Air visibility intelligent monitoring system and method based on three-dimensional space visualization

Abstract

The invention relates to an air visibility intelligent monitoring system and method based on three-dimensional space visualization. The system comprises a data sensing module arranged in a target monitoring area; the data fusion processing module receives the multi-source environment sensing signal and generates a data signal through a spatial interpolation algorithm; the three-dimensional visualization engine module receives the three-dimensional visibility field data signal and generates a visualization scene signal based on a volume rendering technology; the intelligent analysis early warning module receives the three-dimensional visibility field data signal and the dynamic three-dimensional visibility visual scene signal, and generates an early warning information signal through an integrated space analysis and prediction algorithm; and the comprehensive management application module receives the visibility time-space evolution trend and early warning information signal and the dynamic three-dimensional visibility visual scene signal. The air visibility intelligent monitoring system and method based on three-dimensional space visualization can solve the problems that a traditional visibility monitoring method is insufficient in spatial representativeness, lack of three-dimensional visual presentation and difficult to achieve accurate early warning.

Description

Technical Field

[0001] This invention relates to the interdisciplinary field of environmental monitoring and computer visualization, specifically to an intelligent air visibility monitoring system and method based on three-dimensional spatial visualization. Background Technology

[0002] Traditional air visibility monitoring primarily relies on two types of equipment: transmission visibility meters and scattering visibility meters. Transmission visibility meters calculate visibility by measuring light attenuation along a fixed optical path. While offering high accuracy, they require establishing a measurement baseline tens of meters long, resulting in drawbacks such as large installation footprint, high cost, and inconvenient maintenance. Furthermore, they can only provide data from a single measurement point. Scattering visibility meters infer visibility values ​​by detecting the intensity of light scattering by particulate matter. Although smaller and easier to install, their measurement results are susceptible to interference from factors such as optical window contamination and light source attenuation, requiring frequent calibration. Similarly, they only reflect the visibility conditions at a localized point.

[0003] Both methods have fundamental limitations: they cannot capture the distribution of visibility in three-dimensional space, and they struggle to observe the formation, movement, and dissipation of weather phenomena such as fog and haze. In recent years, while some studies have attempted to estimate visibility from surveillance videos using image analysis or utilize lidar for vertical detection, these technologies are mostly applied in isolation, lacking effective spatial integration methods. Existing visibility monitoring systems typically display monitoring point data as two-dimensional map dots, making it difficult for decision-makers to intuitively grasp the overall visibility situation in a region, let alone conduct accurate spatial visibility analysis and intelligent prediction of visibility evolution trends. This situation severely limits the effectiveness of visibility monitoring in important fields requiring three-dimensional spatial perception, such as aviation, navigation, and transportation. Summary of the Invention

[0004] In view of the shortcomings of the existing technology, the purpose of this invention is to provide an intelligent air visibility monitoring system and method based on three-dimensional spatial visualization, which solves the problems of insufficient spatial representativeness, lack of three-dimensional intuitive presentation, and difficulty in achieving accurate early warning in traditional visibility monitoring methods. This invention collects ground and air visibility and meteorological data by deploying a multi-source sensor network, uses a spatial interpolation algorithm to fuse discrete point data to generate a continuous three-dimensional visibility field, and utilizes volume rendering technology to transform the data into a three-dimensional fog effect visualization model with transparency gradients. Combined with visibility analysis, source tracing and inference, and machine learning prediction algorithms, it dynamically presents the distribution and evolution trend of visibility in three-dimensional space, and provides spatial analysis and early warning functions through an integrated interactive interface, achieving a technological leap from point measurement to volume perception, and from digital display to three-dimensional visualization.

[0005] This invention provides an intelligent air visibility monitoring system based on three-dimensional spatial visualization, comprising: The data sensing module is deployed in the target monitoring area to collect multi-source environmental sensing signals, including visibility factors and meteorological factors. The data fusion processing module receives multi-source environmental sensing signals and generates three-dimensional visibility field data signals describing the monitoring area through a spatial interpolation algorithm. The 3D visualization engine module receives 3D visibility field data signals and generates dynamic 3D visibility visualization scene signals that are synchronized with the physical world state based on volume rendering technology. The intelligent analysis and early warning module receives three-dimensional visibility field data signals and dynamic three-dimensional visibility visualization scene signals, and generates visibility spatiotemporal evolution trend and early warning information signals by integrating spatial analysis and prediction algorithms. The integrated management application module receives visibility spatiotemporal evolution trend and early warning information signals and dynamic three-dimensional visibility visualization scene signals, and generates a comprehensive monitoring signal that integrates three-dimensional visualization status, analysis results and early warning instructions.

