A sensor-based three-dimensional model construction method

Abstract

The present invention relates to the field of model construction technology, and in particular to a sensor-based three-dimensional model construction method and system. The method comprises the following steps: obtaining initial data of a rapid current training water area; performing spatiotemporal calibration on the initial data of a rapid current training water area to generate calibration data of a rapid current training water area; performing data preprocessing on the calibration data of a rapid current training water area to generate standard rapid current training water area data; performing water surface line fluctuation averaging on the standard rapid current training water area data to generate a water surface mean curve of a rapid current training water area; using the water surface mean curve of a rapid current training water area to divide the water surface area of ​​a rapid current training water area to obtain an above-water area of ​​a rapid current training water area and an underwater area of ​​a rapid current training water area. The present invention improves the accuracy and quality of water area model construction through spatiotemporal calibration, data preprocessing, hierarchical modeling and dynamic rendering.

Description

Technical Field

[0001] The present invention relates to the technical field of model building, and in particular to a sensor-based three-dimensional model building method and system. Background Art

[0002] With the rapid development of computer technology, computational fluid dynamics (CFD) has become an important tool for water conservancy engineering research. CFD simulates fluid flow through numerical methods, providing an efficient and accurate method for the study of rapids training waters. The original two-dimensional CFD model can only simulate plane flow and cannot accurately reflect the characteristics of complex three-dimensional flow fields. With the advancement of computing power and algorithms, three-dimensional CFD models have gradually matured and can simulate three-dimensional flow characteristics in rapids, including the velocity, pressure distribution and turbulence characteristics of water flow. In recent years, combined with laser scanning, remote sensing and geographic information system (GIS) technology, researchers can obtain high-precision terrain and water data. These data can be used as the basic input of three-dimensional models, significantly improving the accuracy and authenticity of the models. In addition, the application of virtual reality (VR) and augmented reality (AR) technology has enabled three-dimensional models to be used not only for research and design, but also for training and education, improving the training effect. However, the current traditional methods often ignore the complex structure of the water area and the differences in characteristics of different regions, resulting in insufficient modeling results. At the same time, traditional models often only focus on static spatial complexity and ignore the impact of dynamic changes in the water area, which leads to low accuracy and quality in the construction of water area models. Summary of the invention

[0003] Based on this, it is necessary to provide a sensor-based three-dimensional model construction method and system to solve at least one of the above technical problems.

[0004] To achieve the above object, a sensor-based three-dimensional model construction method is provided, the method comprising the following steps:

[0005] Step S1: acquiring initial data of rapids training waters; performing spatiotemporal calibration on the initial data of rapids training waters to generate calibration data of rapids training waters; performing data preprocessing on the calibration data of rapids training waters to generate standard rapids training waters data;

[0006] Step S2: averaging the water surface line fluctuations of the standard rapid current training water area data to generate a rapid current training water area water surface mean curve; using the rapid current training water area water surface mean curve to divide the rapid current training water area into surface areas to obtain the rapid current training water area water surface area and the rapid current training water area underwater area; performing hierarchical modeling on the rapid current training water area and the rapid current training water area underwater area to generate rapid current training water area water surface area modeling data and rapid current training water area underwater area modeling data;

[0007] Step S3: performing static spatial complexity calculation on the modeling data of the water area of ​​the rapid current training water area to obtain static spatial complexity data of the rapid current training water area; performing dynamic spatial complexity calculation on the modeling data of the underwater area of ​​the rapid current training water area to obtain dynamic spatial complexity data of the rapid current training water area; constructing a hierarchical rendering strategy according to the static spatial complexity data of the rapid current training water area and the dynamic spatial complexity data of the rapid current training water area to obtain a hierarchical rendering strategy for the rapid current training water area;

[0008] Step S4: Adopting a hierarchical rendering strategy for the rapids training water area, the modeling data of the surface area of ​​the rapids training water area and the modeling data of the underwater area of ​​the rapids training water area are adapted to the dynamic changes of the water environment to generate a dynamic three-dimensional model of the rapids training water area; the dynamic three-dimensional model of the rapids training water area is containerized and deployed, and the container orchestration technology is used to perform real-time water monitoring of the containerized dynamic three-dimensional model of the rapids training water area, so as to realize the visual monitoring and early warning operation of the rapids training water area.

[0009] The present invention obtains the initial data of the rapid current training water area and performs time and space calibration on the data to generate the calibration data of the rapid current training water area. By calibrating the initial data, the accuracy of subsequent processing and analysis can be ensured. The calibrated rapid current training water area data is preprocessed. First, the water surface line of the water area is subjected to fluctuation averaging processing to generate a mean curve of the water surface. Then, the surface area of ​​the rapid current training water area is divided by the generated water surface mean curve to obtain an above-water area and an underwater area. Next, the above-water area and the underwater area are hierarchically modeled to generate corresponding modeling data for subsequent model rendering and analysis. The above-water area modeling data of the rapid current training water area is subjected to static spatial complexity calculation to obtain static spatial complexity data of the area. At the same time, the underwater area modeling data is subjected to dynamic spatial complexity calculation to obtain dynamic spatial complexity data of the area. According to these data, a hierarchical rendering strategy is constructed for subsequent model rendering and dynamic change adaptation. According to the regional hierarchical rendering strategy of the rapid current training water area, the above-water area modeling data and the underwater area modeling data are subjected to dynamic change adaptation of the water environment to generate a dynamic three-dimensional model of the rapid current training water area. Then, the dynamic three-dimensional model is deployed in a container and monitored in real time using container orchestration technology. This can achieve visual monitoring and early warning of rapids training waters. Therefore, the present invention improves the accuracy and quality of water area model construction through time and space calibration, data preprocessing, hierarchical modeling and dynamic rendering.

[0010] In this specification, a sensor-based three-dimensional model construction system is provided, which is used to execute the above-mentioned sensor-based three-dimensional model construction method. The sensor-based three-dimensional model construction system includes:

[0011] The water area calibration module is used to obtain the initial data of the rapid current training water area; perform time and space calibration on the initial data of the rapid current training water area to generate the rapid current training water area calibration data; perform data preprocessing on the rapid current training water area calibration data to generate the standard rapid current training water area data;

[0012] The water area stratification module is used to average the water surface line fluctuations of the standard rapid current training water area data to generate a rapid current training water area water surface mean curve; divide the rapid current training water area into surface areas using the rapid current training water area water surface mean curve to obtain the rapid current training water area water surface area and the rapid current training water area underwater area; perform hierarchical modeling on the rapid current training water area and the rapid current training water area underwater area to generate rapid current training water area water surface area modeling data and rapid current training water area underwater area modeling data;

[0013] The water area rendering module is used to perform static spatial complexity calculation on the modeling data of the water area above the rapid current training water area to obtain the static spatial complexity data of the rapid current training water area; perform dynamic spatial complexity calculation on the modeling data of the underwater area of ​​the rapid current training water area to obtain the dynamic spatial complexity data of the rapid current training water area; construct a hierarchical rendering strategy based on the static spatial complexity data of the rapid current training water area and the dynamic spatial complexity data of the rapid current training water area to obtain a hierarchical rendering strategy for the rapid current training water area;

[0014] The water area monitoring module is used to adapt the modeling data of the surface area of ​​the rapids training water area and the modeling data of the underwater area of ​​the rapids training water area to the dynamic changes of the water environment through the hierarchical rendering strategy of the rapids training water area, and generate a dynamic three-dimensional model of the rapids training water area; containerize the dynamic three-dimensional model of the rapids training water area, and use container orchestration technology to perform real-time water area monitoring of the containerized dynamic three-dimensional model of the rapids training water area, so as to realize the visual monitoring and early warning operations of the rapids training water area.

[0015] The beneficial effect of the present invention is that through time-space calibration and data preprocessing, the accuracy and completeness of the data of the rapids training water area can be improved, providing a reliable basis for subsequent analysis and modeling. Through the averaging of the water surface line fluctuations and hierarchical modeling of the water area, the above-water and underwater environments of the rapids training water area can be simulated more accurately, and the simulation accuracy and realism can be improved. Through the layered rendering strategy and dynamic change adaptation, the rendering process can be optimized according to the complexity and importance of different areas, the rendering efficiency can be improved, and the resource consumption can be reduced. Through containerized deployment and container orchestration technology, real-time monitoring and early warning of the dynamic three-dimensional model of the rapids training water area can be achieved, which helps to timely discover and respond to abnormal conditions in the water environment and ensure training safety. Through visual monitoring and early warning operations, the state and changes of the rapids training water area can be intuitively displayed, the user experience can be improved, and the timeliness and effectiveness of decision-making and countermeasures can be promoted. Therefore, the present invention improves the accuracy and quality of water area model construction through time-space calibration, data preprocessing, layered modeling and dynamic rendering. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 A schematic diagram of the steps of a sensor-based three-dimensional model construction method;

[0017] Figure 2 for Figure 1 Detailed implementation steps of step S2 in the flowchart;

[0018] Figure 3 for Figure 1 Detailed implementation steps of step S3 in FIG.

