Terrain mapping method and system

Abstract

The application provides a topographic mapping method and system. The method comprises: analyzing the wind speed, precipitation intensity, cloud optical thickness and visibility of the mapping area; calculating the anti-interference coefficient based on the wind speed, precipitation intensity and performance parameters of the unmanned aerial vehicle, if the anti-interference coefficient is lower than the safety threshold, reducing the flight height to a preset safety proportion value of the reference height, starting the laser radar into standby state; according to the comparison result of the cloud optical thickness and the cloud thickness threshold, starting or closing the multispectral camera, controlling the polarization sensor to collect polarization data under the corresponding working waveband; if the visibility is lower than the visibility threshold or the precipitation intensity exceeds the precipitation threshold, waking up the laser radar in standby state to scan the terrain, setting the scanning density to be positively correlated with the precipitation intensity and negatively correlated with the visibility, and generating point cloud data; fusing the point cloud data and the polarization data at the edge end to generate a weather-corrected basic ground model, and if there is ground spectral data, fusing it to generate an enhanced ground model.

Description

Technical Field

[0001] This application relates to the field of meteorological services, and more specifically, to a topographic mapping method and system. Background Technology

[0002] In topographic mapping within the meteorological service sector, existing technologies suffer from a core problem: insufficient coordination between dynamic meteorological factors and mapping equipment. Current UAV mapping relies on offline modeling, failing to dynamically adjust equipment operation based on real-time changing meteorological parameters (such as wind speed, precipitation, and cloud cover), leading to data acquisition interruptions or decreased accuracy under severe weather conditions. Simultaneously, there is a lack of online coordination mechanisms between meteorological elements (such as visibility and cloud optical thickness) and data collected by equipment like lidar and multispectral cameras, and a real-time modeling process for meteorological correction has not been established. Furthermore, fixed threshold control strategies cannot intelligently start and stop equipment under complex weather conditions, resulting in redundant scanning or missing critical data, and increased system power consumption. These problems severely restrict the timeliness and reliability of topographic mapping in meteorologically sensitive scenarios such as typhoon monitoring and flood disaster assessment.

[0003] Therefore, this application provides a topographic mapping method and system to solve one of the aforementioned technical problems. Summary of the Invention

[0004] The purpose of this application is to provide a topographic mapping method and system that can solve at least one of the aforementioned technical problems. The specific solution is as follows:

[0005] According to a specific embodiment of this application, in a first aspect, this application provides a terrain mapping method, comprising: acquiring meteorological data streams in real time, analyzing wind speed, precipitation intensity, cloud optical thickness, and visibility of the mapping area; calculating an anti-interference coefficient based on the wind speed, the precipitation intensity, and the performance parameters of a UAV; if the anti-interference coefficient is lower than a safety threshold, reducing the flight altitude of the UAV to a preset safety ratio value of a reference altitude, and then activating a lidar to put the UAV into a standby state; and, based on the comparison result of the cloud optical thickness and the cloud thickness threshold, activating or deactivating a multispectral camera, and controlling a polarization sensor to collect data in the corresponding working band. The system collects polarization data of the surveyed area, wherein the multispectral camera is used to collect surface spectral data of the surveyed area; in response to determining that the visibility is lower than the visibility threshold or the precipitation intensity exceeds the precipitation threshold, the standby lidar is activated to perform terrain scanning of the surveyed area, the scanning density is set to be positively correlated with the precipitation intensity and negatively correlated with the visibility, and point cloud data is generated; the point cloud data and the polarization data are fused at the edge to generate a meteorologically corrected basic surface model, and if the surface spectral data exists, it is fused with the basic surface model to generate an enhanced surface model.

[0006] According to a specific embodiment of this application, in a second aspect, this application provides a terrain mapping system, comprising: a data parsing unit, configured to acquire meteorological data streams in real time and parse the wind speed, precipitation intensity, cloud optical thickness, and visibility of the mapping area; a processing unit, configured to calculate an anti-interference coefficient based on the wind speed, the precipitation intensity, and the performance parameters of a UAV; if the anti-interference coefficient is lower than a safety threshold, reduce the flight altitude of the UAV to a preset safety ratio value of a reference altitude, and then activate the lidar to put the UAV into a standby state; and based on the comparison result of the cloud optical thickness and the cloud thickness threshold, activate or deactivate a multispectral camera and control the polarization sensor to operate on the corresponding wavelength. The system collects polarization data of the surveyed area, wherein the multispectral camera is used to collect surface spectral data of the surveyed area; in response to determining that the visibility is lower than the visibility threshold or the precipitation intensity exceeds the precipitation threshold, the standby lidar is activated to perform terrain scanning of the surveyed area, the scanning density is set to be positively correlated with the precipitation intensity and negatively correlated with the visibility, and point cloud data is generated; a surface model generation unit is used to fuse the point cloud data and the polarization data at the edge to generate a meteorologically corrected basic surface model, and if the surface spectral data exists, it is fused with the basic surface model to generate an enhanced surface model.

[0007] Compared with the prior art, the above-described solution of this application has at least the following beneficial effects: by constructing a dynamic mapping closed loop driven by meteorological parameters, this application significantly improves the efficiency of terrain data acquisition and modeling accuracy under complex meteorological conditions while ensuring the safe operation of equipment. Specifically:

[0008] (1) By analyzing meteorological data in real time, the limitations of existing offline modeling technology are overcome, providing a real-time data foundation for dynamic decision-making;

[0009] (2) Couple wind speed / precipitation with the UAV’s own performance parameters to calculate the anti-disturbance coefficient, so as to intelligently reduce the flight altitude to avoid the risk of severe weather and reduce the probability of equipment damage;

[0010] (3) Based on the cloud thickness threshold, the multispectral camera is controlled to adaptively start and stop the acquisition of surface spectral data according to the optical thickness, which effectively avoids cloud interference and reduces the amount of invalid data;

[0011] (4) Based on the dual threshold mechanism of visibility and precipitation intensity, the lidar is activated and the scanning density is dynamically matched with the meteorological conditions, which helps to improve the accuracy of point cloud data acquisition and optimize the balance between data quality and energy consumption.

[0012] (5) Multi-source data fusion modeling is carried out through the edge terminal to realize the hierarchical fusion of the meteorological correction base model and the enhanced surface model, solve the data collaboration bottleneck under meteorological interference, and improve the reliability of the surface model. Attached Figure Description

[0013] Figure 1 A flowchart of a topographic mapping method is shown;

[0014] Figure 2 A block diagram of a topographic mapping system according to an embodiment of this application is shown. Detailed Implementation

[0015] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0016] It should be noted that any symbols and / or numbers present in the specification that are not marked in the accompanying drawings are not reference numerals.

[0017] The embodiments provided in this application are embodiments of a topographic mapping method. The following is a combination of... Figure 1 The embodiments of this application will be described in detail.

[0018] Figure 1 A flowchart of a topographic mapping method is shown. The method includes the following steps:

[0019] S101 acquires meteorological data streams in real time and analyzes wind speed, precipitation intensity, cloud optical thickness, and visibility in the surveyed area;

[0020] S102 calculates the anti-interference coefficient based on wind speed, precipitation intensity and drone performance parameters. If the anti-interference coefficient is lower than the safety threshold, the drone's flight altitude is reduced to a preset safety ratio value of the reference altitude, and then the lidar is activated to put the drone into standby mode.

[0021] S103, based on the comparison result of cloud optical thickness and cloud thickness threshold, start or stop the multispectral camera, and control the polarization sensor to collect polarization data of the survey area in the corresponding working band. The multispectral camera is used to collect surface spectral data of the survey area.

[0022] S104, in response to determining that the visibility is below the visibility threshold or the precipitation intensity exceeds the precipitation threshold, wake up the standby lidar to perform terrain scanning on the survey area, set the scanning density to be positively correlated with the precipitation intensity, and set the scanning density to be negatively correlated with the visibility, and generate point cloud data.

[0023] S105 integrates point cloud data and polarization data at the edge to generate a meteorologically corrected basic surface model. If surface spectral data exists, it is integrated with the basic surface model to generate an enhanced surface model.