[0006] In one embodiment of the present invention, the visibility factor collected by the data sensing module originates from a forward scattering visibility meter or a transmission visibility meter. The collected meteorological factors include at least wind speed, wind direction, ambient temperature and relative humidity. The module also integrates an image sensing unit for collecting video image stream signals of the monitored area. All of the above signals together constitute a multi-source environmental sensing signal and are transmitted to the data fusion processing module.

[0007] In one embodiment of the present invention, the spatial interpolation algorithm used by the data fusion processing module is Kriging interpolation or inverse distance weighted interpolation. The module uses this algorithm to fuse the spatially discrete multi-source environmental perception signals from the data perception module into a continuous, high-resolution three-dimensional mesh model. Each mesh node of the model contains an inferred visibility value, thereby forming a three-dimensional visibility field data signal.

[0008] In one embodiment of the present invention, the three-dimensional visualization engine module uses volume rendering technology to convert the three-dimensional visibility field data signal into a visually perceptible fog-like three-dimensional effect with different transparency and color. The area with lower visibility value presents a denser visual effect and the color tends to be a warning color, while the area with higher visibility value is more transparent. This generates a dynamic three-dimensional visibility visualization scene signal that realistically reflects the actual spatial visibility distribution.

[0009] In one embodiment of the present invention, the visibility analysis unit built into the intelligent analysis and early warning module can receive three-dimensional visibility field data signals and automatically calculate the visibility judgment result between the two points based on the user-defined observation point and target point positions. This result is generated and output together with the visibility spatiotemporal evolution trend and early warning information signal.

[0010] In one embodiment of the present invention, the source tracing analysis unit built into the intelligent analysis and early warning module can combine real-time wind direction and speed data to reverse-engineer the three-dimensional visibility field data signal, simulate and identify the possible movement path and source direction of the low visibility air mass, and integrate the source tracing analysis result into the generated visibility spatiotemporal evolution trend and early warning information signal.

[0011] In one embodiment of the present invention, the intelligent analysis and early warning module also integrates a time series prediction unit. This unit uses historical and current three-dimensional visibility field data signals and a prediction model trained by machine learning algorithms to predict the spatial distribution change trend of visibility in the monitoring area within a specific future period. The prediction result constitutes an important part of the spatiotemporal evolution trend of visibility and early warning information signal.

[0012] In one embodiment of the present invention, the interactive interface driven by the integrated management application module provides the functions of roaming and zooming in three-dimensional scenes, and supports users to perform cross-sectional analysis on dynamic three-dimensional visibility visualization scene signals by drawing arbitrary cross-sections to view the visibility status at any position inside. The interactive commands and results are integrated into the integrated monitoring signal.

[0013] In one embodiment of the present invention, after receiving the visibility spatiotemporal evolution trend and warning information signal containing warning information, the integrated management application module automatically generates and sends a standard format warning prompt signal to an external traffic management or public information release system. This warning prompt signal is a specific output form of the integrated monitoring signal.

[0014] This invention also includes a method for intelligent monitoring of air visibility based on three-dimensional spatial visualization, comprising: S1: Deployed in the target monitoring area to collect multi-source environmental sensing signals including visibility factors and meteorological factors; S2: Receives multi-source environmental sensing signals and generates a three-dimensional visibility field data signal to describe the three-dimensional spatial distribution of visibility within the monitoring area through a spatial interpolation algorithm; S3: Receives 3D visibility field data signals and generates dynamic 3D visibility visualization scene signals that are synchronized with the physical world state based on volume rendering technology; S4: Receives three-dimensional visibility field data signals and dynamic three-dimensional visibility visualization scene signals, and generates visibility spatiotemporal evolution trend and early warning information signals through integrated spatial analysis and prediction algorithms; S5: Receives visibility spatiotemporal evolution trend and early warning information signals and dynamic three-dimensional visibility visualization scene signals, generates a comprehensive monitoring signal that integrates three-dimensional visualization situation, analysis results and early warning instructions, and drives the interactive interface to display and publish it.