[0019] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0020] The following is a clear and complete description of the technical method of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by technicians in this field without creative work are within the scope of protection of the present invention.

[0021] In addition, the accompanying drawings are only schematic illustrations of the present invention and are not necessarily drawn to scale. The same reference numerals in the figures represent the same or similar parts, and their repeated description will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities and do not necessarily correspond to physically or logically independent entities. The functional entities can be implemented in software form, or implemented in one or more hardware modules or integrated circuits, or implemented in different networks and / or processor methods and / or microcontroller methods.

[0022] It should be understood that, although the terms "first", "second", etc. may be used herein to describe various units, these units should not be limited by these terms. These terms are used only to distinguish one unit from another unit. For example, without departing from the scope of the exemplary embodiments, the first unit may be referred to as the second unit, and similarly the second unit may be referred to as the first unit. The term "and / or" used herein includes any and all combinations of one or more of the listed associated items.

[0023] To achieve this, please refer to Figures 1 to 3 , a sensor-based three-dimensional model construction method, the method comprising the following steps:

[0024] Step S1: acquiring initial data of rapids training waters; performing spatiotemporal calibration on the initial data of rapids training waters to generate calibration data of rapids training waters; performing data preprocessing on the calibration data of rapids training waters to generate standard rapids training waters data;

[0025] Step S2: averaging the water surface line fluctuations of the standard rapid current training water area data to generate a rapid current training water area water surface mean curve; using the rapid current training water area water surface mean curve to divide the rapid current training water area into surface areas to obtain the rapid current training water area water surface area and the rapid current training water area underwater area; performing hierarchical modeling on the rapid current training water area and the rapid current training water area underwater area to generate rapid current training water area water surface area modeling data and rapid current training water area underwater area modeling data;

[0026] Step S3: performing static spatial complexity calculation on the modeling data of the water area of ​​the rapid current training water area to obtain static spatial complexity data of the rapid current training water area; performing dynamic spatial complexity calculation on the modeling data of the underwater area of ​​the rapid current training water area to obtain dynamic spatial complexity data of the rapid current training water area; constructing a hierarchical rendering strategy according to the static spatial complexity data of the rapid current training water area and the dynamic spatial complexity data of the rapid current training water area to obtain a hierarchical rendering strategy for the rapid current training water area;

[0027] Step S4: Adopting a hierarchical rendering strategy for the rapids training water area, the modeling data of the surface area of ​​the rapids training water area and the modeling data of the underwater area of ​​the rapids training water area are adapted to the dynamic changes of the water environment to generate a dynamic three-dimensional model of the rapids training water area; the dynamic three-dimensional model of the rapids training water area is containerized and deployed, and the container orchestration technology is used to perform real-time water monitoring of the containerized dynamic three-dimensional model of the rapids training water area, so as to realize the visual monitoring and early warning operation of the rapids training water area.

[0028] The present invention obtains the initial data of the rapid current training water area and performs time and space calibration on the data to generate the calibration data of the rapid current training water area. By calibrating the initial data, the accuracy of subsequent processing and analysis can be ensured. The calibrated rapid current training water area data is preprocessed. First, the water surface line of the water area is subjected to fluctuation averaging processing to generate a mean curve of the water surface. Then, the surface area of ​​the rapid current training water area is divided by the generated water surface mean curve to obtain an above-water area and an underwater area. Next, the above-water area and the underwater area are hierarchically modeled to generate corresponding modeling data for subsequent model rendering and analysis. The above-water area modeling data of the rapid current training water area is subjected to static spatial complexity calculation to obtain static spatial complexity data of the area. At the same time, the underwater area modeling data is subjected to dynamic spatial complexity calculation to obtain dynamic spatial complexity data of the area. According to these data, a hierarchical rendering strategy is constructed for subsequent model rendering and dynamic change adaptation. According to the regional hierarchical rendering strategy of the rapid current training water area, the above-water area modeling data and the underwater area modeling data are subjected to dynamic change adaptation of the water environment to generate a dynamic three-dimensional model of the rapid current training water area. Then, the dynamic three-dimensional model is deployed in a container and monitored in real time using container orchestration technology. This can achieve visual monitoring and early warning of rapids training waters. Therefore, the present invention improves the accuracy and quality of water area model construction through time and space calibration, data preprocessing, hierarchical modeling and dynamic rendering.

[0029] In the embodiment of the present invention, reference Figure 1 The above is a schematic diagram of the steps of a sensor-based three-dimensional model construction method of the present invention. In this example, the sensor-based three-dimensional model construction method includes the following steps:

[0030] Step S1: acquiring initial data of rapids training waters; performing spatiotemporal calibration on the initial data of rapids training waters to generate calibration data of rapids training waters; performing data preprocessing on the calibration data of rapids training waters to generate standard rapids training waters data;

[0031] In the embodiment of the present invention, the scope and boundary of the rapids training water area to be acquired, as well as the required data type and format are determined. Use appropriate sensors or data acquisition equipment (such as GPS, remote sensing satellites, water quality sensors, etc.) to collect data in the rapids training water area. The acquired data types specifically include geospatial data, hydrological and meteorological data, water quality data, etc., to fully understand the characteristics and environment of the rapids training water area. The collected initial data of the rapids training water area are calibrated in time and space to ensure that the time and space information of the data are accurately matched. Specifically, GPS positioning data is used for time calibration to unify the time information of different data sources to the same time reference. The spatial information is calibrated to ensure that the spatial information collected by different data sources is in the same coordinate system, and coordinate conversion and registration are performed to ensure the spatial consistency and accuracy of the data. The calibration data of the rapids training water area that has been calibrated in time and space is preprocessed to improve data quality and availability. For different types of data, preprocessing steps such as denoising, missing value filling, and outlier processing are required to ensure the integrity and accuracy of the data. The data is format converted and standardized to meet the requirements and standards of subsequent analysis and modeling.

[0032] Step S2: averaging the water surface line fluctuations of the standard rapid current training water area data to generate a rapid current training water area water surface mean curve; using the rapid current training water area water surface mean curve to divide the rapid current training water area into surface areas to obtain the rapid current training water area water surface area and the rapid current training water area underwater area; performing hierarchical modeling on the rapid current training water area and the rapid current training water area underwater area to generate rapid current training water area water surface area modeling data and rapid current training water area underwater area modeling data;

[0033] In the embodiment of the present invention, the water surface line in the standard rapids training water area data is subjected to fluctuation averaging processing to smooth the water surface fluctuations, reduce the influence of noise, and obtain the water surface mean curve. The water surface line data can be processed by a filtering algorithm (such as mean filtering, median filtering, Gaussian filtering, etc.) to smooth the fluctuation part and retain the main characteristics of the water area. The water surface mean curve is used to divide the water surface area of ​​the rapids training water area into the above-water area and the underwater area. Specifically, a threshold is set or a segmentation algorithm based on feature points is used to divide the water surface into the above-water area and the underwater area according to the fluctuation degree and shape of the water surface mean curve. The above-water area and the underwater area are modeled separately to meet the specific needs and characteristics of different areas. When modeling the above-water area, factors such as terrain, vegetation, and water surface characteristics are considered, and aerial images, terrain data, etc. are specifically used for modeling. When modeling the underwater area, factors such as underwater terrain, water flow, potential obstacles, etc. are considered, and sonar data, underwater photography, etc. are specifically used for modeling.

