. It connects to the station's LAN via a gigabit Ethernet port, enabling local video preview, playback, and export functions to meet the station's local data retrieval needs.

[0029] The provincial centralized storage adopts a distributed storage system. The storage capacity is planned and configured in a coordinated manner according to the number of stations in the province. It supports the associated storage of video data and meteorological data. The data retention period can be flexibly adjusted through system configuration, with a default setting of 90 days to meet the needs of meteorological data archiving and review.

[0030] 5. Aggregation Switch The Hikvision DS-3E2552-H(B) Gigabit Ethernet switch was selected, equipped with 48 Gigabit Ethernet ports (supporting PoE+ power supply) and 4 Gigabit fiber ports, with a packet forwarding rate of no less than 132Mpps, which can meet the video signal aggregation requirements of up to 8 cameras in a single station. The switch supports VLAN segmentation, which can distribute video streams, control signaling, and meteorological data to different VLANs, effectively avoiding data transmission conflicts and ensuring the stability of the transmission link.

[0031] 6. Multi-terminal access unit The B / S web server adopts a reverse proxy + application server cluster architecture, supporting over 1000 concurrent user accesses. The web interface uses mainstream front-end development technologies, supports adaptive resolution, and runs smoothly on PCs and tablets, adapting to different work scenarios for business personnel. The C / S client server is developed based on mainstream frameworks for Windows clients, compatible with Windows 10 / 11 operating systems. It supports video hardware decoding, utilizes GPU acceleration to reduce CPU usage, and provides core functions such as real-time preview, video playback, and alarm handling, meeting the needs of professional business operations. The mobile APP server adopts a mainstream back-end architecture, compatible with Android 8.0 and above, and iOS 12.0 and above systems. It achieves real-time video stream push via 4G / 5G networks, ensuring timely delivery of alarm information to personnel on the field or on duty.

[0032] 7. Access Control Module and Alarm Linkage Module The access control server is equipped with a mainstream database to store user information, role permissions, and station resource association data. It implements a four-level permission system based on the RBAC (Role-Based Access Control) model, adapting to the hierarchical management needs of meteorological operations. The alarm linkage module includes local audible and visual alarm devices and a message push terminal. The audible and visual alarm has an alarm volume of no less than 110dB and a flashing frequency of 1Hz. It connects to the provincial video management system via a standard interface to achieve real-time response to alarm signals, ensuring timely detection of abnormal situations.

[0033] II. System Network Topology and Network Security Configuration (I) Network Topology Design The local network of the station adopts a hierarchical access method. The video cameras are connected to the gigabit port of the aggregation switch through Cat6e twisted pair cables. The local video recorder is also connected to the aggregation switch. All devices are in the same VLAN segment to ensure local data transmission efficiency. The aggregation switch is connected to the fiber optic transceiver through the gigabit optical port, and then connected to the provincial backbone network through the operator's fiber optic link to realize remote communication between the station and the provincial system.

[0034] At the provincial network level, the provincial video management system receives video streams transmitted from various stations via fiber optic transceivers and forms a local area network with B / S web servers, C / S client servers, and mobile APP servers to ensure rapid internal data interaction. The provincial video management system connects to the meteorological big data cloud platform via a dedicated line, and the meteorological comprehensive business real-time monitoring platform establishes two-way communication with the provincial video management system to obtain equipment operating status and event log information in real time, realizing full-link monitoring.

[0035] (ii) Network security configuration Network security protection is built from three aspects: firewall rules, data encryption, and access control. The station's egress firewall only opens the ports required for video streaming, API calls, and signaling interaction, strictly prohibiting external access to unrelated ports to reduce the risk of network attacks; video streaming uses high-strength encryption algorithms, API calls use the HTTPS protocol, and key settings are set with a regular automatic update mechanism to ensure data transmission security; the aggregation switch enables port security functions, binding only the MAC addresses of authorized devices to effectively prevent unauthorized devices from accessing the network and ensure the security of the network environment.

[0036] III. Specific Implementation Plan The core implementation process of this invention includes three main steps: data acquisition, data transmission, and data fusion and display. Extended steps such as data storage, multi-channel signal aggregation, multi-terminal access, access control, and alarm linkage can be selected according to actual application needs. The specific operation process is as follows: (a) Data collection steps 1. Video stream and control signaling acquisition After starting the video camera, log in to the camera management interface through the station's local area network and complete the basic parameter configuration: set the resolution to 4K, frame rate to 25fps, bitrate to 16Mbps (variable bitrate mode), and enable the "full-color night vision" function; enable the UDP protocol for video stream transmission, and enable GB / T28181 signaling, configure the signaling server address and corresponding port to ensure normal interaction of device control commands.