[0015] This invention provides an intelligent air visibility monitoring system and method based on three-dimensional spatial visualization. It collects ground and air visibility and meteorological data by deploying a multi-source sensor network, uses a spatial interpolation algorithm to fuse discrete point data to generate a continuous three-dimensional visibility field, and uses volume rendering technology to transform the data into a three-dimensional fog effect visualization model with transparency gradient. Combined with visibility analysis, source tracing and inference and machine learning prediction algorithms, it dynamically presents the distribution and evolution trend of visibility in three-dimensional space, and provides spatial analysis and early warning functions through an integrated interactive interface, realizing a technological leap from point measurement to volume perception, and from digital display to three-dimensional visualization. Attached Figure Description

[0016] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 This is a system architecture diagram of an intelligent air visibility monitoring system based on three-dimensional spatial visualization. Figure 2 This is a flowchart of a method for intelligent monitoring of air visibility based on three-dimensional spatial visualization. Detailed Implementation

[0018] The following specific examples illustrate the implementation of the present invention. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that, unless otherwise specified, the following embodiments and features described therein can be combined with each other.

[0019] It should be noted that the illustrations provided in the following embodiments are only schematic representations of the basic concept of the present invention. Therefore, the drawings only show the components related to the present invention and are not drawn according to the actual number, shape and size of the components in the actual implementation. In the actual implementation, the form, quantity and proportion of each component can be arbitrarily changed, and the layout of the components may also be more complex.

[0020] In the following description, numerous details are explored to provide a more thorough explanation of embodiments of the invention. However, it will be apparent to those skilled in the art that embodiments of the invention may be practiced without these specific details. In other embodiments, well-known structures and devices are shown in block diagram form rather than in detail to avoid obscuring embodiments of the invention.

[0021] Please see Figure 1-2 The present invention discloses an intelligent air visibility monitoring system based on three-dimensional spatial visualization, comprising: a data sensing module deployed in the target monitoring area for collecting multi-source environmental sensing signals including visibility factors and meteorological factors; a data fusion processing module receiving the multi-source environmental sensing signals and generating a three-dimensional visibility field data signal describing the monitoring area through a spatial interpolation algorithm; a three-dimensional visualization engine module receiving the three-dimensional visibility field data signal and generating a dynamic three-dimensional visibility visualization scene signal synchronized with the physical world state based on volume rendering technology; an intelligent analysis and early warning module receiving the three-dimensional visibility field data signal and the dynamic three-dimensional visibility visualization scene signal, and generating a visibility spatiotemporal evolution trend and early warning information signal through integrated spatial analysis and prediction algorithms; and a comprehensive management application module receiving the visibility spatiotemporal evolution trend and early warning information signal and the dynamic three-dimensional visibility visualization scene signal, and generating a comprehensive monitoring signal integrating three-dimensional visualization status, analysis results, and early warning instructions.

[0022] like Figure 1As shown, its core feature lies in constructing a complete technology chain from front-end data acquisition to back-end integrated application. The system mainly includes five mutually cooperating core modules, which are organically integrated into a whole through specific signal flows. The data sensing module, as the nerve ending of the system, is physically represented by a variety of sensor arrays widely deployed within the target monitoring area. These sensors are not of a single type, but rather a comprehensive system integrating modern sensing technologies. It includes at least a forward-scattering visibility meter for directly measuring atmospheric optical properties or a transmission visibility meter as a reference measurement device; wind speed and direction sensors, temperature sensors, and humidity sensors for capturing key meteorological parameters affecting visibility; and an innovative design of this module that integrates high-resolution network cameras as image sensing units. These cameras continuously acquire real-time video image streams of the monitoring area. All these sensor units work together to collect raw data including visibility factors and meteorological factors. After preliminary analog-to-digital conversion and signal conditioning, the data is packaged into a complete, multi-dimensional, multi-source environmental perception signal. This signal is continuously and stably transmitted to the next core component of the system, namely the data fusion processing module, through wired or wireless transmission networks.