[0034] Step S3: performing static spatial complexity calculation on the modeling data of the water area of ​​the rapid current training water area to obtain static spatial complexity data of the rapid current training water area; performing dynamic spatial complexity calculation on the modeling data of the underwater area of ​​the rapid current training water area to obtain dynamic spatial complexity data of the rapid current training water area; constructing a hierarchical rendering strategy according to the static spatial complexity data of the rapid current training water area and the dynamic spatial complexity data of the rapid current training water area to obtain a hierarchical rendering strategy for the rapid current training water area;

[0035] In an embodiment of the present invention, static spatial complexity calculation is performed on the modeling data of the water area above the rapids training water area to evaluate the static complexity of the water area. The static spatial complexity is calculated specifically based on the characteristics of the water area modeling data, terrain complexity and other factors. Common methods include analyzing the number of grids, feature point density, number of boundaries, etc. Dynamic spatial complexity calculation is performed on the modeling data of the underwater area of ​​the rapids training water area to evaluate the dynamic complexity of the water area. Dynamic spatial complexity specifically considers factors such as the movement of underwater objects and the distribution of potential obstacles, and usually requires analysis and calculation based on information such as the position and speed of the object. Based on the static spatial complexity data and dynamic spatial complexity data of the rapids training water area, a layered rendering strategy is constructed to optimize rendering efficiency and user experience. Static spatial complexity and dynamic spatial complexity can be used as the basis for the layered rendering strategy, and the rendering priority and strategy can be determined according to the complexity level of different areas.

[0036] Step S4: Adopting a hierarchical rendering strategy for the rapids training water area, the modeling data of the surface area of ​​the rapids training water area and the modeling data of the underwater area of ​​the rapids training water area are adapted to the dynamic changes of the water environment to generate a dynamic three-dimensional model of the rapids training water area; the dynamic three-dimensional model of the rapids training water area is containerized and deployed, and the container orchestration technology is used to perform real-time water monitoring of the containerized dynamic three-dimensional model of the rapids training water area, so as to realize the visual monitoring and early warning operation of the rapids training water area.

[0037] In an embodiment of the present invention, according to the hierarchical rendering strategy of the rapids training water area, the modeling data of the water area and the underwater area are dynamically adapted. The details and accuracy of the water environment model can be dynamically adjusted according to the real-time monitoring data or the preset monitoring parameters to adapt to the real-time changes of the water environment. According to the adaptive water environment model, a dynamic three-dimensional model of the rapids training water area is generated. In combination with the hierarchical rendering strategy of the water area, the models of the water area and the underwater area are merged or rendered separately to form a complete dynamic three-dimensional model. The generated dynamic three-dimensional model is containerized and packaged into a container image for easy deployment and management. The container image is deployed to the corresponding computing environment using a containerization platform (such as Docker) to realize the operation and monitoring of the model. The containerized dynamic three-dimensional model is managed and monitored using container orchestration technology (such as Kubernetes). A monitoring program or service is designed to monitor the dynamic three-dimensional model of the rapids training water area in real time to detect changes and abnormalities in the water environment. A visual monitoring interface is designed and implemented to display the dynamic three-dimensional model and monitoring data of the rapids training water area. By combining monitoring data and preset warning rules, real-time monitoring and early warning operations of the water environment can be achieved to ensure the safety and effectiveness of whitewater training.

[0038] Preferably, step S1 comprises the following steps:

[0039] Step S11: obtaining initial data of rapids training water area;

[0040] Step S12: synchronizing the data timestamp of the initial data of the rapids training water area to generate the rapids training water area synchronization data;

[0041] Step S13: performing time-space calibration on the rapids training water area synchronization data to generate rapids training water area calibration data;

[0042] Step S14: performing data denoising on the rapids training water area calibration data to obtain rapids training water area denoised data; performing data standardization on the rapids training water area denoised data using the Z-socre standardization method to generate standard rapids training water area data.

[0043] The present invention ensures that the data is consistent in time by synchronizing the timestamp of the initial data of the rapids training water area. This helps with subsequent data analysis and processing to ensure the accuracy and consistency of the data. There are temporal and spatial deviations or inconsistencies in the rapids training water area. By performing temporal and spatial calibration on the synchronized data, these deviations can be eliminated, making the data more accurate and reliable. Temporal and spatial calibration can improve the accuracy of the data and ensure the consistency of the data in time and space. The rapids training water area data contains noise or interference. By denoising the calibration data, these interference factors can be eliminated and the real signal can be extracted. The denoised data is clearer and easier to analyze and understand. The denoised data is standardized using the Z-score standardization method, and the data can be converted into standard rapids training water area data with the same mean and standard deviation. The standardized data is convenient for comparison and statistical analysis, making the differences between different data clearer.

[0044] In an embodiment of the present invention, relevant data of the rapids training water area are collected, including information such as water flow velocity, water level, water temperature, etc. These data are specifically obtained through sensors, measuring equipment or other data acquisition methods. The collected initial data is time-stamped and synchronized to ensure the consistency of the time information of the data, so that data from different sources can be compared and analyzed on the same time axis. Performing data spatiotemporal calibration involves calibrating the spatial position of the data to ensure the accuracy and reliability of the data, such as correcting the sensor position or correcting the water area map. The calibrated data is denoised to remove existing interference or outliers to ensure the quality and credibility of the data. The denoised data is standardized using the Z-score standardization method so that the data has standard distribution characteristics, which is convenient for subsequent analysis and processing.

[0045] Preferably, step S2 comprises the following steps:

[0046] Step S21: confirming the water surface line of the rapid current training water area with respect to the standard rapid current training water area data, and generating the water surface line data of the rapid current training water area;

[0047] Step S22: performing time series fluctuation analysis on the water surface line data of the rapid current training water area to generate water surface line fluctuation data of the rapid current training water area;

[0048] Step S23: performing curve conversion on the water surface line fluctuation data of the rapid current training water area to generate a water surface fluctuation curve of the rapid current training water area; performing water surface averaging on the water surface fluctuation curve of the rapid current training water area to generate a water surface average curve of the rapid current training water area;

[0049] Step S24: using the mean water surface curve of the rapid current training water area to divide the surface area of ​​the rapid current training water area into the surface area of ​​the rapid current training water area and the underwater area of ​​the rapid current training water area; performing hierarchical modeling on the surface area of ​​the rapid current training water area and the underwater area of ​​the rapid current training water area to generate modeling data of the surface area of ​​the rapid current training water area and modeling data of the underwater area of ​​the rapid current training water area.

[0050] The present invention can determine the boundary line of the water area, that is, the water surface line, by confirming the water surface line of the standard rapids training water area data. This is helpful for subsequent water area analysis and modeling, and provides basic characteristic information of the water area. By performing time series fluctuation analysis on the water surface line data of the rapids training water area, the fluctuation of the water surface can be understood. This is helpful to study the dynamic changes of the water area, the wave characteristics and the influence of the water flow, and provides important data on the motion properties of the water area. By performing curve conversion on the water surface line fluctuation data of the rapids training water area, a water surface fluctuation curve can be generated. At the same time, the fluctuation curve is subjected to water surface averaging processing to obtain the average water surface curve of the water area. These curves provide detailed information on the water surface fluctuation characteristics of the water area, which helps to more deeply understand the fluctuation properties and change laws of the water area. The water area is divided into the above-water area and the underwater area using the water surface mean curve of the rapids training water area. Then, hierarchical modeling is performed for the above-water area and the underwater area respectively to generate corresponding modeling data. Such division and modeling help to study the characteristics and behaviors of different areas of the water area, and provide more detailed and comprehensive water area data.

[0051] As an embodiment of the present invention, refer to Figure 2 As shown, in this example, step S2 includes:

[0052] Step S21: confirming the water surface line of the rapid current training water area with respect to the standard rapid current training water area data, and generating the water surface line data of the rapid current training water area;

[0053] In an embodiment of the present invention, the image data of the rapids training water area is analyzed by using image processing or computer vision technology to identify the water body and non-water body area, and then determine the approximate position of the water surface line. The water edge in the image is detected and extracted. Commonly used algorithms include Canny edge detection, Sobel operator, etc. to obtain the boundary line between the water surface and the non-water surface. The extracted edge is curve fitted to obtain the specific shape and contour of the water surface line. The fitting specifically adopts polynomial fitting, spline curve fitting and other methods to adapt to the irregularity of the water surface line. The fitted water surface line is corrected and adjusted to ensure that it has a high degree of consistency with the actual water body boundary. The confirmed water surface line information is converted into a data format, such as a point set, a line segment or a curve equation to generate the water surface line data of the rapids training water area.