[0037] After the camera is activated, it continuously acquires real-time video streams from the station and stably pushes video data to the provincial video management system via UDP protocol, with transmission latency controlled within 300ms. When users issue control commands such as camera rotation and focus adjustment through the client, the commands are transmitted to the camera after being encapsulated using GB / T28181 signaling standards, with a camera response time of no more than 100ms, ensuring real-time operation. 2. Meteorological element data collection First, configure the network parameters of the meteorological data acquisition unit to ensure that it is on the same communication network segment as the meteorological big data cloud platform and that network connectivity is guaranteed. In the configuration interface of the data acquisition terminal, accurately enter the platform API interface address and interface key, select the meteorological elements to be collected (from temperature, air pressure, precipitation, humidity, wind speed, snow depth, and visibility), and set the data request frequency to once per minute to match the real-time requirements of the video stream.

[0038] The data acquisition terminal initiates API requests at a preset frequency. After receiving the meteorological data in JSON format returned by the platform, it parses the data, extracts key element values ​​and timestamps, and calibrates the local time through the NTP protocol to ensure that the error between the meteorological data timestamp and the video stream timestamp does not exceed 1ms, laying the foundation for subsequent data fusion.

[0039] (II) Data Transmission Steps 1. Single-channel video stream transmission 2. The 4K video stream captured by the video cameras is encapsulated using the UDP protocol and then transmitted sequentially through the station's local area network and fiber optic transmission link to the provincial video management system. The provincial video management system receives the video data packets in real time and verifies their integrity. If a data packet loss is detected, a retransmission request is immediately sent to the camera via the TCP protocol. Upon receiving the request, the camera quickly retransmits the video data for the corresponding time period, ensuring the continuity of the video stream.

[0040] 3. Multi-channel video stream convergence and transmission When a video surveillance station deploys two or more video cameras, the video streams from each camera are connected to the gigabit Ethernet port of the aggregation switch. The switch isolates and forwards the data according to the preset VLAN configuration, efficiently aggregating multiple video streams and then transmitting them to the provincial video surveillance management system via the gigabit optical port through the fiber optic transmission link. The switch supports flow control, which can effectively avoid port congestion during high-concurrency transmission and ensure stable transmission of multiple video streams.

[0041] (III) Data Fusion and Display Steps 1. Data fusion processing The video decoding engine of the provincial video management system first decodes the received UDP video stream and converts it into frame data. At the same time, the data parsing module extracts fields and converts the format of the meteorological data in JSON format, unifying the data units (such as temperature in "℃" and wind speed in "m / s") to ensure that the data format is standardized.

[0042] A dedicated data fusion algorithm is activated to accurately correlate video frame data and meteorological data from the same moment using a timestamp matching mechanism. Based on pixel coordinate mapping rules, semi-transparent data labels are overlaid in designated areas of the video frames, with the label background transparency set to 60% to ensure clear and legible text without obscuring key information in the video. The data label display format is "Element Name: Value + Unit", such as "Temperature: 25.6℃" and "Precipitation: 8.2mm", presenting the meteorological data intuitively.

[0043] The system supports user-defined configurations. Users can flexibly select the meteorological elements to be displayed through the "Data Overlay Configuration" interface on either a B / S web page or a C / S client. They can easily adjust the overlay position of data labels by dragging and dropping with the mouse. Configuration takes effect immediately after completion, and the system automatically updates data fusion rules to adapt to different business scenarios. 2. Display Output Multi-terminal synchronous display: The provincial video management system converts the merged video streams with data tags into RTMP format and distributes them to various terminals through B / S web servers, C / S client servers, and mobile APP servers. The PC webpage supports simultaneous preview of 16 video streams (split-screen display), and the mobile APP supports split-screen preview of 4 video streams. The video display latency is ≤500ms, which is ≥60% lower than that of traditional HTTP protocol transmission.

[0044] Alarm indicator overlay: The system presets alarm thresholds for various meteorological elements (which can be modified through the administrator interface). Default thresholds include wind speed ≥10m / s, visibility ≤1km, and hourly precipitation ≥20mm. When meteorological data exceeds the threshold, the data fusion module automatically overlays a red flashing alarm indicator (size: 80×80 pixels, flashing frequency: 2Hz) in the upper right corner of the video screen. The indicator content is "[Alarm Type]: [Element Name]_[Value]", for example, "[Wind Speed ​​Alarm]: Wind Speed_12.3m / s".