[0023] Furthermore, the data fusion processing module acts as the system's cortex, responsible for processing and understanding the massive amounts of raw information from the perception layer. This module typically consists of a high-performance server cluster in hardware and runs complex spatial data interpolation algorithms and a data fusion engine in software. After receiving multi-source environmental perception signals from the data perception module, it first performs preprocessing such as cleaning, denoising, and standardization to eliminate noise caused by sensor errors or transient interference. Subsequently, the module's core algorithm is activated, employing advanced spatial interpolation algorithms, such as Kriging interpolation based on statistical principles and considering spatial autocorrelation, or inverse distance weighted interpolation based on the principle of geographical distance weighting. The brilliance of this algorithm lies in its ability to creatively and scientifically fuse and extrapolate multi-source environmental perception signals from the data perception module, which are spatially discrete and distributed, ultimately generating a continuous, high-resolution three-dimensional mesh model covering the entire target area. Each grid node in this model is no longer a simple measurement value, but a spatially representative visibility value inferred by an algorithm. This series of grid data, which includes spatial location information and visibility values, together constitutes a three-dimensional visibility field data signal that describes the three-dimensional spatial distribution of visibility within the monitoring area. This signal is the data foundation for all subsequent visualization and analysis work and is completely transmitted to the three-dimensional visualization engine module.

[0024] Specifically, the 3D visualization engine module is crucial for the system's intuitive perception, transforming the abstract data world into an image world that humans can intuitively understand. This module is typically built upon a powerful 3D graphics rendering library or game engine. After receiving the 3D visibility field data signal from the data fusion processing module, it doesn't perform simple surface rendering but instead invokes its internal volume rendering pipeline. This technology treats each data point (voxel) in the 3D data field as a tiny particle with specific optical properties (such as color, transparency, and emitted light intensity). Through a ray casting algorithm, it calculates the color and transparency integral of every ray emitted from the viewpoint that passes through the entire 3D data field. Specifically, in the mapping relationship, areas with extremely low visibility values ​​are assigned lower transparency values ​​and color values ​​tending towards red or gray, resulting in a dense, obstructive fog or haze-like 3D effect during rendering; while areas with high visibility values ​​are assigned extremely high transparency values, rendering them almost invisible. This precise mapping from data to visual effects generates a highly immersive, dynamically updated, and 3D visibility visualization that is in close synchronization with the real physical world. This provides decision-makers with unprecedented situational awareness.

[0025] like Figure 1As shown, the intelligent analysis and early warning module is the system's intelligent center, transcending simple "viewing" and entering the level of "understanding and prediction." This module connects to the data fusion processing module and the 3D visualization engine module, meaning it can directly manipulate raw, precise 3D visibility field data signals for quantitative calculations, and also perform spatial logic analysis by combining rendered scene signals. The module integrates a series of intelligent algorithms. For example, its visibility analysis unit can automatically calculate whether the visibility of all points along the line connecting the user-defined observation point (e.g., airport tower) and target point (e.g., runway end) meets the visibility conditions, providing accurate judgment results. Its source analysis unit can couple real-time wind direction and speed data to reverse-engineer the spatial morphological changes in low-visibility areas, simulating possible source paths. Its time series prediction unit can utilize machine learning models to predict future visibility evolution based on historical data and the current state. All these analysis, deduction, and prediction results are integrated and encapsulated, ultimately generating visibility spatiotemporal evolution trend and early warning information signals with strong decision support value. The integrated management application module serves as the central hub and portal for system-user interaction. It connects the intelligent analysis and early warning module with the 3D visualization engine module, receiving precise analysis results from the former and realistic scene signals from the latter. The core function of this module is to transform information into action. It overlays and merges analysis results with the visualized scene, generating a comprehensive monitoring signal that integrates 3D visualized situational awareness, multi-dimensional analysis results, and specific early warning commands. This signal drives a feature-rich interactive interface displayed on a large-screen command center or terminal computer. Users can not only roam and zoom the 3D scene but also trigger various analysis functions and make rapid command and dispatch decisions based on early warning information. Furthermore, this module is responsible for pushing critical early warning information to external systems through standard interfaces, thus forming a complete closed loop from perception, analysis, early warning to application.