[0024] In some embodiments, the topographic mapping method provided in this application is applied to topographic mapping tasks of unmanned aerial vehicle platforms in meteorologically sensitive scenarios such as typhoon monitoring and flood disaster assessment.

[0025] In step S101, meteorological data streams of the surveyed area are acquired in real time using airborne meteorological sensors, and key meteorological parameters, including wind speed, precipitation intensity, cloud optical thickness, and visibility, are analyzed. Cloud optical thickness is the degree of light attenuation by clouds at a specific wavelength, and its value is equal to the natural logarithm of the ratio of incident radiation intensity to transmitted radiation intensity. Cloud optical thickness is used to quantify the cloud's shading effect on optical imaging equipment; a larger value indicates poorer cloud transmittance. Cloud optical thickness is dimensionless and can be measured in the 550nm band using a multispectral camera or solar radiometer. The cloud thickness threshold is set to 0.1-0.5; a larger value indicates thicker clouds, and the magnitude of cloud optical thickness directly affects the acquisition capability of optical equipment. Wind speed is measured in meters per second (m / s) and is measured in real time using meteorological sensors mounted on the UAV, with a typical measurement range of 0-60 m / s. Precipitation intensity is measured in millimeters per hour (mm / h), representing the amount of precipitation per unit time. It is obtained through raindrop spectrometers or radar inversion, and the precipitation threshold is typically set between 10 and 50 mm / h. Visibility is measured in kilometers (km) and is measured using a forward-scattering visibility meter. The visibility threshold is set between 0.5 and 5 km, with lower values ​​indicating worse weather conditions.

[0026] The immunity coefficient is a quantitative indicator of a drone's ability to maintain stable flight under specific weather conditions. It is defined as the reciprocal of the weather interference weighted value, which is calculated by multiplying the real-time wind speed by a first dynamic calibration coefficient and the precipitation intensity by a second dynamic calibration coefficient. The immunity coefficient comprehensively reflects the impact of wind speed and precipitation on the drone's flight stability; a lower value indicates stronger environmental interference. The immunity coefficient has no dimensionless parameter and is obtained by taking the reciprocal of the product of wind speed (m / s) and dynamic calibration coefficient (s / m), and the product of precipitation intensity (mm / h) and dynamic calibration coefficient (h / mm).

[0027] The flight altitude is measured in meters (m). The reference altitude is the initial cruising altitude of the UAV performing the current mapping task, and can be set between 100 and 1000 meters. The preset safety ratio value refers to a pre-set flight altitude adjustment coefficient used to calculate the safe flight altitude when the anti-interference coefficient is below the safety threshold. Specifically, it is the reference altitude multiplied by this ratio value, for example, in the range of 0.6-0.8. The preset safety ratio value is dimensionless; for example, a preset safety ratio value of 0.7 represents 70% of the reference altitude. The scan density is measured in points per square meter (pts / m²), and the reference density can be set between 100 and 1000 pts / m², dynamically adjusted through the first and second gain coefficients.

[0028] As a feasible implementation, an immunity coefficient is calculated based on real-time wind speed, precipitation intensity, and the performance parameters of the UAV. The immunity coefficient is defined as follows: the product of real-time wind speed and a first dynamic calibration coefficient is calculated, and then the product of precipitation intensity and a second dynamic calibration coefficient is superimposed. The reciprocal of this superimposed value is taken to obtain the immunity coefficient, which is also the meteorological interference weighted value. When the immunity coefficient is lower than a preset safety threshold, the UAV is controlled to reduce its flight altitude to a preset safety proportion of the reference altitude, such as 70% of the reference altitude, and the lidar is activated to enter standby mode to avoid the risk of severe weather.

[0029] In one specific embodiment, the optical equipment is controlled based on a comparison between the cloud optical thickness and a cloud thickness threshold. If the cloud optical thickness does not exceed the threshold (e.g., optical thickness ≤ 0.3), the multispectral camera is activated to collect surface spectral data of the surveyed area; otherwise, the multispectral camera is turned off to avoid cloud interference. Simultaneously, the operating band of the polarization sensor is controlled. When the cloud optical thickness does not exceed the threshold, the visible light band is used to collect polarization data; when it exceeds the threshold, the infrared band is switched to penetrate the cloud.

[0030] For example, when the system responds to visibility falling below the visibility threshold by 1 km, or precipitation intensity exceeding the precipitation threshold by 30 mm / h, it immediately wakes up the standby lidar to perform terrain scanning. The scanning density is dynamically configured to be positively correlated with precipitation intensity and negatively correlated with visibility. Specifically, a preset baseline density multiplied by the product of an altitude attenuation factor and a reflection gain factor is used as the base value for the scanning density. This base value is then multiplied by a first gain coefficient as precipitation intensity increases. The first gain coefficient is determined based on the remaining power of the UAV's power system. As visibility decreases, the base value is multiplied by a second gain coefficient, which is determined based on cloud optical thickness. This generates point cloud data adapted to meteorological conditions.

[0031] In the edge data processing stage, point cloud data and polarization data are first fused to generate a meteorologically corrected basic surface model. Specifically, this includes: constructing an atmospheric transmittance tensor based on precipitation intensity and cloud optical thickness; separating the true surface polarization signal from atmospheric scattering noise using a tensor decomposition algorithm; correcting atmospheric polarization distortion in the polarization data; converting the corrected signal into a terrain normal field; and spatially fusing it with the point cloud data using a non-rigid registration algorithm. If surface spectral data exists, it is fused with the basic surface model to generate an enhanced surface model: extracting the absorption depth of the moisture characteristic absorption bands in the surface spectral data; calculating the elevation correction based on soil type parameters using the dielectric property-water content mapping relationship; and performing dielectric correction on the basic model to generate a humidity-compensated elevation field. When the multispectral camera covers the mineral characteristic response bands, a lithology classification raster is further generated as an independent layer and associated with spatial coordinates, forming an enhanced surface model containing a dielectric-corrected elevation field and a lithology layer.

[0032] The aforementioned topographic mapping method, by constructing a dynamic mapping closed loop driven by meteorological parameters, significantly improves the efficiency of topographic data acquisition and modeling accuracy under complex meteorological conditions while ensuring the safe operation of equipment. Specifically: Real-time analysis of meteorological data overcomes the limitations of existing offline modeling technologies, providing a real-time data foundation for dynamic decision-making; wind speed / precipitation is coupled with the UAV's own performance parameters to calculate the anti-interference coefficient, intelligently reducing flight altitude to avoid severe weather risks and decrease the probability of equipment damage; based on cloud thickness thresholds, the multispectral camera is controlled to adaptively start and stop acquiring surface spectral data according to optical thickness, effectively avoiding cloud interference and reducing invalid data; based on a dual threshold mechanism of visibility and precipitation intensity, the lidar is activated, dynamically matching scanning density with meteorological conditions, which helps improve the accuracy of point cloud data acquisition and optimizes the balance between data quality and energy consumption; multi-source data fusion modeling is performed at the edge, achieving hierarchical fusion of the meteorological correction base model and the enhanced surface model, solving the data collaboration bottleneck under meteorological interference and improving the reliability of the surface model.

[0033] In some embodiments, step S102, which calculates the anti-interference coefficient based on wind speed, precipitation intensity, and the performance parameters of the UAV, includes: calculating the product of real-time wind speed and a first dynamic calibration coefficient, and calculating the product of precipitation intensity and a second dynamic calibration coefficient; adding the two products together to obtain a meteorological interference weighted value; taking the reciprocal of the meteorological interference weighted value to obtain the anti-interference coefficient; wherein the first dynamic calibration coefficient is determined based on the maximum wind resistance level calibrated by the UAV and the real-time flight altitude, and the second dynamic calibration coefficient is determined based on the hydrophobic angle measurement value of the UAV and the remaining power of the power system.

