Small and micro water body water quality inversion method and system based on unmanned aerial vehicle spectrum image
Through drone spectral imaging technology, the shortcomings of traditional water quality monitoring in the water quality in the micro rivers are solved, and high spatial resolution water quality monitoring is achieved, providing fast and real-time water quality inversion and pollution source recognition capabilities, supporting rivers Water environment protection.
Patent Information
- Application Number
- CN202410986634.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-23
- Publication Date
- 2024-11-12
- Estimated Expiration
- 2044-07-23
AI Technical Summary
Traditional water quality monitoring methods are difficult to quickly, in real time and accurately reflect the overall spatial water quality status of small and micro river water bodies, and cannot meet the needs of real-time dynamic monitoring, especially in narrow waters, which is difficult to achieve high spatial resolution water quality inversion.
UAV spectral imaging technology is adopted to uniformly arrange sampling points in the target water area, obtain hyperspectral remote sensing images, perform preprocessing and correlation analysis, select relevant spectral parameters as input to the inversion model, and use traditional regression and machine learning algorithms for Fit, generate a spatial distribution map of water quality, and identify the source and causes of pollution.
It realizes accurate inversion of water quality parameters of small and micro rivers, provides fast and real-time water quality monitoring means, can collect sample data in narrow rivers, identify the degree of pollution and causes, and provide data support for subsequent governance.
Smart Images

Figure CN118937254A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of water quality monitoring, and in particular to a method and system for water quality inversion of small micro water bodies based on unmanned aerial vehicle spectral imaging. Background Art
[0002] As an important part of water resources, rural small and medium-sized rivers play a key role in irrigation, flood control and environmental protection. However, rural rivers face threats from excessive use of pesticides and fertilizers, domestic sewage discharge, and untreated livestock and poultry waste runoff during the rainy season, increasing the risk of water quality deterioration.
[0003] Water quality monitoring is an important measure to protect the river water environment and provides a basis for scientific management of river water quality. The purpose of monitoring is to fully understand the current status, spatiotemporal changes and future trends of river water quality. Understanding water quality changes helps to identify pollution sources, assess pollution intensity and impact range, so fast, real-time, accurate and economical monitoring methods are needed.
[0004] Traditional water quality monitoring methods have high measurement accuracy but are time-consuming and labor-intensive, and the measured data can only represent local sample point information, making it difficult to quickly reflect the overall spatial water quality of the water body and unable to meet the requirements of real-time dynamic water quality monitoring. The development of satellite remote sensing technology has achieved a series of successes in the field of water quality monitoring, but it is currently limited to large-scale water bodies such as rivers, lakes and oceans, and it is difficult to meet the narrow water quality inversion needs of small river waters. Summary of the invention
[0005] The purpose of the present invention is to provide a method and system for water quality inversion of small water bodies based on UAV spectral imaging to solve the following technical problems:
[0006] How to obtain high spatial resolution images of small rivers and achieve accurate inversion of water quality parameters.
[0007] The purpose of the present invention can be achieved through the following technical solutions:
[0008] A method for inverting water quality of small water bodies based on UAV spectral images, including:
[0009] S1. Evenly arrange water sampling points in the target waters and record the location information of each sampling point;
[0010] S2, collecting water samples at each sampling point to obtain the corresponding sampled water body, obtaining a hyperspectral remote sensing image of the target water area, and preprocessing the hyperspectral remote sensing image;
[0011] S3, obtaining the spectral parameter V of the hyperspectral remote sensing image i and the water quality information of the sampled water body, the spectral parameter V iConduct correlation analysis with water quality information;
[0012] Among them, the spectral parameter V i The water quality index information is obtained according to the reflectance information of each band of the hyperspectral remote sensing image, and includes a plurality of water quality indexes and corresponding index concentration values;
[0013] S4, selecting the spectral parameters V that rank top m in correlation with each of the water quality indicators i As the input of the water quality index inversion model, it is divided into a training set and a test set according to a specified ratio, and the output value of the inversion model is the concentration value of the corresponding water quality index;
[0014] S5. After evaluating the accuracy of the model, the best inversion model for each water quality index is selected to obtain the spatial distribution map of water quality. The inversion results are analyzed. The areas with sudden color changes indicate the presence of pollution sources and point out their possible causes of pollution.
[0015] As a further solution of the present invention: Step S1 comprises:
[0016] S11, obtaining a satellite image of the target water area, and evenly distributing the sampling points according to the area and shape of the target water area;
[0017] S12: Determine the longitude and latitude position information of each sampling point.