[0054] Step S22: performing time series fluctuation analysis on the water surface line data of the rapid current training water area to generate water surface line fluctuation data of the rapid current training water area;

[0055] In the embodiment of the present invention, the water surface line data of the rapids training water area is preprocessed, including removing the existing noise, outliers, etc., to ensure the accuracy and reliability of the data. Time series data is extracted from the water surface line data, such as the change of the height or position of the water surface line over time. The time series analysis method is specifically used to extract and process the data. Select a suitable time series fluctuation analysis method, such as Fourier transform: convert the time domain data to the frequency domain for analysis, and identify periodic fluctuations and frequency components. Wavelet transform: suitable for local frequency change analysis of time domain data, which can detect fluctuation characteristics on different time scales. Moving average: used to smooth time series data, highlight long-term trends and periodic fluctuations. Variance analysis: analyze the variance of fluctuations to understand the size and change of fluctuations. Autoregressive model: used to establish a relationship model between time series data and identify the laws and influencing factors of fluctuations. Using the selected fluctuation analysis method, extract the relevant characteristics of the water surface line fluctuation, such as amplitude, period, trend, etc. The results of the fluctuation analysis are visualized to intuitively present the characteristics and change trends of the water surface line fluctuations. Specifically, visualization is performed using line graphs, curve graphs, spectrum graphs, etc. Based on the results of the fluctuation analysis, the water surface fluctuation data of the rapids training water area is generated for subsequent simulation, analysis and application.

[0056] Step S23: performing curve conversion on the water surface line fluctuation data of the rapid current training water area to generate a water surface fluctuation curve of the rapid current training water area; performing water surface averaging on the water surface fluctuation curve of the rapid current training water area to generate a water surface average curve of the rapid current training water area;

[0057] In an embodiment of the present invention, the water surface line fluctuation data of the rapids training water area is converted into a fluctuation curve, which is specifically achieved by drawing the time series fluctuation data into a curve graph. The horizontal axis represents time, and the vertical axis represents the amplitude or position of the water surface fluctuation. Specifically, data processing and visualization tools are used, such as the Matplotlib library in Python, the ggplot2 package in the R language, etc., to convert the fluctuation data into a curve graph. The water surface fluctuation curve of the rapids training water area is averaged, that is, the influence of the fluctuation is removed and the average state of the water surface is retained. A commonly used method is to use a moving average method to calculate the average value of the fluctuation curve within a certain window, and then replace each data point in the original fluctuation curve with the average value to obtain a water surface mean curve.

[0058] Step S24: using the mean water surface curve of the rapid current training water area to divide the surface area of ​​the rapid current training water area into the surface area of ​​the rapid current training water area and the underwater area of ​​the rapid current training water area; performing hierarchical modeling on the surface area of ​​the rapid current training water area and the underwater area of ​​the rapid current training water area to generate modeling data of the surface area of ​​the rapid current training water area and modeling data of the underwater area of ​​the rapid current training water area.

[0059] In an embodiment of the present invention, the average height or position of the water surface can be determined by using the mean water surface curve of the rapids training water area. A threshold value or rule is set, and the rapids training water area is divided into an above-water area and an underwater area according to the height or position of the water surface mean curve. Usually, the part above the water surface mean is divided into the above-water area, and the part below the water surface mean is divided into the underwater area. The above-water area and the underwater area of ​​the rapids training water area are modeled separately. The above-water area modeling usually includes modeling the surface of the water body, and specifically uses geographic information system (GIS) software or three-dimensional modeling software to model the water surface. Underwater area modeling usually includes modeling of underwater terrain, water flow, water quality, etc., and specifically uses hydrological models, fluid dynamics models, etc. for modeling.

[0060] Preferably, the layered modeling of the surface area of ​​the rapid current training water area and the underwater area of ​​the rapid current training water area includes:

[0061] Use aerial cameras and laser radar to obtain aerial images of water areas and radar detection data of water areas;

[0062] Extract visual features from aerial images of water areas to generate water environment feature data for rapids training areas;

[0063] Performing feature point matching on water environment feature data of rapid current training water area to generate water environment feature points of rapid current training water area;

[0064] The radar detection data of the water area is used for triangulation to generate the water geographic elevation data of the rapids training water area;

[0065] Based on the water environment characteristic points of the rapid current training water area and the water geographic elevation data of the rapid current training water area, the water plane fitting is performed to generate the water area modeling data of the rapid current training water area;

[0066] Conduct multi-dimensional perception modeling of the underwater area of ​​the rapids training waters to generate modeling data for the underwater area of ​​the rapids training waters.

[0067] The present invention obtains aerial images and radar detection data of the water area by using aerial cameras and laser radars. These data can provide visual and geographical information of the water area. Visual feature extraction is performed on the aerial images of the water area to extract key feature information in the image. Feature point matching is then performed to match corresponding feature points in different images to establish a correlation relationship between feature points. The radar detection data of the water area is used to perform triangular mesh processing to generate geographic elevation data of the water area. These data can provide topographic and terrain information of the water area. Based on the environmental feature points and geographic elevation data of the water area, water plane fitting is performed, and the feature points and terrain elevation are merged to generate modeling data of the water area. These data can describe the shape, structure and geographical features of the water area. Multi-dimensional perceptual modeling is performed for the underwater area of ​​the rapids training water area. This includes using sonar, underwater cameras and other equipment to obtain data of the underwater area, and performing analysis and modeling to obtain information such as topography, water flow and object distribution in the underwater area.

[0068] In the embodiment of the present invention, an aerial camera is used to obtain an aerial image of the water area of ​​the rapids training water area to capture visual information on the water surface. The laser radar technology is used to obtain radar detection data of the water area to obtain information such as the distance and height of objects on the water surface. The aerial image is processed using computer vision technology to extract visual features in the water environment, such as the color, texture, boundary and other information of the water surface. Feature point matching is performed on the extracted visual feature data, and the corresponding feature points in adjacent images are paired to establish a corresponding relationship between the images. The radar detection data is processed, and the discrete point cloud data is converted into continuous geographic elevation data using a triangular meshing method to reflect the terrain morphology on the water surface. The feature points of the water environment and the geographic elevation data are combined to perform plane fitting to obtain the average shape and position of the water surface, thereby generating modeling data of the water area. Underwater detection equipment (such as sonar, underwater camera, etc.) is used to obtain multi-dimensional perception data of the underwater area, including underwater terrain, water flow, water quality and other information. The underwater area is modeled using methods such as hydrological models and fluid dynamics models to obtain data such as terrain and water flow in the underwater area.

[0069] Preferably, the multi-dimensional perception modeling of the underwater area of ​​the rapids training water area includes:

[0070] Underwater sonar is used to collect underwater sonar data of the underwater area of ​​the rapids training water area to obtain underwater area sonar data;

[0071] Perform sonar echo wavelength analysis on underwater area sonar data to generate sonar echo wavelength data; perform underwater terrain reconstruction based on sonar echo wavelength data to generate underwater area terrain data;

[0072] Perform object motion analysis based on underwater sonar data to generate stationary objects and moving objects in the underwater area; perform object detection on stationary objects in the underwater area to generate feature data of stationary objects in the underwater area; perform biometric identification on moving objects in the underwater area to generate biometric feature data of moving objects in the underwater area;

[0073] The underwater area terrain data is modeled and integrated based on the underwater area moving biological characteristic data and the underwater area stationary object characteristic data, so as to generate underwater area modeling data of rapids training waters.

[0074] The present invention collects sonar data of an underwater area in a rapids training water area by using an underwater sonar device. The sonar device sends a sound wave signal and receives an echo signal, and the information of the underwater area can be obtained by analyzing the echo signal. The echo wavelength analysis is performed on the underwater area sonar data to determine the wavelength of the sound wave propagating in the underwater area. The underwater terrain is reconstructed according to the sonar echo wavelength data to generate the terrain data of the underwater area. These data can provide the terrain and landform information of the underwater area. According to the underwater area sonar data, object motion analysis is performed. By analyzing the changes in the sonar echo signal, the stationary objects and the moving objects in the underwater area can be detected. Object detection is performed on the stationary objects to extract the characteristic data of the stationary objects. The moving objects are subjected to biological identification to extract the characteristic data of the moving organisms. Based on the characteristic data of the moving organisms, the characteristic data of the stationary objects and the terrain data in the underwater area, the modeling data is integrated. These data are fused and combined to generate the modeling data of the underwater area in the rapids training water area. These data can describe the terrain characteristics, object distribution and biological characteristics of the underwater area.

[0075] In the embodiment of the present invention, by using underwater sonar equipment to collect sonar data of the underwater area of ​​the rapids training water area, sonar data of the underwater area is obtained. Sonar specifically detects underwater objects, terrain, water flow and other information. The collected sonar data is subjected to wavelength analysis to identify the wavelength of the sonar echo to understand the terrain characteristics of the underwater area. According to the wavelength data of the sonar echo, a terrain reconstruction algorithm (such as a three-dimensional reconstruction algorithm based on sonar data) is used to reconstruct the terrain of the underwater area and generate underwater area terrain data. The collected sonar data is subjected to object motion analysis to identify stationary objects and moving objects in the underwater area. For the identified stationary objects in the underwater area, object detection and feature extraction are performed to generate feature data of the stationary objects in the underwater area. For the identified moving objects in the underwater area, biometric identification or target tracking is performed to extract their feature data and generate underwater area moving biological feature data. The underwater area terrain data, stationary object feature data and moving biological feature data are integrated to construct a comprehensive model data of the underwater area. Specifically, geographic information system software or professional 3D modeling software is used to integrate different types of data to form a comprehensive model of the underwater area.

