CN-122017176-A - High-precision collaborative emergency method and system for satellite-monitoring vehicle

CN122017176ACN 122017176 ACN122017176 ACN 122017176ACN-122017176-A

Abstract

The application discloses a high-precision collaborative emergency method and system for a satellite-monitoring vehicle. The method comprises the steps of constructing a water quality inversion empirical model based on historical data and monitoring and alarming in real time by a satellite remote sensing monitoring system, controlling an unmanned aerial vehicle to acquire hyperspectral images through an unmanned aerial vehicle sampling area and unmanned ship acquisition point accurate planning algorithm after the intelligent remote sensing monitoring vehicle receives alarming information, controlling the unmanned aerial vehicle to acquire water quality parameters in situ, constructing a ground hyperspectral inversion model based on actual measurement data by a data processing system, acquiring high-precision water quality parameter image layer data of the sampling area, carrying out on-site dynamic correction on the satellite inversion model by adopting a scale elimination algorithm and a weighted least square method, applying the corrected model to satellite data, inverting to obtain high-precision water quality parameters, generating a water quality report, and realizing on-site and command center real-time linkage through a decision command system. The application solves the problems of insufficient traditional monitoring precision and delayed response, and realizes the high-precision and high-efficiency coordination of water quality monitoring.

Inventors

Peng Guozhang
XIE WEIMIN
Tan Zhangru
TU ZIWEN
XU JIE
CHENG JINGHUA
QIN HE
JIN WENJIE
PAN LU
XU ZHENGUO
HUANG XUANMIN
WANG YINGCAI

Assignees

生态环境部长江流域生态环境监督管理局生态环境监测与科学研究中心
南水北调中线水源有限责任公司
武汉润江生态科技有限公司

Dates

Publication Date: 20260512
Application Date: 20260414

Claims (10)