[0018] As a further solution of the present invention: step S2 comprises:
[0019] S21, after acquiring the target water area using a drone equipped with a hyperspectral camera, while collecting water samples, obtain the hyperspectral remote sensing image of the corresponding sampling point at the same time and under the same lighting conditions;
[0020] S22, preprocessing the hyperspectral remote sensing image;
[0021] Wherein, the preprocessing includes mirror transformation, reflectivity calibration, and atmospheric correction.
[0022] As a further solution of the present invention: step S3 comprises:
[0023] S31, extracting the reflectance information of each band corresponding to each sampling point in the hyperspectral remote sensing image;
[0024] S32, band combining the single-band reflectance values, and recording the band combination result as the spectral parameter V i ;
[0025] S33, spectral parameter V i Correlation analysis was performed with water quality information.
[0026] As a further solution of the present invention: the correlation analysis includes:
[0027]
[0028] Among them, k is the total number of sampling points, R xy is the correlation coefficient between spectral parameters and water quality information, x c With y c are the sample values of spectral parameters and water quality information, and are the sample average values of spectral parameters and water quality information, respectively.
[0029] As a further solution of the present invention: in the step S5, the method of selecting the best inversion model for each water quality index after evaluating the model accuracy includes:
[0030] According to the goodness of fit R of the inversion model 2 , mean square error RMSE and average percentage error MAPE to obtain the model score S;
[0031]
[0032] Wherein, α and β are both preset weighting coefficients.
[0033] As a further solution of the present invention: the model accuracy assessment includes:
[0034] The goodness of fit R 2 How to obtain:
[0035]
[0036] The method for obtaining the mean square error RMSE is:
[0037]
[0038] The method for obtaining the mean percentage error MAPE is:
[0039]
[0040] Among them, n is the total number of validation samples, y i is the actual value, f i is the predicted value.
[0041] As a further solution of the present invention: a water quality inversion system for small water bodies based on drone spectral images, comprising:
[0042] A data acquisition module is used to evenly arrange water sampling points in the target water area and record the location information of each sampling point; collect water samples at each sampling point to obtain the corresponding sampled water body, obtain the hyperspectral remote sensing image of the target water area, and pre-process the hyperspectral remote sensing image;
[0043] Feature extraction and analysis module, used to obtain the spectral parameters V of the hyperspectral remote sensing image i and the water quality information of the sampled water body, the spectral parameter V i Conduct correlation analysis with water quality information;
[0044] Among them, the spectral parameter V i The water quality index information is obtained according to the reflectance information of each band of the hyperspectral remote sensing image, and includes a plurality of water quality indexes and corresponding index concentration values;
[0045] A model building module is used to select the spectral parameters V with the top m correlation rankings with each of the water quality indicators. i As the input of the water quality index inversion model, it is divided into a training set and a test set according to a specified ratio, and the output value of the inversion model is the concentration value of the corresponding water quality index;
[0046] The application and analysis module is used to select the best inversion model for each water quality index after evaluating the model accuracy, obtain the water quality spatial distribution map, and analyze the inversion results. The area with sudden color changes indicates the presence of pollution sources and points out the possible causes of pollution.
[0047] Beneficial effects of the invention: The invention provides a method and system for water quality inversion of small water bodies based on UAV hyperspectral imaging. During the detection process, some physical and chemical detection means are used to obtain the concentration values of water quality indicators, and water quality monitoring is achieved based on UAV remote sensing. The collection of measured data from sampling points can be achieved in narrow rivers with complex water quality components. The characteristic spectral parameters obtained by single band / band combination are used to fit and train the model obtained by traditional regression / machine learning algorithms, and then the inverted water quality conditions are analyzed with reference to relevant standards and specifications. This can realize the identification of the degree of water pollution and the preliminary diagnosis of the possible causes of severely polluted river sections, providing reference data support for subsequent governance. BRIEF DESCRIPTION OF THE DRAWINGS
[0048] The present invention will be further described below in conjunction with the accompanying drawings.