[0034] The first dynamic calibration coefficient is measured in seconds per meter (s / m) and is generated by calibrating the ratio of the UAV's maximum wind resistance level (m / s) to its real-time altitude (m). The second dynamic calibration coefficient is measured in hours per millimeter (h / mm) and is calculated based on the product of the sine of the hydrophobic angle (°) and the remaining power of the power system (percentage %). The elevation correction is measured in meters (m) and is calculated through the mapping relationship between the soil dielectric constant and the water content (volume percentage %). The soil dielectric constant is a dimensionless physical quantity that characterizes the soil's ability to store and dissipate electrical energy in an electromagnetic field.

[0035] In some embodiments, the UAV anti-interference coefficient calculation method of this application is specifically implemented in typhoon monitoring and topographic mapping tasks. When the UAV is cruising at a reference altitude of 500 meters in a coastal area, the airborne meteorological station obtains a wind speed of 15 meters per second in real time, equivalent to a level 7 wind, and the millimeter-wave radar simultaneously monitors a precipitation intensity of 25 millimeters per hour, equivalent to a rainstorm. Based on the UAV's pre-calibrated maximum wind resistance level of 20 meters per second and the current flight altitude of 300 meters, the system calculates a first dynamic calibration coefficient of 0.0533 seconds per meter. This coefficient increases as the altitude decreases, reflecting the improvement of wind resistance margin in the low-altitude area. At the same time, a 45-degree hydrophobic angle is measured by the wing hydrophobic angle sensor to characterize the surface waterproof performance. Combined with the 70% remaining power output by the battery management system, a second dynamic calibration coefficient of 0.495 hours per millimeter is calculated. Multiplying the wind speed of 15 by 0.0533 gives 0.8, and multiplying the precipitation intensity of 25 by 0.495 gives 12.375. Adding them together gives a meteorological interference weighted value of 1 / 3.175 seconds. The reciprocal of the result yields an anti-interference coefficient of 0.076, which is lower than the preset safety threshold of 0.1. This triggers the flight altitude to drop to 70% of the baseline altitude of 500 meters, i.e., 350 meters, effectively avoiding the risk of loss of control caused by strong winds and rain.

[0036] As a feasible implementation, in mountainous terrain mapping for flood disaster assessment, a UAV encountered sudden weather changes. Initially, the wind speed was 8 meters per second and the precipitation intensity was 10 millimeters per hour. Based on a flight altitude of 400 meters and a maximum wind resistance level of 20 meters per second, the system calculated a first dynamic calibration coefficient of 0.0425 seconds per meter, combined with an altitude attenuation function; the higher the altitude, the weaker the wind resistance. Simultaneously, based on a 50-degree hydrophobic angle (indicating good wing surface drainage performance) and 80% remaining power, a second dynamic calibration coefficient of 0.657 hours per millimeter was generated. At this point, the weighted value of the meteorological interference was the wind speed contribution of 0.34 (8 × 0.0425) plus the precipitation contribution of 6.57 (10 × 0.657), totaling 1 / 6.91 seconds, corresponding to an immunity coefficient of 0.145, higher than the safety threshold, allowing normal flight to continue. When a sudden weather change causes the wind speed to surge to 18 meters per second, the immunity factor is recalculated and reduced to 0.08. Within 100 milliseconds, the system lowers its altitude to 280 meters to avoid a crash caused by strong wind shear.

[0037] For example, in a high-altitude glacier topographic mapping mission, a UAV encountered freezing rain at an altitude of 2000 meters. Traditional methods, neglecting the hydrophobic angle characteristics, misclassified the freezing rain as ordinary precipitation, resulting in a misclassification rate of over 40%, leading to premature return and mission interruption. This solution measures the hydrophobic angle of the wing in real-time at -20°C, achieving an ice-water contact angle of 110 degrees. Combined with 80% remaining power, a second dynamic calibration coefficient of 0.812 hours per millimeter is calculated, significantly reducing the weight of freezing rain in the meteorological interference weighting value. Simultaneously, based on the thin air characteristics at high altitudes, the maximum wind resistance level is reduced to 15 meters per second, and a first calibration coefficient of 0.025 seconds per meter is dynamically generated. The final interference resistance coefficient of 0.12 is still higher than the safety threshold, allowing the UAV to continuously collect glacier ablation point cloud data, increasing the mission completion rate to 95%.

[0038] The aforementioned method for calculating the anti-disturbance coefficient based on wind speed, precipitation intensity, and UAV performance parameters, through dynamic compensation of the hydrophobic angle and adaptive altitude adjustment, improves the continuous execution rate of mapping tasks under the same meteorological conditions and also enhances the integrity of point cloud data. This performance parameter coupling mechanism addresses the core issue that fixed threshold control cannot adapt to complex meteorological conditions, supporting the beneficial effects of reducing equipment damage probability and ensuring data continuity.

[0039] In some embodiments, step S103, which involves starting or stopping the multispectral camera based on the comparison result between the cloud optical thickness and the cloud thickness threshold, includes: comparing the cloud optical thickness with the cloud thickness threshold; starting the multispectral camera in response to the cloud optical thickness not exceeding the cloud thickness threshold; and stopping the multispectral camera in response to the cloud optical thickness exceeding the cloud thickness threshold.

[0040] In some embodiments, this application provides a specific implementation process for the multispectral camera control mechanism of UAVs in meteorologically sensitive terrain mapping tasks. This embodiment is based on real-time analyzed cloud optical thickness parameters. In one specific embodiment, cloud optical thickness is defined as the degree of attenuation of vertical light radiation by clouds at the 550nm band, and adaptive start / stop decisions for the optical equipment are executed. Specifically, the steps include: comparing the real-time acquired cloud optical thickness with a preset cloud thickness threshold, where a typical value for the cloud thickness threshold is, for example, 0.3; when the cloud optical thickness does not exceed the threshold of 0.3, it is determined that the current cloud transmittance meets the optical imaging conditions, and the multispectral camera is activated to collect surface spectral data of the mapping area; when the cloud optical thickness exceeds the threshold of 0.3, it is determined that cloud obstruction causes the effective spectral signal attenuation to exceed 70%, and the multispectral camera is shut down to avoid invalid data collection.

[0041] As a feasible implementation, in the mapping task of the outer circulation area of ​​a typhoon: the real-time monitoring of the cloud optical thickness is 0.25, which is lower than the threshold of 0.3, and the system automatically starts the multispectral camera; the camera collects surface spectral data of the flooded area in the visible-near infrared band (400nm-2500nm), and focuses on extracting the 1450nm and 1900nm moisture characteristic absorption bands for soil moisture content inversion; the collected data is transmitted to the edge in real time to participate in the subsequent enhanced surface model fusion process.

[0042] As a specific implementation, in a strong convective cloud mapping scenario in the eyewall region of a typhoon: the optical thickness of the cloud layer suddenly increases to 0.8, far exceeding the threshold of 0.3, and the system immediately shuts down the multispectral camera; to avoid strong scattering interference of cumulonimbus clouds on optical signals, the cloud top reflectivity is >80dBz, reducing the amount of invalid data storage by 65%; and to simultaneously switch the polarization sensor to the infrared band, such as 1550nm, to penetrate the cloud layer and collect topographic polarization data, maintaining basic mapping capabilities.

[0043] For example, in mountainous terrain mapping within the meteorological service field: when the cloud optical thickness is 0.1, indicating clear weather, the system continuously collects multispectral data and generates a mineral and lithology classification grid; when the cloud optical thickness rises to 0.4, indicating sudden stratocumulus clouds, the system shuts down the camera within 20ms and records the event log "Cloud optical thickness exceeds threshold 0.3, multispectral camera has been shut down. Polarization sensor infrared band compensation enabled"; this mechanism reduces the power consumption of the UAV during sudden weather changes while ensuring the availability of power for the lidar when visibility deteriorates.

[0044] The aforementioned optical thickness threshold control avoids distortion of multispectral data under thick cloud layers, such as cloud shadows being misidentified as surface shadows. Traditional topographic mapping performs fixed spectral acquisition, and when the cloud optical thickness is >0.3, data availability is less than 30%. This embodiment uses a dynamic start-stop mechanism to ensure that surface spectral data is collected only when meteorological conditions permit, optimizing system resource utilization.