[0026] Furthermore, the data fusion processing module is further defined and deepened. Its main feature lies in revealing the specific optimal type of spatial interpolation algorithm used in this module and its innovative achievements. The spatial interpolation algorithm used in this module is not a simple average or weighted average, but rather a Kriging interpolation method or inverse distance weighted interpolation method with a solid scientific foundation. Kriging interpolation, as an optimal unbiased estimation method, is powerful because it not only considers the distance relationship between the interpolation point and known sample points, but also deeply analyzes the spatial autocorrelation structure between known sample points through the variogram function. This allows it to maximize the utilization of the spatial statistical regularities implicit in the data, making the interpolation result not only an estimate but also providing the variance of the estimation error. Its interpolation result achieves the best balance between smoothness and accuracy. Inverse distance weighted interpolation follows the first law of geography, namely that things closer together are more similar. It assumes that the value of the interpolation point can be calculated by the weighted average of known sample points within a certain range around it, and the weight is inversely proportional to the distance from the sample point to the interpolation point; the closer the distance, the greater the weight and the greater the influence. This method is computationally efficient and intuitive. Only through either of these two algorithms can this module effectively process the multi-source environmental sensing signals from the data sensing module—significantly discrete and isolated in space, representing only the sensor's location—into a comprehensive manner. This involves fusing and inferring a fully continuous, high-spatial-resolution 3D mesh model covering every corner. This model is virtual yet highly realistic; each mesh node contains a reasonable visibility value inferred through complex algorithms. These values, when combined, accurately depict the continuous gradient of visibility in space, ultimately forming the crucial 3D visibility field data signal that forms the basis for all visualization and analysis.

[0027] like Figure 2 The diagram illustrates an intelligent air visibility monitoring method based on three-dimensional spatial visualization according to the present invention. S1: Deployed in the target monitoring area to collect multi-source environmental sensing signals containing visibility factors and meteorological factors; S2: Receive the multi-source environmental sensing signals and generate a three-dimensional visibility field data signal describing the three-dimensional spatial distribution of visibility within the monitoring area using a spatial interpolation algorithm; S3: Receive the three-dimensional visibility field data signal and generate a dynamic three-dimensional visibility visualization scene signal synchronized with the physical world state based on volume rendering technology; S4: Receive the three-dimensional visibility field data signal and the dynamic three-dimensional visibility visualization scene signal, and generate a visibility spatiotemporal evolution trend and early warning information signal through integrated spatial analysis and prediction algorithms; S5: Receive the visibility spatiotemporal evolution trend and early warning information signal and the dynamic three-dimensional visibility visualization scene signal, generate a comprehensive monitoring signal integrating three-dimensional visualization status, analysis results, and early warning instructions, and drive the interactive interface for display and publication.

[0028] Specifically, the detailed explanation of the 3D visualization engine module focuses on clarifying the core technical path and implementation method of how this module transforms dry data into intuitive visual effects. Through its internally implemented advanced volume rendering technology, this module performs a sophisticated transformation on the 3D visibility field data signal received from upstream. Its goal is no longer to generate traditional 3D models with only surfaces, but to generate a semi-transparent volume effect with internal structure and continuous changes. Volume rendering technology is a powerful technique that directly visualizes 3D volume data without the need for intermediate geometric primitive generation. Its basic unit is the voxel. The key step in this technology is assigning optical properties to each voxel, primarily color and opacity (transparency). This process is controlled by a carefully designed transfer function. For this invention, the design principle of the transfer function is: mapping visibility values ​​to transparency. The lower the visibility value, the lower the assigned transparency value (i.e., the less transparent), and the color value tends towards warning colors, such as red, orange, or grayish-black; conversely, the higher the visibility value, the higher the transparency value (the more transparent), and the color tends towards invisible or background colors. During rendering, a ray casting algorithm is employed. A ray is emitted from each pixel on the screen, passing through the 3D volumetric data field that has been endowed with optical properties. The ray is integrated along its path, and the contribution of all voxels along the way to the color of the ray, as well as the occlusion effect caused by opacity, are calculated to ultimately determine the final color value of that pixel. In this way, the entire 3D visibility field is rendered as a realistic, deep, and layered fog-like or hazy 3D effect. Users can observe the internal structure of this "transparent object" from any angle. Dense areas are clearly visible, while transparent areas are almost invisible. This generates a dynamic 3D visibility visualization scene signal that realistically reflects the actual spatial visibility distribution in the physical world and changes dynamically with real-time data updates, achieving a qualitative leap from digital to image.

