CN115524294A - Water leaving type real-time intelligent remote sensing water quality monitoring method - Google Patents
Water leaving type real-time intelligent remote sensing water quality monitoring method Download PDFInfo
- Publication number
- CN115524294A CN115524294A CN202211135571.7A CN202211135571A CN115524294A CN 115524294 A CN115524294 A CN 115524294A CN 202211135571 A CN202211135571 A CN 202211135571A CN 115524294 A CN115524294 A CN 115524294A
- Authority
- CN
- China
- Prior art keywords
- water quality
- water
- remote sensing
- intelligent
- hyperspectral
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/17—Systems in which incident light is modified in accordance with the properties of the material investigated
- G01N21/25—Colour; Spectral properties, i.e. comparison of effect of material on the light at two or more different wavelengths or wavelength bands
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/01—Arrangements or apparatus for facilitating the optical investigation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/01—Arrangements or apparatus for facilitating the optical investigation
- G01N2021/0106—General arrangement of respective parts
- G01N2021/0112—Apparatus in one mechanical, optical or electronic block
Landscapes
- Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- Pathology (AREA)
- Evolutionary Computation (AREA)
- Immunology (AREA)
- General Health & Medical Sciences (AREA)
- Chemical & Material Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Theoretical Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Medical Informatics (AREA)
- Geometry (AREA)
- Computer Hardware Design (AREA)
- Software Systems (AREA)
- Spectroscopy & Molecular Physics (AREA)
- Investigating Or Analysing Materials By Optical Means (AREA)
Abstract
The invention discloses an off-water real-time intelligent remote sensing water quality monitoring method, which comprises the following steps: acquiring hyperspectral remote sensing reflectivity data of a water body to be measured; collecting synchronous surface water quality samples of a water body to be detected, and constructing a water quality hyperspectral remote sensing reflectivity data set; constructing an optimal water quality intelligent inversion model based on a water quality hyperspectral remote sensing reflectivity data set; and obtaining real-time water quality data based on the hyperspectral remote sensing reflectivity data of the water body to be measured and the optimal water quality intelligent inversion model. According to the invention, atmospheric correction is not needed, the monitoring deficiency of satellite and unmanned aerial vehicle remote sensing rainy weather is made up, and the precision of water quality parameter inversion is improved by high spectral resolution; the unattended continuous water body observation is realized by timing acquisition; the three-probe imaging spectrometer carried on the intelligent mobile terminal is small, exquisite, convenient to carry and flexible; the energy consumption, the equipment loss caused by wind waves and the difficulty of equipment maintenance are reduced, and the data error caused by biological attachment pollution is reduced.
Description
Technical Field
The invention relates to the field of water quality monitoring, in particular to an off-water real-time intelligent remote sensing water quality monitoring method.
Background
The method can rapidly and accurately obtain water quality data, objectively reflect the state of the water body, and is an important basis for analyzing a water quality problem cause mechanism, controlling the input of pollutants in a drainage basin, reducing water body pollution and restoring ecology. However, river and lake water quality is affected by various factors such as weather, hydrology, self-properties of pollutants and the like, and the time-space heterogeneity is strong. The traditional manual sampling based on laboratory analysis has the defects of high cost, time and labor waste, time-space dispersion, poor timeliness and the like, and is difficult to capture some rapid lake and reservoir change processes with strong artificial interference and complex water regime in time. Satellite remote sensing has been widely used for water quality monitoring such as transparency, chlorophyll, total nitrogen, total phosphorus and the like due to the advantages of large area, periodicity, economy, high efficiency and the like, but limited time resolution and cloud and rain weather make the satellite remote sensing still have great efforts on solving the problem of long-term and continuous water quality monitoring; in addition, for some small and medium-sized water bodies, the traditional water color satellites such as MODIS, seaWiFS, GOCI and the like cannot provide effective water quality information due to low spatial resolution, and the medium-high resolution terrestrial satellites such as LANDSAT series and Gaofen series cannot provide short-term dynamic information of the water bodies due to long playback period, wide band, low signal-to-noise ratio and the like. The hyperspectral remote sensing of the unmanned aerial vehicle makes up the defects of water quality monitoring of small and medium water bodies due to flexible maneuvering, but is limited by endurance time and cloud and rain weather, so that continuous observation is difficult to realize. The shore-based hyperspectral water quality observation equipment which is developed rapidly in recent years can provide continuous high-frequency observation, but the fixed observation position limits the use of the equipment in lakes with strong spatial heterogeneity. With the development of the technology, the underwater high-frequency probe and the multi-parameter water quality measuring instrument can carry out continuous high-frequency observation on water quality, but the large-range application of the underwater high-frequency probe and the multi-parameter water quality measuring instrument is limited by the defects of low monitoring precision, high price, difficult maintenance, easy adhesion of pollution and the like. Therefore, the existing water quality monitoring means lags behind the requirements of water environment management and decision-making departments in terms of data acquisition frequency, accuracy, timeliness and representativeness, and particularly, some sudden, large-scale sewage discharge and cross-regional pollution events cannot be captured in time.