[0045] In some embodiments, step S103, controlling the polarization sensor to collect polarization data of the survey area in the corresponding working band, includes: obtaining a comparison result between the cloud optical thickness and the cloud thickness threshold; in response to the cloud optical thickness not exceeding the cloud thickness threshold, controlling the deflection sensor to collect polarization data of the survey area in the visible light band; and in response to the cloud optical thickness exceeding the cloud thickness threshold, controlling the deflection sensor to collect polarization data of the survey area in the infrared band.

[0046] As a feasible implementation, an optical thickness threshold of 0.3 is set. In the topographic mapping task for typhoon disaster assessment, if the cloud optical thickness is 0.2, it is determined to be a thin cloud scenario. In this case, a visible light polarization sensor is activated to collect surface polarization angle data of the flooded area with high spatial resolution, such as 0.5m, for inverting micro-topographic changes at water body boundaries. If the cloud optical thickness is 0.6, it is determined to be a thick cloud scenario. In this case, the system automatically switches to the infrared band to penetrate the cumulonimbus cloud layer and collect topographic polarization information, minimizing signal attenuation caused by atmospheric scattering, with a penetration efficiency >80%. This dual-band mechanism ensures the continuous acquisition of effective polarization data during meteorological changes, supporting the construction of basic surface models.

[0047] As a specific implementation, combined with the aforementioned multispectral camera collaborative control: when the cloud optical thickness is 0.25 and does not exceed the cloud thickness threshold of 0.3, the multispectral camera is activated to collect surface spectral data; simultaneously, the visible light polarization sensor is activated, and the two work together to verify the consistency of surface reflectivity. When the cloud optical thickness is 0.5 and exceeds the cloud thickness threshold of 0.3, the multispectral camera is deactivated to avoid invalid data acquisition; the polarization sensor is switched to the infrared band, penetrating the cloud layer through a 1550nm long wavelength: polarization angle data - soil dielectric constant inversion - compensation for lidar elevation moisture difference error. In flood monitoring, this scheme increases the usability of polarization data from 62% of traditional methods to 95%.

[0048] For example, in mountainous terrain mapping, a sudden stratocumulus cloud event occurs: initially, the cloud optical thickness is 0.15 nm, and the polarization sensor is operating in the visible light band (532 nm). After a sudden weather change, the cloud thickness increases to 0.45 nm, exceeding the optical thickness threshold of 0.3 nm. Within 50 ms, the system executes the following: shuts down the multispectral camera; switches the polarization sensor to the infrared band (1550 nm); and records the event log "Cloud optical thickness exceeds limit, polarization switched to IR band." The infrared polarization data successfully penetrates the cloud layer with an attenuation rate of <15%, generating a terrain normal field to participate in edge fusion, ensuring the continuity of basic modeling.

[0049] The above-mentioned method for collecting polarization data in the survey area has the following advantages: the infrared band has a transmittance 3-5 times higher than that of visible light when the optical thickness of the cloud layer is greater than the cloud thickness threshold, which solves the problem of missing polarization data under dense clouds; the visible light band provides sub-meter level terrain details in clear skies, supporting high-precision disaster assessment; the band switching power consumption is low, which is more energy-efficient than the continuous operation of the dual-band scheme, and the overall method breaks through the bottleneck of lack of online coordination between meteorological elements and equipment collection.

[0050] In some embodiments, after waking up the standby lidar in step S104 to perform terrain scanning on the survey area, the method further includes: real-time monitoring of the rate of change of the anti-interference coefficient; if the rate of change exceeds a preset deterioration threshold, stopping the current terrain scanning and using the current flight altitude as a new reference altitude to adjust the flight altitude to a preset safety ratio value of the new reference altitude; encrypting and storing the acquired point cloud data, polarization data, and surface spectral data to obtain an encrypted data packet; starting the millimeter-wave radar to perform a wide-area scan on the survey area to generate a terrain overview map with reduced resolution; and generating a surveying task interruption report based on the terrain overview map and the encrypted data packet.

[0051] In some embodiments, the meteorological deterioration emergency response mechanism of this application is implemented in a coastline topographic mapping mission during a strong typhoon. When the system wakes up the lidar to scan the dike due to a sudden drop in visibility to 0.5 km, the airborne meteorological processor monitors in real time that the rate of change of the anti-interference coefficient is rapidly deteriorating. Specifically, the anti-interference coefficient drops linearly from 0.18 to 0.09 within 3 seconds, with a rate of change of -0.03 per second, far exceeding the preset deterioration threshold of -0.02 per second. The system immediately stops the lidar scanning of the core risk area, sets the current flight altitude of 280 meters as the new reference altitude, and controls the UAV to descend to a preset safety ratio value of 0.7 times that reference altitude, i.e., 196 meters. At the same time, the acquired 1.2 square kilometers of high-precision point cloud data, which includes the characteristics of seawall structural cracks, visible light polarization data, records of abnormal surface seepage, and surface spectral data collected before the rainstorm, is encrypted and packaged for military-grade storage. Simultaneously, millimeter-wave radar was activated to perform a wide-area scan of the disaster area, generating a topographic overview map covering 15 kilometers of coastline, with a resolution reduced to 10 meters. Finally, a mapping interruption report was automatically generated. This report combined the spatial index of the encrypted data packet with the floodplain markers on the overview map, noting "Typhoon eyewall strikes suddenly, core area completion rate 65%, it is recommended to restart the scan at a reference height of 196 meters."

[0052] As a feasible implementation, an extreme weather event occurred during a geological disaster monitoring mission in a mountainous area. After the UAV responded to rainfall intensity exceeding a threshold and initiated lidar scanning of the landslide area, it suddenly detected that the anti-disturbance coefficient was continuously deteriorating at a rate of 0.04 per second (assuming a preset threshold of 0.03 per second). Within 200 milliseconds, the system executed a fully automated emergency procedure: immediately terminating the ongoing cliff area scan, establishing a new safe reference altitude of 320 meters, and rapidly descending to 224 meters to avoid strong turbulence. The collected 0.8 square kilometer rock layer point cloud data and infrared polarization data, which recorded the rock mass displacement vector, as well as the mineral spectral data acquired before the interruption, were encrypted using national cryptographic algorithms and stored in a shock-resistant memory. Simultaneously, the millimeter-wave radar initiated a wide-area scanning mode, generating a 20 square kilometer mountain stability overview map with a resolution of 15 meters, clearly showing the distribution of newly formed crack zones. The automatically generated interruption report included key alarms: "Anti-disturbance coefficient deterioration rate exceeds limit; data for dangerous areas has been saved; wide-area scanning identified 3 high-risk landslide precursors."

[0053] The aforementioned emergency mechanism helps overcome the systemic shortcomings of traditional topographic mapping in the event of sudden weather changes. This embodiment completes altitude reduction decisions in a very short time, ensuring equipment safety and significantly reducing the accident rate. This mechanism enabled the mapping mission to recover at least 90% of the core data even after three weather interruptions, a significant reduction in data loss compared to traditional methods. Simultaneously, the design of dynamically updating the reference altitude in real-time avoids redundant scanning or missing critical data due to fixed thresholds, ensuring that the optimal detection altitude is maintained under complex weather conditions.

[0054] In some embodiments, setting the scan density in step S104 to be positively correlated with precipitation intensity and negatively correlated with visibility includes: calculating an altitude attenuation factor based on real-time flight altitude; calculating surface reflectance based on polarization data or surface spectral data, and calculating a reflection gain factor based on surface reflectance; multiplying a preset baseline density by the product of the altitude attenuation factor and the reflection gain factor to obtain a scan density baseline value; and multiplying the scan density baseline value by a first gain coefficient as precipitation intensity increases and by a second gain coefficient as visibility decreases; wherein the first gain coefficient is determined based on the remaining power of the UAV's power system, and the second gain coefficient is determined based on the cloud optical thickness.