[0049] Figure 1 This is a principle framework diagram of water quality inversion in the present invention. DETAILED DESCRIPTION
[0050] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0051] See also Figure 1 As shown, the present invention is a method for inverting water quality of small water bodies based on UAV spectral images, comprising:
[0052] S1. Evenly arrange water sampling points in the target waters and record the location information of each sampling point;
[0053] S2, collecting water samples at each sampling point to obtain the corresponding sampled water body, obtaining a hyperspectral remote sensing image of the target water area, and preprocessing the hyperspectral remote sensing image;
[0054] S3, obtaining the spectral parameter V of the hyperspectral remote sensing image i and the water quality information of the sampled water body, the spectral parameter V i Conduct correlation analysis with water quality information;
[0055] Among them, the spectral parameter V i The water quality index information is obtained according to the reflectance information of each band of the hyperspectral remote sensing image, and includes a plurality of water quality indexes and corresponding index concentration values;
[0056] S4, selecting the spectral parameters V that rank top m in correlation with each of the water quality indicators i As the input of the water quality index inversion model, it is divided into a training set and a test set according to a specified ratio, and the output value of the inversion model is the concentration value of the corresponding water quality index;
[0057] S5. After evaluating the accuracy of the model, the best inversion model for each water quality index is selected to obtain the spatial distribution map of water quality. The inversion results are analyzed. The areas with sudden color changes indicate the presence of pollution sources and point out their possible causes of pollution.
[0058] In this embodiment of the present invention, m. can be selected as 5.
[0059] In step S1:
[0060] S11, obtaining a satellite image of the target water area, and evenly distributing the sampling points according to the area and shape of the target water area;
[0061] S12: Determine the longitude and latitude position information of each sampling point.
[0062] In step S2, the supporting software can be used to plan the flight route collected in the experiment to drive the drone.
[0063] S21. Use a drone equipped with a hyperspectral camera to obtain a hyperspectral remote sensing image of the target waters. Compared with traditional manual sampling monitoring and automatic monitoring station monitoring, drone monitoring is fast, wide-ranging, and effective. Compared with satellite remote sensing monitoring, drone monitoring has a higher resolution. While collecting water samples, operate the drone equipped with a GaiaSky-mini3-VN airborne spectral imager to shoot over the waters to obtain the hyperspectral remote sensing image of the corresponding sampling point at the same time and under the same lighting conditions. The selected time period is sunny and breezy (level 1 to 2), with no cloud cover.
[0064] When collecting water samples, a glass sampler is used at each sampling point to collect 500 mL of water sample at 0.5 m below the water surface. When taking water samples, care is taken to avoid extremely narrow river channels and places where there are suspended objects such as duckweed. The collected water samples are promptly sent to relevant laboratories for testing to obtain the concentration values of water quality indicators that need to be inverted. The measured water quality indicators include total nitrogen, total phosphorus, ammonia nitrogen, nitrate nitrogen, turbidity, dissolved oxygen, etc.
[0065] S22, pre-processing the hyperspectral remote sensing image, mainly including mirror transformation, reflectivity calibration, and atmospheric correction. The above processing flow is mainly completed using software such as SpectraVIEW.
[0066] In step S3:
[0067] S31, extracting the reflectance information of each band corresponding to each sampling point in the hyperspectral remote sensing image;
[0068] In this embodiment, reflectivity extraction: stitch the study area images for a large amount of data, export the stitched image data to .tiff format, open ArcMap10.8 software, add the longitude and latitude corresponding to each sampling point of the ground water quality data to the layer in the form of .xlsx / .xls data as input point features. Add the stitched drone hyperspectral image to the layer in .tiff format as input raster data. Select System Toolbox-SpatialAnalyst Tool-Extraction Analysis-Multi-value Extraction to Point in the toolbar, open the attribute table, and you can view the spectral reflectance values of each band corresponding to the sampling points of the ground water quality data;
[0069] S32, band combining the single-band reflectance values, and recording the band combination result as the spectral parameter V i Band combination operations such as band ratio, addition, and subtraction can reduce background noise, increase the sensitivity of spectral reflectance to water quality parameters, and improve inversion accuracy.
[0070] The main band combinations used in this study are as follows: a +R b , R a -R b , R a / R b , R a -R b / R a +R b ;
[0071] Among them, R a , R b They represent the reflectivity values of any two bands within the range of 400 to 1000 nm, and a≠b.
[0072] S33, spectral parameter V i The correlation analysis can be carried out with water quality information. Specifically, Preson correlation analysis can be completed in SPSS software to calculate the correlation between the spectral parameter Vi and the concentration values of water quality indicators (total nitrogen, total phosphorus and ammonia nitrogen, nitrate nitrogen, turbidity, dissolved oxygen).