[0076] Preferably, step S3 comprises the following steps:

[0077] Step S31: performing static spatial complexity calculation on the modeling data of the water area of ​​the rapids training water area to obtain static spatial complexity data of the rapids training water area;

[0078] Step S32: performing dynamic spatial complexity calculation on the underwater area modeling data of the rapids training water area to obtain dynamic spatial complexity data of the rapids training water area;

[0079] Step S33: sorting rendering priorities according to the static spatial complexity data of the rapids training water area and the dynamic spatial complexity data of the rapids training water area, and generating rapids training water area rendering priority data;

[0080] Step S34: constructing a hierarchical rendering strategy based on the rapids training water area rendering priority data to obtain a hierarchical rendering strategy for the rapids training water area.

[0081] The present invention obtains static spatial complexity data of a rapid current training water area by performing static spatial complexity calculation on modeling data of an aquatic area of ​​a rapid current training water area. Static spatial complexity refers to the size of space occupied by modeling data of an aquatic area in a static state. The spatial complexity of modeling data of an aquatic area can be evaluated by calculating the volume, area or other indicators of modeling data. Dynamic spatial complexity calculation is performed on modeling data of an underwater area of ​​a rapid current training water area to obtain dynamic spatial complexity data of a rapid current training water area. Dynamic spatial complexity refers to the size of space occupied by modeling data of an underwater area in a moving state. By considering factors such as the flow and change of an underwater area, the spatial complexity of modeling data in a dynamic scene is calculated. Rendering priority data of a rapid current training water area is sorted according to static spatial complexity data of a rapid current training water area and dynamic spatial complexity data of a rapid current training water area to generate rendering priority data of a rapid current training water area. According to the size of spatial complexity data, the aquatic and underwater areas are sorted to determine the rendering priority order. Areas with higher priorities will be rendered first to ensure a balance between rendering effect and performance. A hierarchical rendering strategy is constructed based on the rendering priority data of a rapid current training water area to obtain a hierarchical rendering strategy of a rapid current training water area. Based on the rendering priority data, the above-water and underwater areas are divided into different layers, each with different rendering strategies. For example, high-priority areas can be rendered in fine detail, while low-priority areas can be rendered in simplified detail, thereby improving rendering efficiency and performance.

[0082] As an example of the present invention, refer to Figure 3 As shown, in this example, step S3 includes:

[0083] Step S31: performing static spatial complexity calculation on the modeling data of the water area of ​​the rapids training water area to obtain static spatial complexity data of the rapids training water area;

[0084] In an embodiment of the present invention, static spatial complexity refers to the complexity of describing the structure, organization or layout of an object in the spatial dimension, and generally includes information on the density, distribution, shape, and other aspects of the object. Ensure that the modeling data of the water area in the rapids training waters has been acquired and prepared, including information such as the shape of the water surface, terrain, and distribution of stationary objects. Select a suitable static spatial complexity calculation method, and generally use one or more of the following methods according to the specific situation: Gridding method: Divide the water area modeling data into grid units, and calculate the density, uniformity, and other indicators of each grid unit. Distribution analysis method: Perform distribution analysis on the water area modeling data, including the distribution density and degree of aggregation of objects. Shape analysis method: Analyze the shape complexity of objects in the water area modeling data, the complexity of boundary curves, and the like. According to the selected calculation method, the water area modeling data of the rapids training waters is processed and analyzed, and the relevant indicators of static spatial complexity are calculated.

[0085] Step S32: performing dynamic spatial complexity calculation on the underwater area modeling data of the rapids training water area to obtain dynamic spatial complexity data of the rapids training water area;

[0086] In the embodiment of the present invention, dynamic spatial complexity refers to the complexity of describing the structure, organization or change law of an object in the spatial and temporal dimensions, and generally includes information on the movement, change, interaction, etc. of the object. Ensure that the modeling data of the underwater area in the rapids training waters has been acquired and prepared, including information on underwater terrain, stationary objects, moving objects, etc. Select a suitable dynamic spatial complexity calculation method, and generally adopt one or more of the following methods according to the specific situation: Motion trajectory analysis method: perform trajectory analysis on the moving objects in the underwater area modeling data, including information on speed, acceleration, motion path, etc. Spatiotemporal analysis method: perform spatiotemporal analysis on the underwater area modeling data, including the position of the object, the change of the motion state over time, etc. Interaction analysis method: analyze the interaction relationship between objects in the underwater area modeling data, including collision, aggregation, dispersion, etc. According to the selected calculation method, the underwater area modeling data of the rapids training waters is processed and analyzed, and the relevant indicators of dynamic spatial complexity are calculated. For the motion trajectory analysis method, the trajectory data of the moving objects in the underwater area can be extracted to analyze their motion speed, the tortuosity of the motion path, etc. For the spatiotemporal analysis method, the underwater area modeling data can be analyzed according to the time dimension to observe the changes in the position and state of the objects over time. For the interactive analysis method, the interaction between objects in the underwater area modeling data can be analyzed, such as detecting collisions, aggregation, dispersion, etc. The calculated dynamic spatial complexity data is verified to ensure its accuracy and reliability.

[0087] Step S33: sorting rendering priorities according to the static spatial complexity data of the rapids training water area and the dynamic spatial complexity data of the rapids training water area, and generating rapids training water area rendering priority data;

[0088] In the embodiment of the present invention, by ensuring that the static spatial complexity data and the dynamic spatial complexity data of the rapids training water area have been calculated, a suitable rendering priority sorting method is selected, and the priority is usually determined specifically according to the weights of the static and dynamic spatial complexity data. Other sorting methods can also be selected according to specific needs and scenarios. Common sorting methods include weight-based sorting, ranking-based sorting, threshold-based sorting, and the like.

[0089] Step S34: constructing a hierarchical rendering strategy based on the rapids training water area rendering priority data to obtain a hierarchical rendering strategy for the rapids training water area.

[0090] In the embodiment of the present invention, it is ensured that the rendering priority data of the rapids training water area has been prepared, and the data reflects the rendering importance or priority of different areas. According to the rendering priority data, the rapids training water area is divided into different levels or groups, and each level or group corresponds to a rendering priority. The layering strategy is determined specifically according to factors such as the priority level, the type of rendering object, and the importance in the scene. For each layer or group, a corresponding rendering scheme is formulated. This includes selecting appropriate rendering technology, rendering parameters, rendering effects, etc. High-priority areas use more complex rendering technology or higher-quality rendering effects to improve their visibility and attractiveness. Low-priority areas specifically use simpler rendering technology or low-resolution rendering effects to save computing resources and improve rendering efficiency. For high-priority areas, preloading, detail level rendering and other technologies are specifically used to optimize the rendering process to ensure that it is displayed in the user's field of view in a timely and smooth manner. For low-priority areas, delayed loading, level subdivision rendering and other technologies are specifically used to reduce rendering pressure to improve overall rendering efficiency. According to real-time needs and scene changes, the layered rendering strategy is dynamically adjusted. The rendering schemes and priorities of each level are specifically adjusted according to factors such as changes in user viewpoints and changes in rendering loads. Evaluate and verify the constructed layered rendering strategy to ensure that it can effectively improve rendering efficiency, reduce resource consumption, and meet the user's visual needs and experience.

[0091] Preferably, step S32 includes the following steps:

[0092] Step S321: Performing hydrodynamic simulation on the modeling data of the underwater area of ​​the rapids training water area to generate hydrodynamic simulation data of the underwater area of ​​the rapids training water area;

[0093] Step S322: performing underwater object motion impact analysis on underwater area modeling data of rapid current training water area according to underwater hydrodynamic simulation data of rapid current training water area, and generating underwater area object motion impact data;

[0094] Step S323: extracting spatial distribution features of underwater area object motion impact data to generate underwater area dynamic spatial distribution feature data, wherein the spatial distribution feature analysis includes point density and spatial aggregation;

[0095] Step S324: performing dynamic spatial complexity calculation on the dynamic spatial distribution characteristic data of the underwater area to obtain dynamic spatial complexity data of the rapids training water area.