[0055] In some embodiments, the dynamic adjustment method for lidar scanning density of this application is applied to coastal dike mapping tasks during typhoon passage. In one specific embodiment, a visibility threshold of 1 km is set. When the system responds to a visibility drop to 0.8 km, which is below the visibility threshold, the lidar is activated. Then, an altitude attenuation factor is calculated based on the UAV's current flight altitude of 350 meters and a reference altitude of 500 meters. The altitude attenuation factor characterizes the compensation requirement for point cloud resolution due to the decrease in flight altitude, specifically calculated as 1.48 by (500 / 350) raised to the power of 1.3. Simultaneously, based on the dike surface data collected by the polarization sensor in the visible light band, the surface reflectance is calculated to be 0.25 (clay material). Based on this, a reflection gain factor of 0.8 × 0.25 + 0.6 = 0.8 is calculated using a linear mapping formula. Multiplying the preset reference density of 500 points per square meter by the product of the altitude attenuation factor 1.48 and the reflection gain factor 0.8 yields a base scanning density of 592 points per square meter. In one specific embodiment, a precipitation threshold of 30 mm / h was set. When precipitation intensity reached 40 mm / h, exceeding the threshold, the remaining power of the power system was 60%, resulting in a first gain coefficient of 1.56. A cloud optical thickness of 0.5 corresponds to a visibility of 0.8 km, leading to a second gain coefficient of 2.12. The final scan density was 592 × 1.56 × 2.12 ≈ 1960 points per square meter, a 292% increase over the baseline density to penetrate heavy rain conditions. This triple adjustment mechanism ensures that the lidar can automatically optimize scan intensity under complex weather conditions such as heavy rain and sea fog.

[0056] As a feasible implementation, in the topographic mapping of farmland in flood-stricken areas, a UAV triggers a lidar scan due to rainfall intensity of 35 mm per hour. After the real-time flight altitude decreases from 400 meters to 280 meters, the altitude attenuation factor is calculated as (400 / 280)^1.3 ≈ 1.57. Since the cloud optical thickness of 0.4 causes the multispectral camera to shut down, the system calls polarization data to calculate the surface reflectance: first, the polarization rate is analyzed to be 0.35, which is considered valid as it exceeds the threshold of 0.3. Then, the reflectance of the paddy field is obtained by inverting the polarization angle in the infrared band to 0.15, based on which a reflection gain factor of 0.8 × 0.15 + 0.6 = 0.72 is generated. The basic scan density is 500 × 1.57 × 0.72 ≈ 565 points per square meter. Combining a precipitation intensity of 40 mm / h (with 50% of the remaining power generating a first gain coefficient of 1.44) and a visibility of 0.6 km (with a cloud optical thickness of 0.4 generating a second gain coefficient of 2.0), the final scan density was adjusted to approximately 565 × 1.44 × 2.0 ≈ 1627 points per square meter. This configuration increased the micro-topographic capture rate of flooded farmland ridges from 45% under the traditional fixed density to 92%. The scan density baseline responded to changes in rainfall intensity via the first gain coefficient, while the second gain coefficient dynamically adapted to visibility attenuation caused by dense fog, ensuring that the final scan density reached the optimal configuration under the constraint of the remaining power of the power system.

[0057] The aforementioned mechanism, which sets scanning density to be positively correlated with precipitation intensity and negatively correlated with visibility, effectively overcomes the impact of changes in flight altitude and differences in surface material on scanning quality through the synergistic compensation of altitude attenuation factor and reflection gain factor. It utilizes the first gain coefficient to achieve a positive response to precipitation intensity, ensuring sufficient and effective point cloud data is acquired under heavy rain conditions. By leveraging the second gain coefficient to inversely compensate for visibility attenuation, it overcomes the physical limitations of dense fog / clouds on laser scanning. It significantly enhances the ability to capture terrain details under complex meteorological conditions, making it particularly suitable for critical scenarios such as meteorologically sensitive geological disaster early warning and flood assessment.

[0058] In one specific embodiment, the altitude attenuation factor is a correction coefficient used for lidar scanning density calculation, characterizing the impact of flight altitude on the resolution of point cloud data acquisition. The altitude attenuation factor value is negatively correlated with the real-time flight altitude, specifically obtained through altitude-resolution calibration curve mapping. The altitude attenuation factor is a dimensionless correction coefficient, calculated using the following formula:

[0059]

[0060] Where α is the altitude attenuation factor, which is dimensionless and characterizes the degree to which changes in flight altitude reduce the resolution of the lidar scan. The reference altitude, in meters (m), represents the initial cruising altitude of the UAV during its mapping mission. This is the real-time flight altitude, in meters (m), representing the current adjusted actual flight altitude. n The calibration index is typically set to 1.2-1.5 and is determined by fitting the lidar performance calibration curve. It reflects the nonlinear effect of altitude on point cloud density.

[0061] In one specific embodiment, the reflection gain factor is a scan density correction coefficient calculated based on surface reflectivity, used to compensate for the influence of surface reflection characteristics on laser echo intensity. The reflection gain factor is a dimensionless correction coefficient, positively correlated with surface reflectivity, and obtained through linear mapping or logarithmic transformation. The linear mapping formula is as follows:

[0062]

[0063] in, β It is a dimensionless reflection gain factor that compensates for the gain effect of surface reflection characteristics on laser echo intensity. R Surface reflectance is dimensionless, ranging from 0 to 1, and is measured directly through polarization data inversion or by a multispectral camera. k This is a proportionality coefficient, dimensionless, determined by the lidar receiver sensitivity calibration, and determines the strength of the effect of reflectivity on gain. b This is an offset constant, dimensionless, ensuring the gain factor is greater than 1 for minimum reflectivity. Surface reflectivity is obtained through polarization data inversion or directly from a multispectral camera; the higher the reflectivity, the larger the reflectivity gain factor.

[0064] In one specific embodiment, the first gain coefficient is a dynamic adjustment weight of precipitation intensity on lidar scanning density, and its value is positively correlated with the remaining power of the UAV's power system. The first gain coefficient is used to convert the attenuation effect of precipitation intensity (in mm / h) on point cloud data quality into a compensating gain for scanning density.

[0065] The formula for calculating the first gain coefficient is as follows:

[0066]

[0067] in, k 1 represents the first gain coefficient. The current remaining power of the power system is expressed as a percentage, ranging from 0% to 100%. The nominal maximum power of the power system is expressed as a percentage, and is fixed at 100%. Gain boundary values ​​calibrated for engineering purposes, typical values ​​are as follows: .

[0068] In one specific embodiment, the second gain coefficient is a nonlinear compensation factor for visibility versus lidar scan density, the value of which is directly determined by the cloud optical thickness τ. This coefficient quantifies the degree to which atmospheric extinction (the main cause of reduced visibility) hinders laser transmission.

[0069] In one specific embodiment, the formula for calculating the second gain coefficient is as follows:

[0070]

[0071] in, k 2 represents the second gain coefficient, and τ represents the optical thickness of the cloud layer, which is dimensionless. This is the critical optical thickness threshold, typically 0.4. Beyond this value, the atmospheric extinction effect increases exponentially.

[0072] In one specific embodiment, the final scan density is calculated using the following formula:

[0073]

[0074] in, D This indicates the final scan density, expressed in pts / m². The preset reference density is expressed in pts / m², and β is the reflection gain factor. k 1 represents the first gain coefficient. k 2 represents the second gain coefficient.

[0075] In some embodiments, step S104, calculating surface reflectance based on polarization data or surface spectral data, includes: determining whether a multispectral camera has been activated; in response to the multispectral camera being activated, extracting surface reflectance from the surface spectral data acquired by the multispectral camera; in response to the multispectral camera not being activated, calculating the polarization rate corresponding to the polarization data, and determining whether the polarization data is valid based on the comparison result between the polarization rate and a preset polarization rate threshold; in response to the polarization rate being less than the polarization rate threshold, confirming that the polarization data is invalid, then calling a pre-stored surface reflectance reference map for interpolation to obtain the surface reflectance; in response to the polarization rate being not less than the polarization rate threshold, confirming that the polarization data is valid, then retrieving the surface reflectance based on the polarization data.