[0073] The correlation analysis includes:
[0074]
[0075] Among them, k is the total number of sampling points, R xy is the correlation coefficient between spectral parameters and water quality information, x c With y c are the sample values of spectral parameters and water quality information, and are the sample averages of spectral parameters and water quality information, respectively; R xy The closer the absolute value is to 1, the higher the correlation is.
[0076] In step S4:
[0077] Select the top 5 spectral parameters V that are correlated with each water quality index i As the input of each water quality index inversion model, the output value is the concentration value of the corresponding water quality index;
[0078] The dataset is divided into training set and test set in a ratio of 7:3.
[0079] Use the training set data to train a variety of traditional regression / machine learning models; use the validation set data and combine multiple accuracy evaluation indicators (R 2 , RMSE, and MAPE) were used to evaluate the water quality inversion model.
[0080] The five band spectral parameters with the best correlation divided the dataset into two parts, 70% for training set and 30% for validation set.
[0081] The traditional regression models in the training set mainly include linear models, exponential models, and polynomial models. The machine learning models mainly include partial least squares (PLS), support vector machine (SVM), random forest (RF), and BP neural network (BP).
[0082] In step S5:
[0083] The method of selecting the best inversion model for each water quality index after evaluating the model accuracy includes:
[0084] According to the goodness of fit R of the inversion model 2 , mean square error RMSE and average percentage error MAPE to obtain the model score S; the higher S is, the better the inversion model is.
[0085]
[0086] Wherein, α and β are both preset weighting coefficients.
[0087] Specifically, the model accuracy assessment includes:
[0088] The goodness of fit R 2 How to obtain:
[0089]
[0090] The method for obtaining the mean square error RMSE is:
[0091]
[0092] The method for obtaining the mean percentage error MAPE is:
[0093]
[0094] Among them, n is the total number of validation samples, y i is the actual value, f i is the predicted value.
[0095] As a further solution of the present invention: a water quality inversion system for small water bodies based on drone spectral images, comprising:
[0096] A data acquisition module is used to evenly arrange water sampling points in the target water area and record the location information of each sampling point; collect water samples at each sampling point to obtain the corresponding sampled water body, obtain the hyperspectral remote sensing image of the target water area, and pre-process the hyperspectral remote sensing image;
[0097] Feature extraction and analysis module, used to obtain the spectral parameters V of the hyperspectral remote sensing image i and the water quality information of the sampled water body, the spectral parameter V i Conduct correlation analysis with water quality information;
[0098] Among them, the spectral parameter V i The water quality index information is obtained according to the reflectance information of each band of the hyperspectral remote sensing image, and includes a plurality of water quality indexes and corresponding index concentration values;
[0099] A model building module is used to select the spectral parameters V with the top m correlation rankings with each of the water quality indicators. i As the input of the water quality index inversion model, it is divided into a training set and a test set according to a specified ratio, and the output value of the inversion model is the concentration value of the corresponding water quality index;
[0100] The application and analysis module is used to select the best inversion model for each water quality index after evaluating the model accuracy, obtain the water quality spatial distribution map, and analyze the inversion results. The area with sudden color changes indicates the presence of pollution sources and points out the possible causes of pollution.
[0101] The above is a detailed description of an embodiment of the present invention, but the content is only a preferred embodiment of the present invention and cannot be considered to limit the scope of implementation of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of the patent coverage of the present invention.
Claims
1. A method for water quality inversion of small water bodies based on UAV spectral images, characterized in that: include: S1. Evenly arrange water sampling points in the target waters and record the location information of each sampling point; S2, collecting water samples at each sampling point to obtain the corresponding sampled water body, obtaining a hyperspectral remote sensing image of the target water area, and preprocessing the hyperspectral remote sensing image; S3, obtaining the spectral parameter V of the hyperspectral remote sensing image i and the water quality information of the sampled water body, the spectral parameter V i Conduct correlation analysis with water quality information; Among them, the spectral parameter V i The water quality index information is obtained according to the reflectance information of each band of the hyperspectral remote sensing image, and includes a plurality of water quality indexes and corresponding index concentration values; S4, selecting the spectral parameters V that rank top m in correlation with each of the water quality indicators i As the input of the water quality index inversion model, it is divided into a training set and a test set according to a specified ratio, and the output value of the inversion model is the concentration value of the corresponding water quality index; S5. After evaluating the accuracy of the model, the best inversion model for each water quality index is selected to obtain the spatial distribution map of water quality. The inversion results are analyzed. The areas with sudden color changes indicate the presence of pollution sources and point out their possible causes of pollution.