[0096] The present invention generates hydrodynamic simulation data of underwater area in rapid training water area by performing hydrodynamic simulation on modeling data of underwater area in rapid training water area. By simulating the fluid movement and water flow changes in underwater area, hydrodynamic information of underwater area can be obtained, including parameters such as flow rate, pressure, vortex, etc. According to the underwater hydrodynamic simulation data of rapid training water area, underwater object motion impact analysis is performed on modeling data of underwater area in rapid training water area, and underwater area object motion impact data is generated. According to the hydrodynamic simulation results, the impact of water flow on underwater objects is analyzed, such as the push, resistance and rotation of water flow. By analyzing the motion of underwater objects, the motion characteristics and changes of objects in underwater area can be understood. Spatial distribution feature extraction is performed on underwater area object motion impact data, and underwater area dynamic spatial distribution feature data is generated. Spatial distribution feature analysis includes point density and spatial aggregation, etc. By analyzing underwater area object motion impact data, distribution features of objects in underwater area can be extracted, such as dense areas and aggregation patterns of objects. The dynamic spatial complexity of the underwater area dynamic spatial distribution feature data is calculated to obtain the dynamic spatial complexity data of the rapids training water area. Dynamic spatial complexity refers to the size of the space occupied by the dynamic distribution features of the underwater area. By calculating the volume, area or other indicators of the dynamic distribution feature data, the size of the dynamic spatial complexity of the underwater area can be evaluated.

[0097] In an embodiment of the present invention, a hydrodynamic simulation is performed on the modeling data of the underwater area of ​​the rapids training water area, and a numerical simulation method, such as computational fluid dynamics (CFD) simulation, is specifically used. The flow field of the underwater area of ​​the water area is simulated by CFD software, and parameters such as water flow velocity, flow direction, and pressure are considered to generate hydrodynamic simulation data of the underwater area of ​​the rapids training water area. The underwater hydrodynamic simulation data of the rapids training water area is used to analyze the impact of the movement of objects in the underwater area modeling data, specifically including the analysis of the force conditions, movement trajectories, speeds, etc. of underwater objects, and the impact of the water flow on the objects, resistance, etc. The impact data of the underwater area object movement is generated through the simulation results. The spatial distribution features of the underwater area object movement impact data are extracted. This includes analyzing the spatial distribution of the underwater object movement, such as the distribution density and aggregation of the object. Specifically, spatial analysis is performed using tools such as geographic information systems (GIS) to obtain dynamic spatial distribution feature data of the underwater area. Dynamic spatial complexity calculation is performed on the dynamic spatial distribution feature data of the underwater area. This step specifically uses various complexity calculation methods, such as the calculation method based on spatial entropy, to measure the complexity of the dynamic spatial distribution of the underwater area. Finally, the dynamic spatial complexity data of the rapids training water area is obtained.

[0098] Preferably, step S34 includes the following steps:

[0099] Step S341: Modeling hierarchical division of the rapids training water area rendering priority data to generate a rapids training water area modeling division hierarchy, wherein the rapids training water area modeling division hierarchy includes a basic hierarchy, a landscape hierarchy, a dynamic element hierarchy, an underwater area hierarchy, and a special effect hierarchy;

[0100] Step S342: sorting the base layer with high priority rendering, and performing high-precision rendering on the base layer to generate a high-precision rendering strategy; sorting the special effect layer with low priority rendering, and performing low-precision rendering on the special effect layer to generate a low-precision rendering strategy;

[0101] Step S343: performing rendering load analysis according to the high-precision rendering base level and the low-precision rendering special effect level to generate rendering load analysis data; calculating the remaining rendering load capacity on the rendering load analysis data based on a preset rendering load threshold to obtain the remaining rendering load capacity;

[0102] Step S344: Based on the remaining renderable load, the landscape level, dynamic element level and underwater area level are randomly rendered with dynamic precision to generate a dynamic adjustment rendering strategy; the high-precision rendering strategy, the dynamic adjustment rendering strategy and the low-precision rendering strategy are integrated to generate a graded rendering strategy for the rapids training water area.

[0103] The present invention generates a modeling division hierarchy of a rapids training water area by modeling hierarchical division of the rendering priority data of the rapids training water area. The rendering priority data is divided into different hierarchies, including a basic hierarchical level, a landscape hierarchical level, a dynamic element hierarchical level, an underwater area hierarchical level and a special effect hierarchical level. Through hierarchical division, the priority order of rendering can be determined according to the importance and characteristics of different areas. The basic hierarchical level is sorted for high priority rendering, and the basic hierarchical level is rendered with high precision to generate a high precision rendering strategy. The special effect hierarchical level is sorted for low priority rendering, and the special effect hierarchical level is rendered with low precision to generate a low precision rendering strategy. By sorting priorities and setting precisions for different hierarchical levels, the rendering methods of the basic hierarchical level and the special effect hierarchical level can be determined to achieve a balance between rendering effect and performance. Rendering load analysis is performed based on the high precision rendering basic hierarchical level and the low precision rendering special effect hierarchical level to generate rendering load analysis data. Based on a preset rendering load threshold, the rendering load analysis data is calculated for the remaining rendering load capacity to obtain the remaining rendering load capacity. By analyzing the load of high-precision rendering and low-precision rendering, the rendering capability and resource utilization of the system are evaluated. Based on the remaining renderable load, random dynamic precision rendering is performed on the landscape level, dynamic element level, and underwater area level to generate a dynamically adjusted rendering strategy. The high-precision rendering strategy, dynamic adjustment rendering strategy, and low-precision rendering strategy are integrated to generate a hierarchical rendering strategy for the rapids training water area. According to the remaining renderable load and the rendering requirements of different levels, the rendering strategy is flexibly adjusted to maximize the use of system resources and improve rendering efficiency.

[0104] In an embodiment of the present invention, by modeling the hierarchical division of the rendering priority data of the rapids training water area, specifically by analyzing and classifying the different characteristics and elements of the water area, it is divided into different levels such as the basic level, the landscape level, the dynamic element level, the underwater area level and the special effect level. Rendering sorting and rendering strategy formulation of different priorities are performed for each level. The basic level usually includes the basic structure and terrain of the water area, which requires high priority and high precision rendering; while the special effect level includes the light and shadow effects, ripple effects, etc. in the water area, and specifically adopts low priority and low precision rendering. According to these requirements, high precision rendering strategy and low precision rendering strategy are formulated. Rendering load analysis is performed according to the formulated rendering strategy. This includes analyzing the rendering tasks of each level and calculating the rendering load. According to the preset rendering load threshold, the remaining rendering load is calculated to determine the remaining available resources of the system. Random dynamic precision rendering is performed on the landscape level, the dynamic element level and the underwater area level based on the remaining renderable load. According to the remaining resources, the rendering accuracy and priority of each level are dynamically adjusted to achieve the best rendering effect. The high-precision rendering strategy, the dynamically adjusted rendering strategy and the low-precision rendering strategy are integrated to form a graded rendering strategy for the rapids training water area.

[0105] Preferably, step S4 comprises the following steps:

[0106] Step S41: Model rendering is performed on the modeling data of the surface area of ​​the rapid current training water area and the modeling data of the underwater area of ​​the rapid current training water area by using the rapid current training water area hierarchical rendering strategy to generate a rapid current training water area rendering three-dimensional model;

[0107] Step S42: using a machine learning method to perform adaptive operation on the three-dimensional model of the rapids training water area to dynamically change the water area environment, thereby generating a dynamic three-dimensional model of the rapids training water area;

[0108] Step S43: containerize and deploy the dynamic three-dimensional model of the rapids training water area, and use container orchestration technology to perform real-time monitoring of the containerized dynamic three-dimensional model of the rapids training water area, so as to realize visual monitoring and early warning operations of the rapids training water area.

[0109] The present invention renders the modeling data of the water area and the underwater area of ​​the rapid current training water area according to the hierarchical rendering strategy of the rapid current training water area, and generates a rendered three-dimensional model of the rapid current training water area. According to the setting of the hierarchical rendering strategy, the modeling data of different levels are rendered to generate a detailed and realistic three-dimensional model. The rendered three-dimensional model of the rapid current training water area is adaptive to the dynamic changes of the water environment by using a machine learning method. The dynamic changes of the rapid current training water area are predicted and adapted by the machine learning algorithm and model to generate a dynamic three-dimensional model of the rapid current training water area. This step can realize the dynamic simulation of environmental factors such as water flow, water level, and waves, so that the rendering model can reflect the actual changes of the rapid current training water area. The dynamic three-dimensional model of the rapid current training water area is containerized and deployed, and the containerized model is used to monitor the water area in real time by using the container orchestration technology. The dynamic three-dimensional model is deployed in the container environment, and the container orchestration technology is used to manage and monitor it. By real-time monitoring of the dynamic model of the rapid current training water area, visual monitoring and early warning operations of the water area can be performed, and potential problems or abnormal situations can be discovered in time.