[0045] (iv) Data storage steps For local storage, the local recorder receives video streams pushed by cameras in real time, compresses and stores them in H.265 encoding format, and plans the storage path according to the "station ID / date / time period" standard. It supports quick retrieval of local recordings by time and camera number, and can export recording files through a standard interface, making it convenient for local data retention and viewing at the station.

[0046] At the centralized storage level, the provincial video management system associates the merged video stream with meteorological data and stores it in a distributed storage system. It adopts a storage structure of "video stream + metadata". The metadata includes meteorological element values, timestamps, alarm information, etc. It supports accurate retrieval and related queries by time range, station ID, and event type (such as alarm events and normal monitoring) to meet the data management needs of the entire province.

[0047] (V) Access Control Steps Account creation and permission allocation are handled by the system administrator through the "User Management" module on the B / S web interface. Access permissions are assigned according to a four-level permission model: The system administrator has full operation permissions, including creating / deleting accounts, modifying system parameters, and accessing all station resources; provincial accounts can access real-time video, historical recordings, and meteorological data from all stations in the province, and support data and recording export, but have no system configuration modification permissions; municipal / prefectural bureau accounts can only access resources of stations under their own jurisdiction, and support real-time preview, recording playback, and local export, but have no cross-municipal / prefectural access permissions; county-level station accounts are limited to accessing the video and data of their own station, supporting real-time monitoring and local recording playback, but have no export or cross-station access permissions, strictly adhering to the hierarchical management principle.

[0048] When users log in to various terminals, the system verifies their identity using "username + password + verification code" (mobile terminals support SMS verification code). After successful verification, the system loads the corresponding station resource list according to the account permissions, preventing unauthorized access and ensuring data security.

[0049] (vi) Alarm linkage steps When meteorological data exceeds the preset threshold, the alarm linkage module of the provincial video management system immediately triggers a triple alarm: a control signal is sent to the audible and visual alarm at the station, activating the alarm (which lasts for 5 minutes by default and can be manually stopped via the client), quickly alerting the station staff; an alarm message is sent to the relevant users' C / S clients and mobile APPs, including the alarm station, alarm time, alarm type, current value, and real-time video link, facilitating remote awareness of the situation; and the system automatically records alarm events (including alarm trigger time, recovery time, element value, and handler), storing them in the alarm log database for at least one year for subsequent traceability and review.

[0050] When the meteorological data recovers to the threshold range, the system automatically stops the audible and visual alarms and message pushes, removes the red alarm marker from the video screen, and updates the alarm log to the "recovered" status, ensuring a closed loop in the alarm process.

[0051] (V) Steps for Historical Data Retrieval and Analysis Users can use the "Event Search" function on the B / S web interface to input search criteria such as time range, event type, and station ID. The system quickly filters corresponding video files and meteorological data from the distributed storage system and generates a search results list. Users can preview the video (supporting playback speeds of 0.5x-4x) and view associated meteorological data curves for intuitive analysis of weather processes. Search results can be downloaded (videos in MP4 format and data in Excel format) and archived with tags. Users can manually label event types (such as "heavy rain event" or "strong wind event"), and the label information is stored in association with the video and data for easy subsequent classification, management, and retrieval.

[0052] IV. Implementation Plans for Each Section The key technologies required are as follows: (I) Achieving Low-Latency Transmission of 4K Full-Color Video This invention ensures low-latency transmission of 4K video streams through a dual approach of "hardware selection + protocol optimization": 1. Select a camera and decoder that support H.265 encoding. Compared to H.264, H.265 encoding can save 50% of bandwidth and reduce transmission latency by 30% at the same bitrate. 2. The UDP protocol transmission adopts "simplified packet header + multicast optimization", which removes redundant transmission control fields and distributes video streams to multiple terminals through multicast to avoid bandwidth consumption caused by repeated transmission; 3. The transmission link uses optical fiber media to reduce signal attenuation and interference. Combined with NTP time synchronization, it ensures that the end-to-end delay of the video stream from acquisition to display is ≤500ms.