Therefore, a real-time water quality monitoring method which has high space-time resolution, is suitable for various water qualities and various weathers and is flexible and mobile is urgently needed to be provided, the defects of the existing observation means are overcome, and the power is provided for diagnosing water quality pollution, analyzing a cause mechanism and scientifically preventing and controlling.
Disclosure of Invention
In order to solve the problems in the prior art, the invention provides an off-water real-time intelligent remote sensing water quality monitoring method, which combines a three-probe hyperspectral imager with an intelligent mobile terminal, matches synchronous water quality data, and is based on an intelligent water quality inversion method of deep learning, so that the rapid real-time monitoring of water quality of different types of water bodies under complex weather is realized, the intellectualization, miniaturization, simple and easy operation and motorized level of water environment monitoring are improved, and the access threshold to equipment and personnel in water quality monitoring is reduced.
In order to achieve the technical purpose, the invention provides an off-water real-time intelligent remote sensing water quality monitoring method, which comprises the following steps:
s1, acquiring hyperspectral remote sensing reflectivity data of a water body to be detected;
s2, collecting synchronous surface layer water quality samples of the water body to be detected, and constructing a water quality hyperspectral remote sensing reflectivity data set;
s3, constructing an optimal water quality intelligent inversion model based on the water quality hyperspectral remote sensing reflectivity data set;
and S4, obtaining real-time water quality data based on the hyperspectral remote sensing reflectivity data of the water body to be detected and the optimal water quality intelligent inversion model, and realizing the water leaving type real-time intelligent remote sensing water quality monitoring.
Optionally, a spectrum acquisition system is used to acquire the hyperspectral data of the water body, and the spectrum acquisition system includes: the system comprises an intelligent mobile terminal, a three-probe hyperspectral imager and a cloud deck;
the three-probe hyperspectral imager is connected with the intelligent mobile terminal, and the intelligent mobile terminal is connected with the holder through a telescopic buckle; the holder fixes an observation angle through an adjusting nut, and the observation angle ensures that the angle between the long side of the three-probe hyperspectral imager and the horizontal plane is 0 degree.
Optionally, the hyperspectral remote sensing reflectance data comprises: and (3) water body hyperspectral remote sensing reflectivity data of different types of water bodies, water bodies with different water quality conditions and water bodies with different weather conditions.
Optionally, the different types of water bodies include inland water bodies and clear ocean water bodies of different nutrient levels and different degrees of clarity; the water bodies with different water quality conditions comprise water bodies with different wind wave sizes, different algal bloom outbreak conditions and different turbidity degrees; the different weather conditions include sunny days, cloudy days and light rains;
optionally, the construction process of the optimal water quality intelligent inversion model is as follows:
constructing an optimal water quality intelligent inversion model by adopting various machine learning algorithms based on the water quality hyperspectral remote sensing reflectivity data set;
the multiple machine learning algorithms include a random forest algorithm, a neural network algorithm and a Gaussian process regression algorithm.