[0110] In an embodiment of the present invention, the modeling data of the water area and the underwater area are assigned to different rendering levels according to the hierarchical rendering strategy of the rapids training water area. The rendering details and priorities of each level are determined to ensure that the key areas are rendered with high quality, and the secondary areas are simplified to improve efficiency. Appropriate rendering technology and tools (such as OpenGL, DirectX, Unity, Unreal Engine, etc.) are selected, and rendering parameters are configured to meet the needs of the water environment. Technologies such as lighting models, texture mapping, and shadow effects are applied to enhance the realism of the three-dimensional model. The modeling data of the water area is rendered to generate a three-dimensional model including elements such as terrain, water surface, and vegetation. The modeling data of the underwater area is rendered to generate a three-dimensional model including elements such as underwater terrain, objects, and water flow. The rendering results are merged to generate a complete three-dimensional model of the rapids training water area. Relevant environmental data, such as meteorological data, hydrological data, historical data, etc., are collected for training machine learning models. The data is preprocessed, including steps such as data cleaning, normalization, and feature extraction. An appropriate machine learning algorithm (such as a neural network, a random forest, a support vector machine, etc.) is selected, and the preprocessed data is used for training. Optimize model parameters to improve the prediction accuracy and adaptability of the model. Use the trained model to dynamically predict the environmental data of the rapids training waters and simulate the changes in the water environment. According to the model prediction results, adjust the rendering effect of the three-dimensional model in real time so that it can adaptively reflect the dynamic changes of the water environment. Update the machine learning model regularly, use new data for training and optimization, and ensure that the model can accurately reflect the latest changes in the water environment. Select appropriate containerization technology (such as Docker) to package the dynamic three-dimensional model of the rapids training waters and its dependent environment into a container. Write Dockerfile to configure the startup command and environment variables of the container to ensure that the model can run correctly in the container. Select a container orchestration tool (such as Kubernetes) to deploy and manage multiple containers to achieve high availability and scalability of the dynamic three-dimensional model. Configure orchestration strategies, including container scheduling, expansion, load balancing, etc., to ensure the stable operation of the system. Use monitoring tools in the container (such as Prometheus, Grafana) to monitor the operating status of the dynamic three-dimensional model and the water environment in real time. Set monitoring indicators and warning thresholds. When abnormal conditions (such as sudden rise in water level, abnormal temperature, etc.) are detected, the warning mechanism is triggered and the warning information is displayed through a visual interface. Combine monitoring data with 3D models to display the real-time status and changes of the water environment through a visual interface (such as Web, desktop applications, etc.). Provide interactive functions to allow users to view specific data, historical records, warning information, etc., to help users make timely decisions and response measures.

[0111] In this specification, a sensor-based three-dimensional model construction system is provided, which is used to execute the above-mentioned sensor-based three-dimensional model construction method. The sensor-based three-dimensional model construction system includes:

[0112] The water area calibration module is used to obtain the initial data of the rapid current training water area; perform time and space calibration on the initial data of the rapid current training water area to generate the rapid current training water area calibration data; perform data preprocessing on the rapid current training water area calibration data to generate the standard rapid current training water area data;

[0113] The water area stratification module is used to average the water surface line fluctuations of the standard rapid current training water area data to generate a rapid current training water area water surface mean curve; divide the rapid current training water area into surface areas using the rapid current training water area water surface mean curve to obtain the rapid current training water area water surface area and the rapid current training water area underwater area; perform hierarchical modeling on the rapid current training water area and the rapid current training water area underwater area to generate rapid current training water area water surface area modeling data and rapid current training water area underwater area modeling data;

[0114] The water area rendering module is used to perform static spatial complexity calculation on the modeling data of the water area above the rapid current training water area to obtain the static spatial complexity data of the rapid current training water area; perform dynamic spatial complexity calculation on the modeling data of the underwater area of ​​the rapid current training water area to obtain the dynamic spatial complexity data of the rapid current training water area; construct a hierarchical rendering strategy based on the static spatial complexity data of the rapid current training water area and the dynamic spatial complexity data of the rapid current training water area to obtain a hierarchical rendering strategy for the rapid current training water area;

[0115] The water area monitoring module is used to adapt the modeling data of the surface area of ​​the rapids training water area and the modeling data of the underwater area of ​​the rapids training water area to the dynamic changes of the water environment through the hierarchical rendering strategy of the rapids training water area, and generate a dynamic three-dimensional model of the rapids training water area; containerize the dynamic three-dimensional model of the rapids training water area, and use container orchestration technology to perform real-time water area monitoring of the containerized dynamic three-dimensional model of the rapids training water area, so as to realize the visual monitoring and early warning operations of the rapids training water area.

[0116] The beneficial effect of the present invention is that through time-space calibration and data preprocessing, the accuracy and completeness of the data of the rapids training water area can be improved, providing a reliable basis for subsequent analysis and modeling. Through the averaging of the water surface line fluctuations and hierarchical modeling of the water area, the above-water and underwater environments of the rapids training water area can be simulated more accurately, and the simulation accuracy and realism can be improved. Through the layered rendering strategy and dynamic change adaptation, the rendering process can be optimized according to the complexity and importance of different areas, the rendering efficiency can be improved, and the resource consumption can be reduced. Through containerized deployment and container orchestration technology, real-time monitoring and early warning of the dynamic three-dimensional model of the rapids training water area can be achieved, which helps to timely discover and respond to abnormal conditions in the water environment and ensure training safety. Through visual monitoring and early warning operations, the state and changes of the rapids training water area can be intuitively displayed, the user experience can be improved, and the timeliness and effectiveness of decision-making and countermeasures can be promoted. Therefore, the present invention improves the accuracy and quality of water area model construction through time-space calibration, data preprocessing, layered modeling and dynamic rendering.

[0117] Therefore, the embodiments should be regarded as illustrative and non-restrictive from all points, and the scope of the present invention is limited by the appended claims rather than the above description, and it is intended that all changes falling within the meaning and range of equivalent elements of the application documents are included in the present invention.

[0118] The above description is only a specific embodiment of the present invention, so that those skilled in the art can understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but should conform to the widest scope consistent with the principles and novel features invented herein.