[0053] (II) Core Logic of Data Fusion Algorithm The core process of the dedicated data fusion algorithm is as follows: 1. Timestamp Alignment: A linear interpolation method is used. When the timestamps of meteorological data (1 per minute) and video stream timestamps (1 per 40ms) do not match perfectly, the meteorological element interpolation at the corresponding video frame time is calculated based on the values ​​of the two consecutive meteorological data points to ensure accurate synchronization between data and video. 2. Pixel coordinate mapping: Establish a mapping relationship between the pixel coordinates of video frames and the screen display coordinates. The user-defined label position (screen coordinates) is converted into the pixel coordinates of video frames through a mapping formula to ensure the consistency of the label position on display devices with different resolutions. 3. Tag overlay optimization: Using layer overlay technology, data tags are overlaid as independent layers on top of video frames. The transparency is adjusted to avoid obscuring key video information. At the same time, automatic tag avoidance is supported (when a key area in the video frame, such as the device, conflicts with the tag position, the tag position is automatically adjusted).

[0054] (III) Alarm Linkage Response Mechanism The alarm linkage module adopts a "tiered response + rapid triggering" design approach. In the data exceedance detection phase, the provincial video management system performs threshold judgment on the latest received meteorological data every 100ms. Once an exceedance is detected, the alarm process is immediately triggered, with an alarm response time of no more than 1 minute, ensuring rapid response to abnormal situations. For multi-dimensional alarm collaboration, local audible and visual alarms are triggered first (response time no more than 2 seconds), while simultaneously pushing messages to the user client. The client notifies the user via pop-up windows and ringtones, ensuring no alarm information is missed and adapting to different duty scenarios. In the alarm recovery judgment phase, the system continuously monitors meteorological data. When the data is within the threshold range for three consecutive times (each time with a 10-second interval), the alarm automatically stops, ensuring alarm stability and avoiding false alarms caused by instantaneous fluctuations that could interfere with business operations.

[0055] V. Troubleshooting and System Maintenance (I) Common Exception Handling Procedures In practical applications, you may encounter abnormal situations such as video stream not displaying, meteorological data not being acquired, data and video being out of sync, and alarms not being triggered. The specific handling methods are as follows: If the video stream cannot be displayed: First, log in to the camera management interface to check the device's operating status, verify the camera's network connectivity using a network testing tool, and check the connection status of the corresponding port on the aggregation switch; the corresponding solutions are to remotely restart the camera (or power off and restart it on-site), check the network cable connection and replace the faulty network cable, and restart the aggregation switch to quickly restore video transmission.

[0056] Unable to acquire meteorological data: Check system logs to confirm if there are API interface authentication failures or network connection timeout errors. Verify the network connectivity between the provincial video management system and the meteorological big data cloud platform using the ping command. The corresponding solutions are to contact the platform administrator to reset the interface key and update the system configuration, check the network link status and contact the operator to repair the fault, and restore data acquisition.

[0057] Data and video are out of sync: Check the NTP synchronization status of the meteorological data acquisition unit and compare the difference between the video frame timestamp and the meteorological data timestamp; the corresponding solution is to reconfigure the NTP server address and calibrate the time, check the running status of the data fusion algorithm, and restart the data fusion service if necessary to restore the synchronization effect.

[0058] Alarms not triggered: Check whether the threshold configuration of meteorological elements in the system is reasonable (e.g., whether it meets the business early warning standards), and check the running status of the alarm service through the service management tool; the corresponding solution is to modify the threshold configuration and restart the alarm service, and start the alarm service that is not running to ensure that the alarm function is working properly.

[0059] (II) System Maintenance Cycle and Operation System maintenance is divided into three categories: daily maintenance, periodic maintenance, and quarterly maintenance, to ensure the long-term stable operation of the system. Daily maintenance requires checking the system operation log to confirm that there are no error messages, checking the synchronization status of video stream transmission and meteorological data acquisition, and ensuring that core functions are normal. Periodic maintenance requires backing up system configuration files and alarm logs, and timely updating system security patches to improve system security. Quarterly maintenance requires checking the remaining storage capacity of devices such as local recorders and distributed storage, cleaning up expired recording data that is more than 90 days old according to rules to avoid storage overflow, and calibrating the camera time and NTP synchronization status to ensure data timeliness and accuracy.