[0029] Furthermore, the functional enhancement of the intelligent analysis and early warning module is highlighted by its core feature: the reveal of a specific analysis unit integrated within the module—the visibility analysis unit—and its unique working mechanism and application value. This visibility analysis unit is not an independent tool, but a key functional component deeply embedded within the algorithmic framework of the intelligent analysis and early warning module. It directly receives and processes 3D visibility field data signals with precise spatial coordinates and visibility values, generated and transmitted by the data fusion processing module. The unit's working principle is based on ray tracing in 3D computer graphics, but its application purpose is not for rendering; rather, it is for performing precise spatial accessibility calculations. When a user (e.g., an air traffic controller or port dispatcher) defines an observation point location (such as the top observation window of an airport control tower) and one or more target point locations (such as the runway entrance, taxiway intersection, or distant navigation points) on the system interface, these spatial coordinate requests are sent to the visibility analysis unit. The unit is then activated, and its internal algorithm precisely constructs a virtual line of sight from the observation point to the target point within a digitized 3D visibility field. This line of sight is not a simple straight line but fully considers the Earth's curvature, terrain elevation obstacles (such as mountains or buildings, if the model contains such data), and most importantly, the visibility values ​​within all the 3D mesh cells along the path. It indexes the corresponding visibility value in the 3D visibility field data signal using its 3D spatial coordinates and compares this value in real time with a preset, operationally configurable visibility pass-through threshold. If the visibility values ​​at all sampling points along the entire line of sight are better than the threshold, it is considered "pass-through"; conversely, if the visibility value at any point on the path is worse than the threshold, theoretically obstructing visible light transmission, it is considered "no pass-through". This calculation process is dynamic and rapid, responding in real time to updates to the 3D visibility field data signal. Ultimately, this unit generates a quantitative, black-and-white visibility assessment result. This result is not merely a simple "yes" or "no" Boolean value; it often includes detailed information such as the location of the first obstacle causing the lack of visibility or the lowest visibility value along the entire path. This highly business-instructive visibility assessment result does not exist in isolation. Instead, it is integrated into the intelligent analysis and early warning module as a core element, along with other early warning and predictive information generated by the module. Together, they constitute a comprehensive, decision-support-rich signal of the spatiotemporal evolution trend of visibility and early warning information. This signal is then transmitted to the integrated management application module for display and use, enabling decision-makers to scientifically, intuitively, and quantitatively assess the impact of visibility on key sight corridors.

[0030] Furthermore, the core algorithm of the source tracing analysis unit is based on the principle of the reverse trajectory model in atmospheric physics. Its operation can be figuratively understood as a kind of "digital backtracking": it uses the core location of the currently identified low-visibility area as a starting point (or multiple starting points), and along a direction opposite to the real-time wind direction, at a certain step size and frequency, it gradually calculates the most likely path trajectory of these low-visibility air masses over a previous period (e.g., the past few hours). This process needs to consider the possible changes in wind speed with altitude (if wind profile data is available) to construct a more accurate three-dimensional backtracking path. Through this complex calculation and deduction, the unit can simulate and identify in digital space the possible movement path of the air mass causing the low visibility in the current monitoring area and its final source direction. This "source path map" is extremely valuable; it not only helps meteorological analysts understand the mechanisms of weather processes but also provides crucial evidence for early warning—for example, if the current dense fog is identified as advection fog from a lake area to the west, and the westerly winds continue, it can be predicted that the fog mass will continue to move eastward and affect downstream areas. Ultimately, this valuable source analysis result will be seamlessly integrated into the visibility spatiotemporal evolution trend and early warning information signals generated by the intelligent analysis and early warning module. This ensures that the signal not only includes the current situation and predictions but also contains in-depth interpretations of the causes of phenomena, greatly enhancing the scientific rigor and foresight of the system output. Extending the scope of protection to the predictive capabilities of the intelligent analysis and early warning module, its key feature is the explicit inclusion of a more advanced time-series prediction unit. This unit's function is to achieve spatiotemporal prediction of future visibility changes, a significant leap beyond current situation perception and real-time analysis. The operation of this unit relies on three-dimensional visibility field data signals across two time dimensions: historical and current. This means that the system needs to continuously store continuous three-dimensional visibility field snapshots over a past period (e.g., the past 24 hours or longer), forming a spatiotemporal data cube. Telescopes (historical data) alone are insufficient for accurate future prediction; the core of this unit lies in its internally built and running specialized prediction model trained using machine learning algorithms. Machine learning algorithms, such as Long Short-Term Memory networks or other advanced time-series forecasting models, can automatically learn from massive amounts of historical spatiotemporal data and capture the complex patterns, nonlinear relationships, periodicities (such as diurnal variations and sea-land breeze variations) hidden behind visibility changes, as well as the strong correlations between states at different times. After training, the model can receive three-dimensional visibility field data sequences of several recent consecutive time steps, as well as other related meteorological data sequences that may be input simultaneously (such as spatiotemporal data of humidity, wind field, and temperature), and perform forward inference calculations through its complex internal network structure and memory mechanism.Its output is no longer a description of the current or past state, but a prediction of the visibility value on every three-dimensional grid cell within the monitoring area for a specific future time period (e.g., 1 hour, 3 hours, or 6 hours). Essentially, it generates a predictive three-dimensional visibility field data signal for a future moment. This prediction result is no longer a single value, but includes the changing trend of the entire space in the future. It constitutes the most forward-looking and important component of the spatiotemporal evolution trend of visibility and early warning information signals, providing crucial scientific basis and time windows for issuing early warnings and initiating emergency response procedures.