2. The method for water quality inversion of small and micro water bodies based on UAV spectral imaging according to claim 1 is characterized in that: The step S1 comprises: S11, obtaining a satellite image of the target water area, and evenly distributing the sampling points according to the area and shape of the target water area; S12: Determine the longitude and latitude position information of each sampling point.
3. The method for water quality inversion of small and micro water bodies based on UAV spectral imaging according to claim 1 is characterized in that: The step S2 comprises: S21, after acquiring the target water area using a drone equipped with a hyperspectral camera, while collecting water samples, obtain the hyperspectral remote sensing image of the corresponding sampling point at the same time and under the same lighting conditions; S22, preprocessing the hyperspectral remote sensing image; Wherein, the preprocessing includes mirror transformation, reflectivity calibration, and atmospheric correction.
4. The method for water quality inversion of small and micro water bodies based on UAV spectral imaging according to claim 1 is characterized in that: The step S3 comprises: S31, extracting the reflectance information of each band corresponding to each sampling point in the hyperspectral remote sensing image; S32, band combining the single-band reflectance values, and recording the band combination result as the spectral parameter V i ; S33, spectral parameter V i Correlation analysis was performed with water quality information.
5. The method for water quality inversion of small water bodies based on drone spectral images according to claim 4 is characterized in that: The correlation analysis includes: Among them, k is the total number of sampling points, R xy is the correlation coefficient between spectral parameters and water quality information, x c With y c are the sample values of spectral parameters and water quality information, and are the sample average values of spectral parameters and water quality information, respectively.
6. The method for water quality inversion of small and micro water bodies based on UAV spectral imaging according to claim 5 is characterized in that: In step S5, the method of selecting the best inversion model for each water quality index after evaluating the model accuracy includes: According to the goodness of fit R of the inversion model 2 , mean square error RMSE and average percentage error MAPE to obtain the model score S; Wherein, α and β are both preset weighting coefficients.
7. The method for water quality inversion of small and micro water bodies based on drone spectral images according to claim 6 is characterized in that: The model accuracy assessment includes: The goodness of fit R 2 How to obtain: The method for obtaining the mean square error RMSE is: The method for obtaining the mean percentage error MAPE is: Among them, n is the total number of validation samples, y i is the actual value, f i is the predicted value.
8. A water quality inversion system for small water bodies based on drone spectral images, characterized in that: include: A data acquisition module is used to evenly distribute water sampling points in the target water area and record the location information of each sampling point; While collecting water samples at each sampling point to obtain the corresponding sampled water body, a hyperspectral remote sensing image of the target water area is obtained, and the hyperspectral remote sensing image is preprocessed; Feature extraction and analysis module, used to obtain the spectral parameters V of the hyperspectral remote sensing image i and the water quality information of the sampled water body, the spectral parameter V i Conduct correlation analysis with water quality information; Among them, the spectral parameter V i The water quality index information is obtained according to the reflectance information of each band of the hyperspectral remote sensing image, and includes a plurality of water quality indexes and corresponding index concentration values; A model building module is used to select the spectral parameters V with the top m correlation rankings with each of the water quality indicators. i As the input of the water quality index inversion model, it is divided into a training set and a test set according to a specified ratio, and the output value of the inversion model is the concentration value of the corresponding water quality index; The application and analysis module is used to select the best inversion model for each water quality index after evaluating the model accuracy, obtain the water quality spatial distribution map, and analyze the inversion results. The area with sudden color changes indicates the presence of pollution sources and points out the possible causes of pollution.
Citation Information
Patent Citations
Gamma radiation source item division method and system based on information criterion
CN109521457A
Air-ground-space integrated hyperspectral water quality monitoring and analyzing method
CN110865040A
Load model goodness of fit evaluation method based on cloud matter element theory
CN111105168A
Chlorophyll a remote sensing inversion method and device based on genetic neural network model
CN112666328A
Water quality inversion method, device and equipment based on remote sensing image
CN115410086A
Cited By
Integrated hyperspectral water quality analysis method
CN119337251A
Water quality parameter inversion method based on hyperspectral data analysis
CN119600454A
A method for retrieving water quality parameters based on hyperspectral data analysis
CN119600454B
Unmanned aerial vehicle-mounted hyperspectral water quality monitoring system and method with automatic calibration
CN119845865A
Method for monitoring water quality of aquaculture water area
CN120558872A