[0060] VI. Verification of Implementation Results This invention has completed a three-month practical operation verification at multiple pilot meteorological stations, and all technical indicators have met the design requirements: the measured end-to-end latency of video transmission is 320-480ms, with an average latency of 400ms, which is significantly reduced compared to the traditional HTTP protocol transmission latency, meeting the needs of real-time monitoring; in nighttime (illuminance 0.01Lux) environments, 4K full-color video can clearly identify the operating status of equipment at 50m, the visibility recognition range is no less than 2km, and the detail recognition rate is 80% higher than that of traditional 1080P black and white night vision, solving the shortcomings of monitoring in low-light environments; the timestamp error between meteorological data and video stream is no more than 1ms, and the data tags are completely synchronized with the image without significant delay or offset, ensuring the linkage effect between data and real scene; after alarm triggering, the local audible and visual alarm response time is no more than 2s, and the client message push response time is no more than 5s, which is significantly improved compared to the lag of manual inspection and improves the efficiency of anomaly handling; during 90 days of continuous operation, there were no equipment failures, data loss, or transmission interruptions, and the system availability is no less than 99.9%, meeting the 7×24-hour operation requirements of meteorological services.

[0061] (I) Comparison of key technical indicators Compared with traditional station systems, this invention achieves significant improvements in core technical indicators: video transmission latency is reduced from the traditional 1200-1500ms (average 1350ms) to 320-480ms (average 400ms), a reduction of 66.7%; in low-light conditions at night, the detail recognition distance is increased from 20m to 50m, and the visibility recognition range is expanded from ≤1km to ≥2km, representing improvements of 150% and 100% respectively; the data-video synchronization error is reduced from ≥500ms to ≤1ms, a reduction of 99.8%; the alarm response time is shortened from ≥1 hour (manual inspection) to ≤5s (client push), a 720-fold increase in response efficiency; and system availability is improved from ≤95% to ≥99.9%, an increase of 4.9 percentage points, comprehensively optimizing meteorological monitoring efficiency.

[0062] (II) Typical Application Scenarios and Effects In heavy rain monitoring scenarios, traditional technologies require switching between video systems to observe rainfall patterns and data systems to check precipitation amounts, taking at least 5 minutes to determine precipitation intensity. This invention, by directly overlaying minute-by-minute precipitation data onto video footage, can determine precipitation intensity within 10 seconds, increasing efficiency by 30 times and saving valuable time for heavy rain warnings. In equipment fault diagnosis scenarios, traditional technologies require on-site inspections after data anomalies are detected, taking at least 2 hours. This invention, through "abnormal data + video footage," can directly confirm equipment status (such as sensor obstruction or equipment displacement), allowing remote diagnosis in less than 10 minutes, increasing efficiency by 12 times and reducing maintenance costs.

[0063] Through the above-described embodiments, this invention successfully solves the technical problems of traditional meteorological stations, such as "separation of data and real-world scene," "poor video acquisition and transmission quality," and "insufficient scene adaptability." It achieves real-time overlay display of real-world monitoring and minute-level meteorological data, flexible configuration, and intelligent alarms, providing strong technical support for meteorological monitoring and early warning, disaster reconnaissance, and public services. Those skilled in the art can make appropriate adjustments to component selection and parameter configuration according to actual application scenarios, all of which fall within the scope of protection of this invention.

[0064] The embodiments described above are merely preferred embodiments of the invention and are not intended to limit the scope of the invention. Any modifications and improvements made to the technical solutions of the invention by those skilled in the art without departing from the spirit of the invention should fall within the protection scope defined by the claims.

Claims

1. A weather station real scene monitoring and minute-level weather data superimposed display method, characterized in that, Comprise: S1 data acquisition: Collect real scene video stream through video camera deployed by station, and transmit the real scene video stream by using UDP protocol; Real-time read at least one meteorological element data of the station through the open API interface of the meteorological big data cloud platform, the meteorological element data includes one or more of temperature, air pressure, precipitation, humidity, wind speed, snow depth, and visibility, the data update frequency is minute level and matches the frame rate of the real scene video stream to realize timestamp alignment; S2 data transmission: Transmit the real scene video stream to the provincial video management system by using the mode of "UDP protocol as the main transmission channel and TCP protocol as the auxiliary signaling verification"; S3. Data fusion and display: Align the meteorological element data obtained by the open API interface with the parsed video stream through a special data fusion algorithm, and map the meteorological element data label to any area of the video picture through pixel coordinates; Pre-set alarm threshold of each meteorological element, when the meteorological element data exceeds the alarm threshold, the system automatically adds alarm mark on the video picture.

2. The weather station real scene monitoring and minute-level weather data superimposed display method according to claim 1, characterized in that, Also includes data storage step: real-time store the real scene video stream in the local video recorder of the station, or realize centralized storage through the provincial video management system, or simultaneously realize local real-time storage and provincial centralized storage.