Optionally, the optimal water quality intelligent inversion model is written into application software suitable for an intelligent mobile terminal, and the application software can call a calculation function, a storage function, a data display function and a downloading function of the intelligent mobile terminal.
Optionally, the real-time water quality data comprises: water quality parameters and environmental condition parameters;
the water quality parameters comprise: total nitrogen, total phosphorus, chlorophyll, transparency, suspended matter, permanganate index, turbidity, extinction coefficient, ammonia nitrogen, colored soluble organic matter absorption coefficient, soluble organic carbon, granular organic carbon and eutrophication index;
the environmental condition parameters include: observation point longitude and latitude, observation time, air temperature, observation angle, weather and wind speed.
Optionally, preprocessing the hyperspectral remote sensing reflectivity data and the water sample to be detected;
the preprocessing of the hyperspectral remote sensing reflectivity data comprises the following steps: rejecting a non-water body spectrum, rejecting an abnormal water body spectrum and normalizing spectrum data;
the pretreatment of the water sample to be detected comprises the following steps: suction filtration, low-temperature sample storage, laboratory measurement and water sample data abnormal value elimination.
Optionally, when the water quality index exceeds a specific threshold value during the water leaving type real-time intelligent remote sensing water quality monitoring, the alarm reminding is carried out by changing the display color and popping up a window;
the specific threshold is a national water quality evaluation standard or a self-defined threshold.
The invention has the following technical effects:
1. compared with the traditional space foundation, space foundation and unmanned aerial vehicle remote sensing, the water leaving type real-time intelligent remote sensing monitoring method for water quality provided by the invention has the advantages that the high-spectrum remote sensing reflectivity data of the water body is collected near the water surface, the signal-to-noise ratio of the data is increased, and the influence of atmosphere and aerosol is basically avoided, so that atmosphere correction is not needed; in addition, the wide observation weather conditions comprise light rain, cloudy days and cloudy days, the spectrum imaging time range is expanded, and the monitoring deficiency of satellite and unmanned aerial vehicle remote sensing cloudy weather is overcome; finally, the high spectral resolution improves the accuracy of water quality parameter inversion;
2. compared with the existing ground feature spectrograph or hyperspectral imager, the three-probe hyperspectral imager can simultaneously acquire downward irradiance on the water surface, skylight and water body leaving radiance without a grey standard plate and a complex operation and processing process, so that the leaving remote sensing reflectivity deviation caused by rapid change of the light environment is avoided; in addition, the timing acquisition realizes the unattended continuous water body observation; compared with the traditional ground object spectrometer, the three-probe imaging spectrometer carried on the intelligent mobile terminal is small, exquisite, convenient to carry and flexible;
3. the three-probe hyperspectral imager, the holder and the intelligent algorithm application software do not depend on a fixed smartphone brand, can realize the use of various intelligent terminal devices, and reduce the requirements on spectrum acquisition equipment;
4. compared with land-based hyperspectral remote sensing equipment, the three-probe hyperspectral imager can simultaneously measure downlink irradiance, uplink radiance and sky radiance, eliminates errors of sky light in the process of calculating remote sensing reflectivity, obtains more accurate water-leaving radiance information, and provides a foundation for high performance and precision of a model;
5. compared with the traditional underwater probe equipment, the method and the device for inverting the water quality parameter information do not need to directly contact with a water body, so that not only are the energy consumption, the equipment loss caused by wind and waves and the difficulty of equipment maintenance reduced, but also the data error caused by biological attachment pollution is reduced.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings needed to be used in the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings without inventive exercise.