[0031] This invention discloses an intelligent air visibility monitoring system and method based on three-dimensional spatial visualization. It collects ground and air visibility and meteorological data through a multi-source sensor network, uses spatial interpolation algorithms to fuse discrete point data to generate a continuous three-dimensional visibility field, and utilizes volume rendering technology to transform the data into a three-dimensional fog effect visualization model with transparency gradients. Combined with visibility analysis, source tracing, and machine learning prediction algorithms, it dynamically presents the distribution and evolution trend of visibility in three-dimensional space. Furthermore, it provides spatial analysis and early warning functions through an integrated interactive interface, achieving a technological leap from point measurement to volume perception, and from digital display to three-dimensional visualization.

[0032] Therefore, the present invention provides an intelligent air visibility monitoring system and method based on three-dimensional spatial visualization, which solves the problems of insufficient spatial representativeness, lack of three-dimensional and intuitive presentation, and difficulty in achieving accurate early warning in traditional visibility monitoring methods.

[0033] The above embodiments are merely illustrative of the principles and effects of the present invention and are not intended to limit the invention. Any person skilled in the art can modify or alter the above embodiments without departing from the spirit and scope of the present invention. Therefore, all equivalent modifications or alterations made by those skilled in the art without departing from the spirit and technical concept disclosed in the present invention should still be covered by the claims of the present invention.

Claims

1. An intelligent air visibility monitoring system based on three-dimensional spatial visualization, characterized in that, include: A data sensing module is deployed in the target monitoring area to collect multi-source environmental sensing signals including visibility factors and meteorological factors. The data fusion processing module receives the multi-source environmental sensing signals and generates a three-dimensional visibility field data signal describing the monitoring area through a spatial interpolation algorithm. A 3D visualization engine module receives the 3D visibility field data signal and generates a dynamic 3D visibility visualization scene signal that is synchronized with the physical world state based on volume rendering technology. The intelligent analysis and early warning module receives the three-dimensional visibility field data signal and the dynamic three-dimensional visibility visualization scene signal, and generates visibility spatiotemporal evolution trend and early warning information signal by integrating spatial analysis and prediction algorithms. The integrated management application module receives the spatiotemporal evolution trend of visibility, the early warning information signal, and the dynamic three-dimensional visibility visualization scene signal, and generates an integrated monitoring signal that integrates the three-dimensional visualization situation, analysis results, and early warning instructions.

2. The intelligent air visibility monitoring system based on three-dimensional spatial visualization according to claim 1, characterized in that, The visibility factors collected by the data sensing module originate from forward scattering visibility meters or transmission visibility meters. The collected meteorological factors include at least wind speed, wind direction, ambient temperature, and relative humidity. The module also integrates an image sensing unit for collecting video image stream signals of the monitored area. All of the above signals together constitute the multi-source environmental sensing signal and are transmitted to the data fusion processing module.