Claims

1. A sensor-based three-dimensional model construction method, characterized in that: The following steps are involved: Step S1: using a sensor to collect data in the rapids training water area to obtain initial data of the rapids training water area; Performing time-space calibration on the initial data of the rapids training water area to generate calibration data of the rapids training water area; Preprocessing the calibration data of rapids training waters to generate standard rapids training waters data; Step S2: averaging the water surface line fluctuations of the standard rapids training water area data to generate a rapids training water area water surface mean curve; using the rapids training water area water surface mean curve to divide the rapids training water area into surface areas to obtain the rapids training water area above the water area and the rapids training water area below the water area; Performing hierarchical modeling on the surface area of ​​the rapid current training water area and the underwater area of ​​the rapid current training water area to generate modeling data of the surface area of ​​the rapid current training water area and modeling data of the underwater area of ​​the rapid current training water area; Step S3: performing static spatial complexity calculation on the modeling data of the surface area of ​​the rapid current training water area to obtain static spatial complexity data of the rapid current training water area; performing dynamic spatial complexity calculation on the modeling data of the underwater area of ​​the rapid current training water area to obtain dynamic spatial complexity data of the rapid current training water area; A hierarchical rendering strategy is constructed according to the static spatial complexity data of the rapids training water area and the dynamic spatial complexity data of the rapids training water area to obtain a hierarchical rendering strategy for the rapids training water area; Step S3 includes the following steps: Step S31: performing static spatial complexity calculation on the modeling data of the water area of ​​the rapids training water area to obtain static spatial complexity data of the rapids training water area; Step S32: performing dynamic spatial complexity calculation on the underwater area modeling data of the rapids training water area to obtain dynamic spatial complexity data of the rapids training water area; Step S32 includes the following steps: Step S321: Performing hydrodynamic simulation on the modeling data of the underwater area of ​​the rapids training water area to generate hydrodynamic simulation data of the underwater area of ​​the rapids training water area; Step S322: performing underwater object motion impact analysis on underwater area modeling data of rapid current training water area according to underwater hydrodynamic simulation data of rapid current training water area, and generating underwater area object motion impact data; Step S323: extracting spatial distribution features of underwater area object motion impact data to generate underwater area dynamic spatial distribution feature data, wherein the spatial distribution feature analysis includes point density and spatial aggregation; Step S324: performing dynamic spatial complexity calculation on the underwater area dynamic spatial distribution characteristic data to obtain dynamic spatial complexity data of the rapids training water area; Step S33: sorting rendering priorities according to the static spatial complexity data of the rapids training water area and the dynamic spatial complexity data of the rapids training water area, and generating rapids training water area rendering priority data; Step S34: constructing a hierarchical rendering strategy based on the rendering priority data of the rapids training water area to obtain a hierarchical rendering strategy for the rapids training water area; Step S34 includes the following steps: Step S341: Modeling hierarchical division of the rapids training water area rendering priority data to generate a rapids training water area modeling division hierarchy, wherein the rapids training water area modeling division hierarchy includes a basic hierarchy, a landscape hierarchy, a dynamic element hierarchy, an underwater area hierarchy, and a special effect hierarchy; Step S342: sorting the base layer with high priority rendering, and performing high-precision rendering on the base layer to generate a high-precision rendering strategy; sorting the special effect layer with low priority rendering, and performing low-precision rendering on the special effect layer to generate a low-precision rendering strategy; Step S343: performing rendering load analysis according to the high-precision rendering base level and the low-precision rendering special effect level to generate rendering load analysis data; calculating the remaining rendering load capacity on the rendering load analysis data based on a preset rendering load threshold to obtain the remaining rendering load capacity; Step S344: Based on the remaining renderable load, the landscape level, the dynamic element level and the underwater area level are randomly rendered with dynamic precision to generate a dynamic adjustment rendering strategy; the high-precision rendering strategy, the dynamic adjustment rendering strategy and the low-precision rendering strategy are integrated to generate a hierarchical rendering strategy for the rapids training water area; Step S4: Adopting a hierarchical rendering strategy for the rapids training water area, the modeling data of the surface area of ​​the rapids training water area and the modeling data of the underwater area of ​​the rapids training water area are adapted to the dynamic changes of the water environment to generate a dynamic three-dimensional model of the rapids training water area; the dynamic three-dimensional model of the rapids training water area is containerized and deployed, and the container orchestration technology is used to perform real-time water monitoring of the containerized dynamic three-dimensional model of the rapids training water area, so as to realize the visual monitoring and early warning operation of the rapids training water area.

2. The sensor-based three-dimensional model construction method according to claim 1, characterized in that: Step S1 includes the following steps: Step S11: obtaining initial data of rapids training water area; Step S12: synchronizing the data timestamp of the initial data of the rapids training water area to generate the rapids training water area synchronization data; Step S13: performing time-space calibration on the rapids training water area synchronization data to generate rapids training water area calibration data; Step S14: performing data denoising on the rapids training water area calibration data to obtain rapids training water area denoised data; performing data standardization on the rapids training water area denoised data using the Z-socre standardization method to generate standard rapids training water area data.

3. The sensor-based three-dimensional model construction method according to claim 1, characterized in that: Step S2 includes the following steps: Step S21: confirming the water surface line of the rapid current training water area with respect to the standard rapid current training water area data, and generating the water surface line data of the rapid current training water area; Step S22: performing time series fluctuation analysis on the water surface line data of the rapid current training water area to generate water surface line fluctuation data of the rapid current training water area; Step S23: performing curve conversion on the water surface line fluctuation data of the rapid current training water area to generate a water surface fluctuation curve of the rapid current training water area; performing water surface averaging on the water surface fluctuation curve of the rapid current training water area to generate a water surface average curve of the rapid current training water area; Step S24: using the mean water surface curve of the rapid current training water area to divide the surface area of ​​the rapid current training water area into the surface area of ​​the rapid current training water area and the underwater area of ​​the rapid current training water area; performing hierarchical modeling on the surface area of ​​the rapid current training water area and the underwater area of ​​the rapid current training water area to generate modeling data of the surface area of ​​the rapid current training water area and modeling data of the underwater area of ​​the rapid current training water area.

4. The sensor-based three-dimensional model construction method according to claim 3, characterized in that: The hierarchical modeling of the surface area and underwater area of ​​the rapids training water area includes: Use aerial cameras and laser radar to obtain aerial images of water areas and radar detection data of water areas; Extract visual features from aerial images of water areas to generate water environment feature data for rapids training areas; Performing feature point matching on water environment feature data of rapid current training water area to generate water environment feature points of rapid current training water area; The radar detection data of the water area is used for triangulation to generate the water geographic elevation data of the rapids training water area; Based on the water environment characteristic points of the rapid current training water area and the water geographic elevation data of the rapid current training water area, the water plane fitting is performed to generate the water area modeling data of the rapid current training water area; Conduct multi-dimensional perception modeling of the underwater area of ​​the rapids training waters to generate modeling data for the underwater area of ​​the rapids training waters.

5. The sensor-based three-dimensional model construction method according to claim 4, characterized in that: Multi-dimensional perception modeling of underwater areas in rapids training waters includes: Underwater sonar is used to collect underwater sonar data of the underwater area of ​​the rapids training water area to obtain underwater area sonar data; Perform sonar echo wavelength analysis on underwater area sonar data to generate sonar echo wavelength data; perform underwater terrain reconstruction based on sonar echo wavelength data to generate underwater area terrain data; Perform object motion analysis based on underwater sonar data to generate stationary objects and moving objects in the underwater area; perform object detection on stationary objects in the underwater area to generate feature data of stationary objects in the underwater area; perform biometric identification on moving objects in the underwater area to generate biometric feature data of moving objects in the underwater area; The underwater area terrain data is modeled and integrated based on the underwater area moving biological characteristic data and the underwater area stationary object characteristic data, so as to generate underwater area modeling data of rapids training waters.

6. The sensor-based three-dimensional model construction method according to claim 1, characterized in that: Step S4 includes the following steps: Step S41: Model rendering is performed on the modeling data of the surface area of ​​the rapid current training water area and the modeling data of the underwater area of ​​the rapid current training water area by using the rapid current training water area hierarchical rendering strategy to generate a rapid current training water area rendering three-dimensional model; Step S42: using a machine learning method to perform adaptive operation on the three-dimensional model of the rapids training water area to dynamically change the water area environment, thereby generating a dynamic three-dimensional model of the rapids training water area; Step S43: containerize and deploy the dynamic three-dimensional model of the rapids training water area, and use container orchestration technology to perform real-time monitoring of the containerized dynamic three-dimensional model of the rapids training water area, so as to realize visual monitoring and early warning operations of the rapids training water area.

7. A sensor-based three-dimensional model building system, characterized in that: For executing the sensor-based three-dimensional model building method according to claim 1, the sensor-based three-dimensional model building system comprises: The water area calibration module is used to obtain the initial data of the rapid current training water area; perform time and space calibration on the initial data of the rapid current training water area to generate the rapid current training water area calibration data; perform data preprocessing on the rapid current training water area calibration data to generate the standard rapid current training water area data; The water area stratification module is used to average the water surface line fluctuations of the standard rapid current training water area data to generate a rapid current training water area water surface mean curve; divide the rapid current training water area into surface areas using the rapid current training water area water surface mean curve to obtain the rapid current training water area water surface area and the rapid current training water area underwater area; perform hierarchical modeling on the rapid current training water area and the rapid current training water area underwater area to generate rapid current training water area water surface area modeling data and rapid current training water area underwater area modeling data; The water area rendering module is used to perform static spatial complexity calculation on the modeling data of the water area above the rapid current training water area to obtain the static spatial complexity data of the rapid current training water area; perform dynamic spatial complexity calculation on the modeling data of the underwater area of ​​the rapid current training water area to obtain the dynamic spatial complexity data of the rapid current training water area; construct a hierarchical rendering strategy based on the static spatial complexity data of the rapid current training water area and the dynamic spatial complexity data of the rapid current training water area to obtain a hierarchical rendering strategy for the rapid current training water area; The water area monitoring module is used to adapt the modeling data of the surface area of ​​the rapids training water area and the modeling data of the underwater area of ​​the rapids training water area to the dynamic changes of the water environment through the hierarchical rendering strategy of the rapids training water area, and generate a dynamic three-dimensional model of the rapids training water area; containerize the dynamic three-dimensional model of the rapids training water area, and use container orchestration technology to perform real-time water area monitoring of the containerized dynamic three-dimensional model of the rapids training water area, so as to realize the visual monitoring and early warning operations of the rapids training water area.

Images