FIG. 1 is a flow chart of an off-water real-time intelligent remote sensing water quality monitoring method according to an embodiment of the invention;
FIG. 2 is a schematic structural diagram of a three-probe hyperspectral imager in an embodiment of the invention;
FIG. 3 is a schematic diagram of a side view and a front view of a spectrum acquisition system of the water-leaving real-time intelligent remote sensing water quality monitoring method in the embodiment of the invention;
FIG. 4 shows three observation modes of the water-leaving real-time intelligent remote sensing water quality monitoring method of the embodiment of the invention, wherein (a) is land-based fixed type, (b) is ship-borne type, and (c) is hand-held type;
FIG. 5 is a schematic diagram of a relationship between a hyperspectral remote sensing reflectance ratio and a surface feature spectrometer before and after normalization according to an embodiment of the invention;
fig. 6 validation results of suspended particulate concentration in gaussian regression model (top), neural network model (middle), and random forest model (bottom) in training set (left), validation set (middle), and whole data set (right).
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
As shown in figure 1, the invention discloses an off-water real-time intelligent remote sensing water quality monitoring method, which comprises the following steps:
s1, acquiring hyperspectral remote sensing reflectivity data of a water body to be detected;
in 2022, 21 days in 7 months to 3 days in 8 months, a field synchronous experiment of a continuous water-leaving type real-time intelligent remote sensing monitoring method for water quality is developed at a research station of a Tai lake ecosystem of the Wuxi Chinese academy of sciences. Firstly, a three-probe hyperspectral imager (shown in figure 2) which is designed autonomously and comprises an irradiance meter with the top end collecting downward irradiance and a horizontally rotatable radiance meter with the zenith angles of 40 degrees and 140 degrees is carried on the back of an intelligent mobile terminal (a smart phone) by using an adjustable buckle, then the imager is placed on a holder with a vertical telescopic rod by using the adjustable buckle, a spectrum collection system of an off-water real-time water quality intelligent remote sensing monitoring method is built together, and figure 3 is a side view and a front view of the built spectrum collection system.
As shown in fig. 4, in order to test different carrying modes, the present embodiment respectively collects hyperspectral remote sensing reflectance data of a water body by using three carrying modes, namely a fixed ground-based vertical rod, a motorized observation ship and a manual hand-held carrying mode. The land-based fixed type comprises: firstly, a spectrum acquisition system is horizontally fixed on a vertical rod of a fixed shore base, and according to the distance between the water surface and the horizontal distance of 2 meters, the hyperspectral remote sensing reflectivity data of the water body to be detected are acquired by a Bluetooth transmission timing acquisition instruction. The motor-driven observation ship comprises: the spectrum acquisition system is carried on a motor monitoring ship, a telescopic rod of a tripod head is fixed on a deck by a powerful sucker, the tripod head horizontally extends out of the ship to enable the equipment to be horizontally 2 meters away from the ship, and the hyperspectral remote sensing reflectivity data are acquired by timing acquisition instructions through Bluetooth transmission. The manual hand-held type includes: the device is manually held, the volume key of the smart phone is manually controlled, and the data of the hyperspectral remote sensing reflectivity of the water body are acquired by transmitting and acquiring instructions through Bluetooth. Although the carrying modes are different, the three-probe hyperspectral imager is required to be kept in a horizontal state when the hyperspectral remote sensing reflectivity data are collected, the collected spectral range is 400-900nm, and the spectral resolution is 1nm.
The obtained observation scene of the hyperspectral remote sensing reflectivity data of the water body to be detected comprises different water body types (inland water bodies and clear ocean water bodies with different nutrition levels and different clarity degrees), different weather conditions (sunny days, cloudy days and light rain), different water quality conditions (different stormy waves, different algal conditions and different turbidity degrees), different stormy wave intensities comprising small stormy waves (the wind speed is lower than 2 m/s), medium stormy waves (the wind speed is between 2 and 5 m/s) and big stormy waves (the wind speed is higher than 5 m/s).
S2, collecting synchronous surface layer water quality samples of the water body to be detected, and constructing a water quality hyperspectral remote sensing reflectivity data set;
when the spectrum acquisition system acquires the data of the high spectrum remote sensing reflectivity, surface water around the observation point is acquired as a synchronous surface water quality sample, and the acquisition of the synchronous surface water quality sample follows the following standards: the synchronous surface water quality sample is a mixed water body with water columns of 0-50cm below the water surface within 10cm around a water body hyperspectral remote sensing reflectivity data acquisition point, within 1 minute of acquisition time difference, and the sampling time, wind speed, transparency, weather, temperature and the like are recorded.