[0076] In some embodiments, the surface reflectance acquisition method of this application is implemented in flood assessment tasks after a typhoon. When the UAV flies to the disaster area, the system first determines whether the multispectral camera has been activated. At this time, the cloud optical thickness is 0.25, which does not exceed the cloud thickness threshold of 0.3, so the multispectral camera is in working condition. The system directly extracts the reflectance value of 0.18 in the 1450nm band from the surface spectral data collected by the camera. This parameter accurately reflects the soil saturated water content characteristics and provides a key input for subsequent scanning density calculation.

[0077] As a feasible implementation, in a mapping scenario involving sudden dense fog in mountainous areas, the cloud optical thickness increases to 0.45, causing the multispectral camera to shut down. The system then analyzes the data collected by the polarization sensor: first, it calculates the polarization rate (defined as the ratio of polarized signal intensity to total light intensity) to be 0.35, which is higher than the preset polarization rate threshold of 0.3, thus determining the polarization data to be valid. Subsequently, based on the 1550nm infrared band polarization angle data, the Fresnel reflection model is used to inversely derive the granite surface reflectance of 0.32, supporting the high-precision scanning of rock fissures by the lidar.

[0078] For example, during levee monitoring in the eyewall area of ​​a coastal typhoon, extreme weather conditions were encountered: a cloud optical thickness of 0.6 forced the multispectral camera to shut down, and heavy rainfall caused severe attenuation of the polarization signal. The system detected a polarization rate of only 0.15, below the threshold of 0.3, and determined that the current polarization data was invalid. The system immediately retrieved a pre-stored regional surface reflectance benchmark map, generated from a historical survey task, with a resolution of 10m × 10m. Bilinear interpolation was used to obtain the reflectance of the concrete levee at the current location as 0.41, ensuring that the scan density calculation was not interrupted by meteorological factors.

[0079] Understandably, a polarization rate threshold of 0.3 is the critical threshold for atmospheric scattering noise; values ​​below this threshold indicate that aerosols / rain / fog are causing polarization signal distortion. Pre-stored baseline maps are established through multispectral scanning of historical missions and stored in a reflectivity feature library categorized by soil type / surface material.

[0080] The aforementioned method for calculating surface reflectance based on a triple protection mechanism prioritizes the use of multispectral cameras to acquire accurate reflectance under favorable optical conditions, supporting detailed analyses such as soil moisture content inversion; when cloud cover thickens, surface characteristics are inverted using effective polarization data to maintain mapping continuity under complex weather conditions; and when polarization signals are interfered with, pre-stored reference maps are intelligently invoked to avoid scanning control failures caused by missing reflectance. This triple protection mechanism significantly improves the reliability of terrain data in meteorologically sensitive scenarios such as geological disaster early warning and flood assessment.

[0081] In some embodiments, step S105, fusing point cloud data and polarization data at the edge, includes: constructing an atmospheric transmittance tensor based on precipitation intensity and cloud optical thickness; separating the true polarization signal of the ground surface from atmospheric scattering noise using a tensor decomposition algorithm based on the atmospheric transmittance tensor to correct atmospheric polarization distortion of the polarization data; converting the corrected polarization signal into a terrain normal vector field, and spatially fusing it with the point cloud data using a non-rigid registration algorithm.

[0082] The atmospheric transmittance tensor is a three-dimensional matrix describing the atmospheric transmission characteristics of light waves with different polarization states. Its construction depends on precipitation intensity and cloud optical thickness parameters. Tensor elements represent the atmospheric transmittance of electromagnetic waves in a specific direction and polarization state, used to quantify the distortion effect of atmospheric scattering on polarization signals. Non-rigid registration algorithms refer to elastic matching algorithms suitable for spatial fusion of terrain normal fields and point cloud data. Typical implementations include ThinPlate Splines or Bézier Deformation Networks, which can handle local data deformation caused by meteorological interference.

[0083] In some embodiments, the meteorological correction data fusion method of this application is implemented in a coastal topography modeling task after a typhoon disaster. After the UAV completes the levee scanning under heavy precipitation conditions, the edge processor constructs an atmospheric transmittance tensor based on the real-time precipitation intensity of 45 mm / h and the cloud optical thickness of 0.5. This tensor is a three-dimensional matrix structure, with its three dimensions representing the transmission characteristics of different polarization states (0°, 45°, 90°), the working wavelength (visible / infrared), and the spatial azimuth angle, respectively. The noise components interfered by rain and fog scattering are separated by the CP tensor decomposition algorithm (CANDECOMP / PARAFAC Decomposition), and atmospheric polarization distortion correction is performed on the raw data collected by the polarization sensor. The corrected polarization signal is converted into a topographic normal vector field describing the surface undulation characteristics, and non-rigid spatial fusion is performed with the point cloud data obtained by lidar through a thin-plate spline transformation algorithm to generate a basic surface model resistant to meteorological interference.

[0084] As a feasible implementation, in the topographic reconstruction of flood-inundated areas, the system drives the fusion process through real-time meteorological parameters. Edge devices construct an atmospheric transmittance tensor containing dual-band (532nm visible light and 1550nm infrared) transmission characteristics based on a precipitation intensity of 30 mm / h and a cloud optical thickness of 0.4. The Tucker decomposition algorithm is used to remove signal distortion caused by multiple scattering of raindrops from the polarization data, restoring the true polarization angle distribution of the water body boundary. The corrected polarization field is converted into a high-precision topographic normal vector, which is spatially registered with point cloud data representing elevation information through a Bezier deformation network. This successfully reconstructs the 0.3-meter-level micro-topography of field ridges in flooded farmland areas, a feature previously overlooked by traditional methods.

[0085] For example, edge-end fusion exhibits strong robustness when monitoring geological hazards in mountainous areas encounters persistent dense fog. An atmospheric transmittance tensor, constructed based on a precipitation intensity of 15 mm / h and a cloud optical thickness of 0.7, quantifies the attenuation differences of polarization signals caused by fog droplets from different directions. By using non-negative matrix factorization to separate 30% of atmospheric scattering noise, the quality of polarization data in rock fracture areas is significantly improved. The corrected terrain normal field and laser point cloud are non-rigidly registered using feature constraints, generating centimeter-level accuracy vector maps of unstable rock mass displacement even under visibility conditions of less than 500 meters.

[0086] Understandably, the thin-plate spline transformation compensates for UAV attitude drift caused by strong winds by elastically matching surfaces with control points; the Bezier deformation network, on the other hand, uses parametric surface fitting, making it particularly suitable for terrain reconstruction in vegetated areas. In the atmospheric transmittance tensor, precipitation intensity mainly affects the diagonal element, i.e., forward transmittance, while cloud optical thickness dominates the off-diagonal element, i.e., multipath scattering.

[0087] The aforementioned method for fusing point cloud data and polarization data utilizes atmospheric transmittance tensors to transform complex meteorological parameters into a computable model, overcoming the limitations of traditional linear correction. Tensor decomposition algorithms effectively separate meteorological noise from surface signals, solving the polarization distortion problem under heavy precipitation / dense fog. Non-rigid registration technology adaptively compensates for spatial deviations caused by equipment jitter and meteorological disturbances, improving fusion accuracy in complex environments. The entire process is processed in real time at the edge, meeting the stringent timeliness requirements of scenarios such as geological disaster early warning and flood assessment.

[0088] In some embodiments, step S105, if surface spectral data exists, fuses it with the basic surface model to generate an enhanced surface model, including: if surface spectral data exists, extracting the absorption depth of the water characteristic absorption band in the surface spectral data; based on the absorption depth and the pre-acquired soil type parameters of the survey area, calculating the elevation correction amount according to the mapping relationship between soil dielectric properties and water content, and performing dielectric correction on the elevation field of the basic surface model using the elevation correction amount to generate a humidity-compensated elevation field; if the working band of the multispectral camera covers the pre-calibrated mineral characteristic response band, extracting the mineral characteristic reflectance and matching it with a preset mineral spectral library to generate a lithology classification raster, associating the lithology classification raster as an independent geographic layer with the spatial coordinates of the humidity-compensated elevation model, and generating an enhanced surface model containing a dielectric-corrected elevation field and an optional lithology layer.