1. The high-precision collaborative emergency system for the satellite-monitoring vehicle is characterized by comprising a central control platform integrated in the intelligent remote sensing monitoring vehicle, a satellite remote sensing monitoring system, an unmanned aerial vehicle and unmanned ship management and control system and a data processing system, wherein the satellite remote sensing monitoring system is in communication connection with the central control platform; the satellite remote sensing monitoring system is used for constructing a water quality parameter inversion empirical model based on historical ground measured water quality parameter data of a target water area, ground water body remote sensing reflectivity data and synchronous satellite image spectrum data, inverting the water quality parameter inversion empirical model to obtain a water quality inversion layer of the whole target water area, and generating and sending alarm information when the water quality parameter of a certain area in the water quality inversion layer reaches or exceeds a preset alarm threshold value; The central control platform is used for receiving the alarm information, determining an unmanned aerial vehicle monitoring area and unmanned ship acquisition points according to the alarm information and further generating a route path plan; The unmanned aerial vehicle and unmanned ship control system is used for controlling the unmanned aerial vehicle to collect hyperspectral image data in a sampling area according to the route planning command and transmitting the collected hyperspectral image data back to the data processing system in real time; The data processing system is used for receiving the hyperspectral image data and the actually measured water quality parameter data, constructing a ground hyperspectral inversion model, obtaining high-precision water quality parameter layer data of a sampling area, dynamically correcting the water quality parameter inversion empirical model based on the high-precision water quality parameter layer data and satellite spectrum data to generate a corrected satellite inversion model, applying the corrected satellite inversion model to satellite remote sensing data on the same day, inverting to obtain high-precision water quality parameter data, and carrying out comprehensive water quality analysis by combining the hyperspectral data of the unmanned aerial vehicle and the actually measured data of the unmanned aerial vehicle to generate a water quality report.
2. The satellite-monitor high-precision collaborative emergency system according to claim 1, wherein the central control platform determines an unmanned aerial vehicle monitoring area and an unmanned ship acquisition point according to the alert information, specifically comprising: S201, acquiring an alarm area A alert , an area B of a single shooting range of the unmanned aerial vehicle and a preset coverage proportion P; S202, calculating the alarm area A coverage =A alert x P to be covered according to the alarm area A alert and the preset coverage proportion P; S203, determining the number N=ceil (A coverage /B) of sampling areas according to the alarm area A coverage to be covered and the single shooting range area B of the unmanned aerial vehicle, wherein ceil represents upward rounding; s204, determining the side length of a single sampling area according to the single shooting range area B of the unmanned aerial vehicle ; S205, generating an outsourcing rectangle according to four-to-coordinate of the alarm area, and determining the width W and the height H of the outsourcing rectangle; s206, carrying out grid division on the outer package rectangle according to the side length S, and calculating the number n=W/S of the sampling areas which can be divided in each row and the number m=H/S of the sampling areas which can be divided in each column; s207, generating a set of potential sampling region positions C potential = { (j x S, i x S) i=0, 1, & m-1, j=0, 1, & n-1}, to ensure that the potential sampling regions are uniformly distributed in the outer packing rectangle; S208, calculating a uniform selection step= (n×m)/N according to the number N of sampling areas and the total number n×m of potential positions, and uniformly selecting indexes from the potential sampling area position sets according to the step length step to obtain an unmanned plane monitoring area set C drone = {C potential [ k ] |k=0, step, 2×step,. }; S209, for each selected sampling region, determining its central location as an unmanned ship acquisition point P boat ,k =(C drone,k,x +(s/2),C drone,k,y + (S/2)), where k=0, 1,..n-1.
3. The satellite-monitor vehicle high-precision collaborative emergency system according to claim 1, wherein the data processing system dynamically corrects the water quality parameter inversion empirical model based on the high-precision water quality parameter layer data and satellite spectrum data, and generates a corrected satellite inversion model, comprising: s601, resampling the high-precision water quality parameter layer data of the sampling area obtained in the step S5 to be consistent with the resolution of satellite data by adopting a scale elimination algorithm to obtain satellite scale water quality parameter data; s602, matching the satellite scale water quality parameter data with satellite spectrum data of a corresponding area to construct a correction data set; And S603, adopting a weighted least square method, taking the water quality parameter data of the satellite scale in the corrected data set as a real label, and carrying out parameter iterative optimization on the water quality parameter inversion empirical model constructed in the step S1 until a preset error threshold is met, so as to obtain a corrected satellite inversion model.
4. The satellite-monitor vehicle high precision collaborative emergency system according to claim 3, wherein the scale cancellation algorithm comprises: calculating a spatial weight factor based on Euclidean distance between sub-meter level pixels and satellite pixel centers Calculating a spectral similarity weight factor based on cosine similarity of sub-meter pixel hyperspectral reflectivity and satellite pixel average hyperspectral reflectivity ; By spatial weighting factors Weighting factors for spectral similarity Weighting and fusing after normalization to obtain comprehensive weight factors ; According to comprehensive weight factors Weighting and summing the water quality parameter values of the sub-meter level pixels to obtain the water quality parameter values of the satellite pixels 。
5. The satellite-monitor vehicle high precision collaborative emergency system according to claim 3, wherein the objective function of the weighted least squares method is: ; ; ; Wherein, the As a function of the object to be processed, To correct the total number of samples in the dataset; true tag value for satellite pixel Confidence weights of (2); mathematical expression of the satellite optimal inversion model; spectral feature vectors for the j-th satellite pixel; The real tag value of the satellite scale water quality parameter is obtained; Sub-meter-level pixel for unmanned aerial vehicle Parameters of water quality Is calculated based on unmanned ship in-situ measured data and unmanned aerial vehicle inversion data; Sub-meter-level pixel for unmanned aerial vehicle In-situ monitoring of water quality parameter values by unmanned ships with nearest spatial positions; Sub-meter-level pixel for unmanned aerial vehicle Is obtained by inversion of a ground hyperspectral inversion model.