All collected synchronous surface water quality samples are immediately pretreated, including chlorophyll and suspended particle samples are filtered, filtered water is stored, phytoplankton is fixed, and the like, all the treated samples are stored in a dark environment at 4 ℃ and are conveyed to a laboratory within 4 hours to be measured for chlorophyll, extinction coefficient, chemical oxygen demand, absorption coefficient of colored soluble organic matters, diffusion attenuation coefficient, total nitrogen, total phosphorus, suspended particle concentration, ammonia nitrogen and turbidity; in order to ensure the accuracy and reliability of data, negative values of water quality parameters measured in a laboratory are removed;
because the wave can cause the flare effect, covers water information, consequently this embodiment carries out data cleaning to hyperspectral remote sensing reflectivity data, rejects abnormal data, specifically is: eliminating the spectrum with the reflectivity of 550nm higher than 0.1 as an abnormal spectrum, and eliminating the spectrum with the reflectivity of 440nm higher than 0.05 as abnormal data;
because the response relation between the hyperspectral remote sensing reflectivity data collected under different weather conditions and the synchronous hyperspectral response measured by the surface feature spectrometer is inconsistent, the average value of the remote sensing reflectivity of 575-585nm is used as a reference, ratio normalization processing is carried out on each wave band, the result is shown in figure 5, and the hyperspectral remote sensing reflectivity data and the synchronous hyperspectral response measured by the surface feature spectrometer are distributed in the following steps: around the line 1, the coefficient of determination higher than 0.99 shows that the consistency of the two is very high under various weather conditions, and the method can be used for water quality parameter inversion.
Collecting a synchronous surface water quality sample of a water body to be measured, and constructing to obtain a water quality hyperspectral remote sensing reflectivity data set; in order to accurately obtain a water quality intelligent inversion model, a water quality hyperspectral reflectivity data set matched with the acquisition time is randomly divided into a training data set and a verification data set, wherein two thirds of the data set are randomly selected as the training set, and the other third of the data set is selected as the verification set.
S3, constructing an optimal water quality intelligent inversion model by adopting various machine learning algorithms based on the water quality hyperspectral remote sensing reflectivity data set;
taking 400-900nm hyperspectral remote sensing reflectivity data and water quality parameter data in a training set as input data, totaling 501 wave bands, and inputting the input data into a water quality intelligent inversion model constructed by adopting a multi-machine learning algorithm. The multi-machine learning algorithm comprises a Gaussian regression algorithm, a neural network algorithm and a random forest algorithm, and the parameter settings of machine learning are respectively as follows: the Kernel function of the Gaussian regression algorithm is set as Square expanded Kernel, the display basis is Constant, the Sigma initial value is set as 0.1, the parameter inversion method is set as SD, the prediction method is Exact, and the parameter optimizer is quasinesewton; the training function of the neural network algorithm is a tangent S-type function, the number of input nodes is 501, the number of output nodes is 1, a hidden layer is 10 layers, the learning rate is 0.1, and the iteration number is 2000; the number of the decision trees is set to be 60 according to the random forest algorithm parameters, the minimum leaf node sample number is 5, the maximum leaf node sample number is unlimited, and regressions are used in the training and predicting method.