[0089] Dielectric correction refers to a method for compensating for humidity in lidar elevation data based on the dielectric properties of soil. By mapping the absorption depth of the water-characteristic absorption bands in the surface spectral data to a dielectric constant-water content relationship model combined with soil type parameters, the elevation correction amount is calculated to eliminate the error in topographic elevation measurement caused by soil moisture. The lithological classification raster is a rasterized lithological distribution map generated by matching mineral characteristic reflectance with a preset mineral spectral library. Its pixel value represents a lithological classification code, such as granite code 1 and sandstone code 2. The spatial resolution is consistent with the data acquired by the multispectral camera, and it is spatially correlated with the elevation model as an independent geographic layer. Polarization rate is a core indicator for evaluating the validity of polarization data, defined as the ratio of polarized signal intensity to total light intensity. When the polarization rate is lower than a preset polarization rate threshold (usually set at 0.2-0.3), it indicates that atmospheric scattering noise has severely contaminated the polarization signal.

[0090] In some embodiments, the enhanced surface model construction method of this application is applied to the assessment of coastal dike repair after a typhoon. When the UAV completes mapping under a cloud optical thickness of 0.25, the surface spectral data collected by the multispectral camera is used for enhanced fusion. The edge processor first extracts the absorption depth of the 1450nm and 1900nm moisture characteristic absorption bands, which is defined as the normalized difference between the reference reflectance and the absorption valley reflectance. Combined with pre-loaded soil type parameters of the mapping area, such as 70% clay and 30% sand, the elevation correction is calculated through the soil dielectric constant-water content mapping curve. This correction is used to correct the measurement deviation caused by soil saturation in the lidar base elevation field, generating a humidity-compensated elevation field. At the same time, since the multispectral camera covers the 2100nm mineral characteristic response band, the system extracts the granite characteristic reflectance and matches it with the preset spectral library to generate a lithological classification raster map with a resolution of 0.5 meters. Finally, it is associated with the spatial coordinates of the humidity-compensated elevation model to form an enhanced surface model.

[0091] As a feasible implementation, in the topographic modeling of farmland in flood-inundated areas, the soil moisture content was inferred to be 42% from the 1450nm absorption depth of the surface spectral data. Based on the dielectric properties of loam in the pre-stored soil database, with the real part of the dielectric constant ε'=18.2, an elevation correction of 0.28 meters was calculated. Dielectric correction was applied to the paddy field area in the basic model to eliminate the artificial elevation caused by water infiltration. A multispectral camera detected mineral reflection characteristics in the 2200nm band, which were identified as kaolinite veins through spectral library matching. The generated lithology classification raster was superimposed as an independent layer onto the corrected elevation field, providing geological basis for post-disaster farmland restoration planning.

[0092] For example, during emergency mapping of geological disasters in mountainous areas, a brief window of clear skies was encountered: when the cloud optical thickness was 0.28, a multispectral camera was activated to acquire the rock layer spectrum. After extracting the 1900nm absorption depth to reflect surface humidity, and combining this with bedrock type parameters (granite accounting for 85%), an elevation correction of 0.15 meters was calculated to eliminate dielectric errors caused by previous rainfall. Chlorite characteristic peaks were identified in the 1650-2400nm shortwave infrared band, and a lithology classification raster was generated by matching it with a mineral spectral library. This raster was then superimposed onto a humidity-compensated elevation field to form an enhanced model. This model successfully identified altered rock zones neglected by traditional topographic data, providing crucial evidence for topographic mapping of landslide risk.

[0093] It is understandable that the aforementioned water characteristic absorption bands specifically refer to the water molecule vibration absorption peaks such as 1450nm and 1900nm; dielectric correction is based on the elevation error compensation caused by the change in the propagation speed of electromagnetic waves in moist soil; the lithological classification grid uses pixel value to encode geological types, such as 0 water body / 1 granite / 2 sandstone, and maintains spatial consistency with the elevation model through georegistration.

[0094] The aforementioned enhanced surface model construction method eliminates radar altimetry errors caused by meteorological interference by dynamically calculating elevation corrections based on water absorption depth and soil parameters; mineral feature band analysis generates independent geographic layers, achieving spatial coupling between the terrain model and geological attributes; a humidity-compensated elevation field ensures the accuracy of the basic terrain, and optional lithology layers expand the dimensions of disaster assessment; it significantly improves the surface modeling depth for meteorological disaster scenarios such as typhoons and floods, supporting geological risk early warning and post-disaster reconstruction planning.

[0095] This application also provides system embodiments that follow the above embodiments, for implementing the method steps of the above embodiments. The interpretation of the same names is the same as that of the above embodiments, and they have the same technical effects as those of the above embodiments, so they will not be repeated here.

[0096] like Figure 2 As shown, this application provides a topographic mapping system, including:

[0097] The data parsing unit 1001 is used to acquire meteorological data streams in real time and analyze the wind speed, precipitation intensity, cloud optical thickness and visibility of the survey area.

[0098] The processing unit 1002 is used to calculate the anti-interference coefficient based on wind speed, precipitation intensity, and the performance parameters of the UAV. If the anti-interference coefficient is lower than the safety threshold, the UAV's flight altitude is reduced to a preset safety ratio value of the reference altitude, and then the lidar is activated to put the UAV into standby mode. According to the comparison result of cloud optical thickness and cloud thickness threshold, the multispectral camera is activated or deactivated, and the polarization sensor is controlled to collect polarization data of the survey area in the corresponding working band. The multispectral camera is used to collect surface spectral data of the survey area. In response to the judgment that the visibility is lower than the visibility threshold or the precipitation intensity exceeds the precipitation threshold, the lidar in standby mode is woken up to perform terrain scanning of the survey area. The scanning density is set to be positively correlated with the precipitation intensity and negatively correlated with the visibility to generate point cloud data.

[0099] The surface model generation unit 1003 is used to fuse point cloud data and polarization data at the edge to generate a meteorological correction base surface model. If surface spectral data exists, it is fused with the base surface model to generate an enhanced surface model.

[0100] Regarding the system in the above embodiments, the specific ways in which each module performs operations have been described in detail in the embodiments related to the method, and will not be elaborated here.

[0101] Although the operations are described in a specific order in the accompanying drawings, this should not be construed as requiring these operations to be performed in the specific order or serial order shown, or requiring all of the operations shown to obtain the desired result. In certain environments, multitasking and parallel processing may be advantageous.

[0102] The methods and systems of this application can be implemented using standard programming techniques, and various method steps can be implemented using rule-based logic or other logic. It should also be noted that the terms "system" and "module" as used herein and in the claims are intended to include implementations using one or more lines of software code and / or hardware implementations and / or devices for receiving input.

[0103] Any step, operation, or procedure described herein may be performed or implemented using one or more hardware or software modules, either alone or in combination with other devices. In one embodiment, the software module is implemented using a computer program product comprising a computer-readable medium containing computer program code, which is executable by a computer processor to perform any or all of the described steps, operations, or procedures.

[0104] The foregoing description of implementations of this application has been provided for illustrative and descriptive purposes. The foregoing description is not exhaustive and is not intended to limit this application to the exact forms disclosed. Various modifications and variations may exist in accordance with the foregoing teachings, or may arise from practice of this application. These embodiments were chosen and described to illustrate the principles of this application and its practical application, enabling those skilled in the art to utilize this application in various implementations and modifications to suit the specific purpose of the concept.

[0105] It is further understood that although the operations are described in a specific order in the accompanying drawings in the embodiments of this application, this should not be construed as requiring these operations to be performed in the specific order or serial order shown, or requiring all the operations shown to obtain the desired result. In certain environments, multitasking and parallel processing may be advantageous.

[0106] Other embodiments of this application will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of this application that follow the general principles of this application and include common knowledge or customary techniques in the field of this application that are not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of this application are indicated by the following claims.

[0107] It should be understood that this application is not limited to the precise structure described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this application is limited only by the appended claims.