3. The intelligent air visibility monitoring system based on three-dimensional spatial visualization according to claim 1, characterized in that, The spatial interpolation algorithm used by the data fusion processing module is either Kriging interpolation or inverse distance weighted interpolation. This module uses this algorithm to fuse spatially discrete multi-source environmental perception signals from the data perception module into a continuous, high-resolution three-dimensional mesh model. Each mesh node of this model contains an inferred visibility value, thereby forming the three-dimensional visibility field data signal.

4. The intelligent air visibility monitoring system based on three-dimensional spatial visualization according to claim 1, characterized in that, The 3D visualization engine module uses volume rendering technology to convert 3D visibility field data signals into visually perceptible fog-like 3D effects with different transparency and colors. The lower the visibility value, the denser the visual effect and the more warning color the area. The higher the visibility value, the more transparent the area. This generates a dynamic 3D visibility visualization scene signal that realistically reflects the actual spatial visibility distribution.

5. The intelligent air visibility monitoring system based on three-dimensional spatial visualization according to claim 1, characterized in that, The built-in visibility analysis unit of the intelligent analysis and early warning module can receive three-dimensional visibility field data signals and automatically calculate the visibility judgment result between the two points based on the user-defined observation point and target point positions. This result, together with the spatiotemporal evolution trend of visibility and early warning information signals, is generated and output.

6. The intelligent air visibility monitoring system based on three-dimensional spatial visualization according to claim 1, characterized in that, The intelligent analysis and early warning module has a built-in source tracing analysis unit that can combine real-time wind direction and speed data to reverse-engineer the three-dimensional visibility field data signal, simulate and identify the possible movement path and source direction of the low visibility air mass, and integrate the source tracing analysis results into the generated visibility spatiotemporal evolution trend and early warning information signal.

7. The intelligent air visibility monitoring system based on three-dimensional spatial visualization according to claim 1, characterized in that, The intelligent analysis and early warning module also integrates a time series prediction unit. This unit uses historical and current three-dimensional visibility field data signals and a prediction model trained by machine learning algorithms to predict the spatial distribution change trend of visibility in the monitoring area within a specific future period. The prediction result constitutes an important part of the spatiotemporal evolution trend of visibility and early warning information signals.

8. The intelligent air visibility monitoring system based on three-dimensional spatial visualization according to claim 1, characterized in that, The interactive interface driven by the integrated management application module provides 3D scene roaming and zooming functions, and supports users to perform cross-sectional analysis of dynamic 3D visibility visualization scene signals by drawing arbitrary cross-sections to view the visibility status at any position inside. The interactive commands and results are integrated into the integrated monitoring signal.

9. The intelligent air visibility monitoring system based on three-dimensional spatial visualization according to claim 1, characterized in that, After receiving the visibility spatiotemporal evolution trend and early warning information signal containing early warning information, the integrated management application module will automatically generate and send a standard format early warning signal to the external traffic management or public information release system. This early warning signal is a specific output form of the integrated monitoring signal.

10. A method for an intelligent air visibility monitoring system based on three-dimensional spatial visualization according to any one of claims 1-9, characterized in that, include: S1: Deployed in the target monitoring area to collect multi-source environmental sensing signals including visibility factors and meteorological factors; S2: Receives multi-source environmental sensing signals and generates a three-dimensional visibility field data signal to describe the three-dimensional spatial distribution of visibility within the monitoring area through a spatial interpolation algorithm; S3: Receives 3D visibility field data signals and generates dynamic 3D visibility visualization scene signals that are synchronized with the physical world state based on volume rendering technology; S4: Receives three-dimensional visibility field data signals and dynamic three-dimensional visibility visualization scene signals, and generates visibility spatiotemporal evolution trend and early warning information signals through integrated spatial analysis and prediction algorithms; S5: Receives visibility spatiotemporal evolution trend and early warning information signals and dynamic three-dimensional visibility visualization scene signals, generates a comprehensive monitoring signal that integrates three-dimensional visualization situation, analysis results and early warning instructions, and drives the interactive interface to display and publish it.

Images