6. A satellite-monitor vehicle high precision collaborative emergency method performed with a system according to any one of claims 1-5, the method comprising the steps of: S1, a satellite remote sensing monitoring system constructs a water quality parameter inversion empirical model based on historical ground measured water quality parameter data of a target water area, ground water body remote sensing reflectivity data and geostationary satellite image spectrum data, a water quality inversion image layer of the whole target water area is obtained through inversion of the water quality parameter inversion empirical model, and alarm information is generated and sent when water quality parameters of a certain area in the water quality inversion image layer reach or exceed a preset alarm threshold value; S2, the central control platform of the intelligent remote sensing monitoring vehicle receives the alarm information, determines an unmanned aerial vehicle monitoring area and unmanned ship acquisition points according to the alarm area range and the unmanned aerial vehicle single shooting range so as to generate a route planning, and sends the route planning to an unmanned aerial vehicle and an unmanned ship management and control system respectively; S3, controlling the unmanned aerial vehicle and the unmanned ship management and control system to acquire hyperspectral image data of a sampling area according to the route planning, transmitting the acquired hyperspectral image data back to the data processing system in real time, controlling the unmanned ship to reach a preset acquisition point for in-situ monitoring of water quality parameters, and transmitting actually measured water quality parameter data back to the data processing system in real time; S4, the data processing system receives the hyperspectral image data and the actually measured water quality parameter data, a ground hyperspectral inversion model is constructed, and high-precision water quality parameter image layer data of a sampling area are obtained; S5, the data processing system dynamically corrects the water quality parameter inversion empirical model constructed in the step S1 based on the sampling area high-precision water quality parameter map layer data and satellite spectrum data obtained in the step S4, and a corrected satellite inversion model is generated; s6, applying the corrected satellite inversion model to satellite remote sensing data on the same day, inverting to obtain high-precision water quality parameter data, and carrying out comprehensive water quality analysis by combining unmanned aerial vehicle hyperspectral data and unmanned ship measured data to generate a water quality report.
7. The method according to claim 6, wherein in step S2, the determining the unmanned aerial vehicle monitoring area and the unmanned ship acquisition point according to the alarm information specifically includes: S201, acquiring an alarm area A alert , an area B of a single shooting range of the unmanned aerial vehicle and a preset coverage proportion P; S202, calculating the alarm area A coverage =A alert x P to be covered according to the alarm area A alert and the preset coverage proportion P; S203, determining the number N=ceil (A coverage /B) of sampling areas according to the alarm area A coverage to be covered and the single shooting range area B of the unmanned aerial vehicle, wherein ceil represents upward rounding; s204, determining the side length of a single sampling area according to the single shooting range area B of the unmanned aerial vehicle ; S205, generating an outsourcing rectangle according to four-to-coordinate of the alarm area, and determining the width W and the height H of the outsourcing rectangle; s206, carrying out grid division on the outer package rectangle according to the side length S, and calculating the number n=W/S of the sampling areas which can be divided in each row and the number m=H/S of the sampling areas which can be divided in each column; s207, generating a set of potential sampling region positions C potential = { (j x S, i x S) i=0, 1, & m-1, j=0, 1, & n-1}, to ensure that the potential sampling regions are uniformly distributed in the outer packing rectangle; S208, calculating a uniform selection step= (n×m)/N according to the number N of sampling areas and the total number n×m of potential positions, and uniformly selecting indexes from the potential sampling area position sets according to the step length step to obtain an unmanned plane monitoring area set C drone = {C potential [ k ] |k=0, step, 2×step,. }; S209, for each selected sampling region, determining its central location as an unmanned ship acquisition point P boat ,k =(C drone,k,x +(s/2),C drone,k,y + (S/2)), where k=0, 1,..n-1.
8. The satellite-monitor vehicle high-precision collaborative emergency method according to claim 7, wherein step S6 specifically comprises: s601, resampling the high-precision water quality parameter layer data of the sampling area obtained in the step S5 to be consistent with the resolution of satellite data by adopting a scale elimination algorithm to obtain satellite scale water quality parameter data; s602, matching the satellite scale water quality parameter data with satellite spectrum data of a corresponding area to construct a correction data set; And S603, adopting a weighted least square method, taking the water quality parameter data of the satellite scale in the corrected data set as a real label, and carrying out parameter iterative optimization on the water quality parameter inversion empirical model constructed in the step S1 until a preset error threshold is met, so as to obtain a corrected satellite inversion model.
9. The satellite-monitor vehicle high-precision collaborative emergency method according to claim 8, wherein the scale elimination algorithm in step S601 includes: calculating a spatial weight factor based on Euclidean distance between sub-meter level pixels and satellite pixel centers Calculating a spectral similarity weight factor based on cosine similarity of sub-meter pixel hyperspectral reflectivity and satellite pixel average hyperspectral reflectivity ; By spatial weighting factors Weighting factors for spectral similarity Weighting and fusing after normalization to obtain comprehensive weight factors ; According to comprehensive weight factors Weighting and summing the water quality parameter values of the sub-meter level pixels to obtain the water quality parameter values of the satellite pixels 。
10. The satellite-monitor vehicle high-precision collaborative emergency method according to claim 8, wherein the objective function of the weighted least squares method in step S603 is: ; ; ; Wherein, the As a function of the object to be processed, To correct the total number of samples in the dataset; true tag value for satellite pixel Confidence weights of (2); mathematical expression of the satellite optimal inversion model; spectral feature vectors for the j-th satellite pixel; The real tag value of the satellite scale water quality parameter is obtained; Sub-meter-level pixel for unmanned aerial vehicle Parameters of water quality Is calculated based on unmanned ship in-situ measured data and unmanned aerial vehicle inversion data; Sub-meter-level pixel for unmanned aerial vehicle In-situ monitoring of water quality parameter values by unmanned ships with nearest spatial positions; Sub-meter-level pixel for unmanned aerial vehicle Is obtained by inversion of a ground hyperspectral inversion model.