3. The weather station real scene monitoring and minute-level weather data superimposed display method according to claim 1, characterized in that, When multiple video cameras are deployed in the station, after converging multiple video signals through the convergence switch, the mode of "UDP protocol as the main transmission channel and TCP protocol as the auxiliary signaling verification" is used to transmit to the provincial video management system; the convergence switch has gigabit Ethernet interface, including at least 48 gigabit electrical ports and 4 gigabit optical ports, supports PoE+ power supply, and the packet forwarding rate is ≥78Mpps / 132Mpps.

4. The weather station real scene monitoring and minute-level weather data superimposed display method according to claim 1, characterized in that, The video camera supports 24h full-color display, the resolution is not less than 4K, and is compatible with GB / T28181 national standard signaling to collect device control signaling; the video camera is connected to the station local area network through gigabit network port, supports POE power supply, and the frame rate is set to 25fps.

5. The weather station real scene monitoring and minute-level weather data superimposed display method according to claim 2, characterized in that, The local video recorder supports H.265 encoding format, and is configured with storage capacity ≥2TB, and is compatible with H.264, Smart265, Smart264 video decoding format; the meteorological big data cloud platform is "Tiancheng" of Qinghai Province meteorological big data cloud platform, the meteorological element data request frequency of the open API interface is 1 time per minute, and the return data format is JSON; the parsing format of the video stream is RTSP.

6. The weather station real scene monitoring and minute-level weather data superimposed display method according to claim 5, characterized in that, Support user to customize the superposition position of meteorological element data label and the type of meteorological element to be displayed, the meteorological element data label is semi-transparent display; when the meteorological element data exceeds the threshold, the sound and light alarm of the local device is triggered synchronously and the message push to the related user client is triggered, the alarm response time is ≤1 minute, and the alarm threshold includes wind speed ≥10m / s and visibility ≤1km.

7. A weather station real scene monitoring and minute-level weather data superimposed display system, characterized in that, Comprise data acquisition module, data transmission module, data fusion and display module; The data acquisition module includes a video camera and a meteorological data acquisition unit, the video camera is used to collect the real scene video stream of the station and transmit by using the UDP protocol, the meteorological data acquisition unit is used to read at least one kind of meteorological element data of the station in real time through the open API interface of the meteorological big data cloud platform, the meteorological element data includes one or more of temperature, air pressure, precipitation, humidity, wind speed, snow depth and visibility, the data update frequency is minute level and matches the frame rate of the real scene video stream to realize timestamp alignment; the data transmission module transmits the real scene video stream to the provincial video management system in the mode of "UDP protocol as the main mode, TCP protocol as the auxiliary signaling verification"; the data fusion and display module is built-in with a special data fusion algorithm, which is used to realize the timestamp alignment of the meteorological element data and the video stream and the data label superposition, and a preset alarm threshold, when the data is out of limit, the red alarm mark is automatically superimposed on the video picture.

8. The system of claim 7, wherein, It also includes a data storage module, the data storage module includes a local video recorder and / or a storage unit of the provincial video management system, the local video recorder is used to store the real scene video stream in real time, and the storage unit of the provincial video management system is used to realize centralized storage; the video camera is a Hikvision DS-2DY54QAZ-HJ type, and the local video recorder is a Hikvision DS-7604N-E1-V3 type.

9. The system of claim 7, wherein, It also includes a convergence switch, a multi-terminal access unit, a permission management module and an alarm linkage module; the convergence switch is used to converge multiple video signals when multiple video cameras are deployed at the station, the convergence switch is a Hikvision DS-3E2552-H(B) type, and the video signals are transmitted to the provincial video management system through optical fiber or twisted pair; the multi-terminal access unit includes a B / S web server, a C / S client server and a mobile terminal APP server, which are connected with the provincial video management system to realize multi-terminal access; the permission management module divides four levels of permissions based on a role-based access control model, which are system administrator permission, provincial account permission, city and prefecture bureau account permission and county station level account permission; The alarm linkage module is used to trigger local audible and light alarms and client message push when the meteorological element data exceeds the threshold, the alarm response time is less than or equal to 1 minute, and the alarm threshold includes wind speed greater than or equal to 10 m / s and visibility less than or equal to 1 km; the provincial video management system and the meteorological comprehensive business real-time monitoring platform realize two-way communication, and the meteorological comprehensive business real-time monitoring platform is used to monitor the device state and event log of the whole link.

10. The system of claim 9, wherein, The B / S web server supports historical data retrieval and analysis functions, users can filter videos according to time range and event type, the event type includes device alarm and wind speed exceeding the standard, and the retrieval result supports downloading and label archiving.

Images