[0108] The above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims

1. A topographic mapping method, characterized in that, include: Real-time acquisition of meteorological data streams, and analysis of wind speed, precipitation intensity, cloud optical thickness, and visibility in the surveyed area; The anti-interference coefficient is calculated based on the wind speed, the precipitation intensity and the performance parameters of the UAV. If the anti-interference coefficient is lower than the safety threshold, the flight altitude of the UAV is reduced to a preset safety ratio value of the reference altitude, and then the lidar is activated to put the UAV into standby mode. Based on the comparison result between the cloud optical thickness and the cloud thickness threshold, the multispectral camera is started or stopped, and the polarization sensor is controlled to collect polarization data of the survey area in the corresponding working band. The multispectral camera is used to collect the surface spectral data of the survey area. In response to determining that the visibility is below the visibility threshold or the precipitation intensity exceeds the precipitation threshold, the standby lidar is activated to perform terrain scanning on the surveyed area. The scanning density is set to be positively correlated with the precipitation intensity. An altitude attenuation factor is calculated based on the real-time flight altitude. The surface reflectivity is calculated based on polarization data or surface spectral data, and a reflection gain factor is determined. The preset baseline density is multiplied by the product of the altitude attenuation factor and the reflection gain factor to obtain the scanning density base value. The scanning density base value is multiplied by a first gain factor as the precipitation intensity increases and by a second gain factor as the visibility decreases, thereby generating point cloud data. At the edge, an atmospheric transmittance tensor is constructed based on precipitation intensity and cloud optical thickness. The true polarization signal of the land surface and atmospheric scattering noise are separated by tensor decomposition algorithm. Atmospheric polarization distortion correction is performed on the polarization data. The corrected polarization signal is converted into a terrain normal vector field. It is then spatially fused with the point cloud data through a non-rigid registration algorithm to generate a basic surface model for meteorological correction. If the surface spectral data exists, the absorption depth of the water characteristic absorption band is extracted. Based on the absorption depth and the soil type parameters of the pre-acquired survey area, the elevation correction is calculated. Dielectric correction is performed on the basic surface model to generate a humidity-compensated elevation field. If the working band of the multispectral camera covers the pre-calibrated mineral characteristic response band, the mineral characteristic reflectance is extracted and matched with a preset mineral spectral library to generate a lithology classification raster. The lithology classification raster is associated as an independent geographic layer with the spatial coordinates of the humidity-compensated elevation field to generate an enhanced surface model that includes the dielectric-corrected elevation field and an optional lithology layer.

2. The method according to claim 1, characterized in that, The step of starting or stopping the multispectral camera based on the comparison result of the cloud optical thickness and the cloud thickness threshold includes: Compare the optical thickness of the cloud layer with the cloud thickness threshold; The multispectral camera is activated in response to the cloud optical thickness not exceeding the cloud thickness threshold. In response to the cloud optical thickness exceeding the cloud thickness threshold, the multispectral camera is turned off.

3. The method according to claim 1, characterized in that, The controlled polarization sensor acquires polarization data of the surveyed area under the corresponding operating wavelength band, including: Obtain the comparison result between the cloud optical thickness and the cloud thickness threshold; In response to the fact that the optical thickness of the cloud layer does not exceed the cloud layer thickness threshold, the deflection sensor is controlled to collect polarization data of the survey area in the visible light band. In response to the cloud optical thickness exceeding the cloud thickness threshold, the deflection sensor is controlled to collect polarization data of the survey area in the infrared band.

4. The method according to claim 1, characterized in that, The calculation of the anti-interference coefficient based on the wind speed, the precipitation intensity, and the performance parameters of the UAV includes: Calculate the product of real-time wind speed and the first dynamic calibration coefficient, and calculate the product of precipitation intensity and the second dynamic calibration coefficient. Add the two products together to obtain the meteorological disturbance weighted value. The anti-interference coefficient is obtained by taking the reciprocal of the weighted value of the meteorological interference. The first dynamic calibration coefficient is determined based on the maximum wind resistance level and real-time flight altitude of the UAV, while the second dynamic calibration coefficient is determined based on the measured hydrophobic angle of the UAV and the remaining power of the power system.

5. The method according to claim 1, characterized in that, After waking up the standby lidar to perform terrain scanning of the survey area, the method further includes: Real-time monitoring of the rate of change of the disturbance rejection coefficient; If the rate of change exceeds a preset deterioration threshold, the current terrain scan is stopped, and the current flight altitude is used as a new reference altitude to adjust the flight altitude to a preset safety ratio value of the new reference altitude. The acquired point cloud data, polarization data, and surface spectral data are encrypted and stored to obtain an encrypted data package; The millimeter-wave radar is activated to perform a wide-area scan of the surveyed area, generating a terrain overview map with reduced resolution. Based on the terrain overview map and the encrypted data packet, a surveying mission interruption report is generated.

6. The method according to claim 1, characterized in that, The calculation of surface reflectance based on the polarization data or the surface spectral data includes: Determine whether the multispectral camera has been started; In response to the activation of the multispectral camera, surface reflectance is extracted from the surface spectral data acquired by the multispectral camera; In response to the multispectral camera not being activated, the polarization rate corresponding to the polarization data is calculated, and the validity of the polarization data is determined based on the comparison result between the polarization rate and a preset polarization rate threshold. If the polarization rate is less than the polarization rate threshold and the polarization data is confirmed to be invalid, the surface reflectance is obtained by interpolation using a pre-stored surface reflectance reference map. In response to the polarization rate being not less than the polarization rate threshold, the polarization data is confirmed to be valid, and the surface reflectance is then inverted based on the polarization data.

7. A topographic mapping system, characterized in that, include: The data parsing unit is used to acquire meteorological data streams in real time and analyze wind speed, precipitation intensity, cloud optical thickness, and visibility in the surveyed area. The processing unit is used to calculate the anti-interference coefficient based on the wind speed, the precipitation intensity and the performance parameters of the UAV. If the anti-interference coefficient is lower than the safety threshold, the UAV's flight altitude is reduced to a preset safety ratio value of the reference altitude, and then the lidar is activated to put the UAV into standby mode. Based on the comparison between the cloud optical thickness and the cloud thickness threshold, the multispectral camera is activated or deactivated, and the polarization sensor is controlled to collect polarization data of the surveyed area in the corresponding working band. The multispectral camera is used to collect surface spectral data of the surveyed area. In response to the determination that the visibility is lower than the visibility threshold or the precipitation intensity exceeds the precipitation threshold, the standby lidar is activated to perform terrain scanning of the surveyed area. The scanning density is set to be positively correlated with the precipitation intensity. The altitude attenuation factor is calculated based on the real-time flight altitude. The surface reflectivity is calculated based on the polarization data or surface spectral data, and the reflection gain factor is determined. The preset reference density is multiplied by the product of the altitude attenuation factor and the reflection gain factor to obtain the scanning density base value. The scanning density base value is multiplied by a first gain coefficient as the precipitation intensity increases and by a second gain coefficient as the visibility decreases, thereby generating point cloud data. The surface model generation unit is used to construct an atmospheric transmittance tensor based on precipitation intensity and cloud optical thickness at the edge. It separates the true polarization signal from atmospheric scattering noise using a tensor decomposition algorithm, corrects atmospheric polarization distortion in the polarization data, converts the corrected polarization signal into a terrain normal field, and spatially fuses it with the point cloud data using a non-rigid registration algorithm to generate a meteorologically corrected basic surface model. If the surface spectral data exists, it extracts the absorption depth of the moisture characteristic absorption bands, calculates the elevation correction based on the absorption depth and pre-acquired soil type parameters of the survey area, and performs dielectric correction on the basic surface model to generate a humidity-compensated elevation field. If the working band of the multispectral camera covers the pre-calibrated mineral characteristic response bands, it extracts the mineral characteristic reflectance and matches it with a preset mineral spectral library to generate a lithology classification raster. The lithology classification raster is then associated as an independent geographic layer with the spatial coordinates of the humidity-compensated elevation field, generating an enhanced surface model containing the dielectrically corrected elevation field and an optional lithology layer.

Images