Description

High-precision collaborative emergency method and system for satellite-monitoring vehicle Technical Field The invention relates to the technical field of water quality monitoring, in particular to a satellite-monitoring vehicle high-precision collaborative emergency method and system. Background Along with the acceleration of global industrialization and urban process, the problem of water resource pollution is more serious, the frequency of sudden water pollution events is increased, the situation is complicated, and the emergency monitoring of water quality becomes a key defense line for guaranteeing ecological environment safety and people's life health. The traditional water quality monitoring means mainly depend on manual sampling and fixed monitoring stations, but both have obvious short plates when handling sudden water pollution events, the manual sampling needs to consume a large amount of manpower, material resources and time, the monitoring range is limited, the real-time dynamic response to a large-area water area is difficult to realize, the fixed monitoring stations are sparsely distributed, a large amount of monitoring dead areas exist, pollution signals cannot be captured at the first time, and the monitoring layout is difficult to flexibly adjust according to pollution diffusion paths. Satellite remote sensing technology is becoming an important means for water quality monitoring due to its wide coverage and timely monitoring capability. The water quality parameter inversion based on satellite remote sensing is generally constructed by combining historical and long-term ground actual measurement data with synchronous satellite image spectrum data and adopting an empirical model. The empirical models show good robustness and applicability when monitoring water quality abnormality, and can effectively reflect the trend of water quality parameters. However, weather environmental conditions (such as weather changes and aerosol concentration) of the current day can significantly affect the inversion accuracy of satellite remote sensing data, so that the water quality parameter results of the empirical model inversion constructed based on historical data are difficult to meet the requirements of accurate monitoring and evaluation, and further affect the decision support of related departments. With the rapid development of unmanned technology, unmanned aerial vehicles and unmanned ships are gradually applied in the field of water quality monitoring. The unmanned aerial vehicle can quickly reach a target water area by virtue of the characteristics of strong maneuverability and wide monitoring range, acquire high-resolution image data and partial water quality parameters, and the unmanned ship can navigate on the water surface and carry water quality monitoring equipment to realize in-situ monitoring of the water body. The two are combined, and actually measured water sample data can be provided for the unmanned aerial vehicle, so that the water quality inversion model of the unmanned aerial vehicle can be accurately marked. However, unmanned aerial vehicles often have limited flight duration and coverage, and when facing large lakes and reservoirs, cannot comprehensively reflect the water quality condition of the entire water body. At present, no effective means is available for carrying out system integration and cooperative application on the large-scale monitoring capability of satellite remote sensing and the on-site accurate monitoring capability of unmanned aerial vehicle and unmanned ship. Disclosure of Invention The invention aims to provide a high-precision collaborative emergency method and system for a satellite-monitoring vehicle, which integrate the core advantages of satellite remote sensing, unmanned aerial vehicle and unmanned ship, and aim to comprehensively improve the efficiency, accuracy, timeliness and coverage of water quality monitoring; when the satellite remote sensing monitoring system gives out a water quality abnormality alarm, the intelligent remote sensing monitoring vehicle (carrying unmanned aerial vehicle or unmanned ship) carries out on-site actual measurement on an abnormal area to obtain real-time water quality parameters and hyperspectral data on the same day, and based on the actual measurement data, the on-site correction of a satellite inversion model can be realized, the accuracy of a satellite remote sensing inversion image layer is obviously improved, the specificity and the practical value of the satellite remote sensing inversion image layer are enhanced, the related departments are ensured to be capable of obtaining accurate and reliable water quality parameter data, and scientific basis is provided for environmental management and decision making. The high-precision collaborative emergency system of the satellite-monitoring vehicle comprises a central control platform integrated in the intelligent remote sensing monitoring vehicle, a satellite remote