Then, taking 400-900nm hyperspectral remote sensing reflectance data in the verification data set as input data, totaling 501 wave bands, inputting the data into the constructed water quality intelligent inversion model to obtain a water quality parameter inversion result, and calculating a decision coefficient (R) of the water quality intelligent inversion model by combining synchronous water quality parameter data 2 ) The Root Mean Square Error (RMSE) and the average relative error (MRE), judging the performance of the water quality intelligent inversion model and screening the indexes of the optimal water quality intelligent inversion model, wherein the screening standard is as follows: r 2 And the model with high simultaneous RMSE and MAPE being small is the optimal water quality intelligent inversion model. It should be noted that, because different machine learning algorithm models have different precision fitting effects on the same parameter, machine learning algorithms selected by different parameters may be different when selecting an optimal water quality intelligent model, and the comparison process takes the concentration of suspended particulate matter as an example, and includes the following contents:
fig. 6 shows the verification results of the concentrations of the suspended particles in the gaussian regression model (top), the neural network model (middle) and the random forest model (bottom) in the training set (left), the verification set (middle) and the whole data set (right), and after comparison, the random forest model is selected as the water quality intelligent inversion model with the optimal concentration of the suspended particles.
Further, writing the constructed optimal water quality intelligent inversion model into application software suitable for the smart phone, wherein the application software can call the calculation function, the storage function, the data display function and the downloading function of the smart phone; the calculation function is that acquired data is applied to the optimal water quality intelligent inversion model as input parameters through a three-probe hyperspectral imager to directly obtain a water quality parameter result; the display function comprises displaying the water quality parameter inversion result in a form of a table, a time series diagram and a classification color setting form; the download function can download data to a local computer through a download button provided in the application software through wireless Bluetooth transmission or smart phone data line connection.
S4, obtaining real-time water quality data based on the hyperspectral remote sensing reflectivity data of the water body to be measured and the optimal water quality intelligent inversion model, and realizing water leaving type real-time intelligent remote sensing water quality monitoring;
and (4) compiling the optimal water quality intelligent inversion model obtained in the step (3) into application software by using an android system language and an apple system language respectively, transmitting and receiving data and instructions through wireless Bluetooth, applying the acquired and input hyperspectral remote sensing reflectivity data of the water body to be detected as input data to the optimal water quality intelligent inversion model, and obtaining a water quality inversion result and storing the water quality inversion result in a local hard disk. The application software interface can display the water quality parameter inversion result in a form of a table, a time sequence diagram and a classification coloring form, is additionally provided with a downloading function, and can be connected and downloaded to a local computer through a downloading button in the application software through a wireless Bluetooth or smart phone data line.
And (3) applying the hyperspectral remote sensing reflectivity data of the water body to be detected acquired in the S1 to compiled application software to acquire real-time water quality data, setting a water quality parameter threshold value or selecting a national water quality standard threshold value according to the user, triggering a popup window and a short message alarm when a water quality index exceeds the threshold value, realizing the off-water real-time in-situ monitoring and the unattended high-frequency water quality monitoring of sailing, and assisting in early warning and handling sudden water quality events in time.
The foregoing illustrates and describes the principles, general features, and advantages of the present invention. It will be understood by those skilled in the art that the present invention is not limited to the embodiments described above, which are given by way of illustration of the principles of the present invention, but that various changes and modifications may be made without departing from the spirit and scope of the invention, and such changes and modifications are within the scope of the invention as claimed. The scope of the invention is defined by the appended claims and equivalents thereof.
Claims (9)
1. An off-water real-time intelligent remote sensing water quality monitoring method is characterized by comprising the following steps:
s1, acquiring hyperspectral remote sensing reflectivity data of a water body to be detected;
s2, collecting synchronous surface water quality samples of the water body to be detected, and constructing a water quality hyperspectral remote sensing reflectivity data set;
s3, constructing an optimal water quality intelligent inversion model based on the water quality hyperspectral remote sensing reflectivity data set;
and S4, obtaining real-time water quality data based on the hyperspectral remote sensing reflectivity data of the water body to be detected and the optimal water quality intelligent inversion model, and realizing the water leaving type real-time intelligent remote sensing water quality monitoring.
2. The method for monitoring the water quality by the intelligent water-leaving real-time remote sensing according to claim 1, wherein a spectrum acquisition system is adopted to acquire the hyperspectral data of the water body, and the spectrum acquisition system comprises: the system comprises an intelligent mobile terminal, a three-probe hyperspectral imager and a cloud deck;
the three-probe hyperspectral imager is connected with the intelligent mobile terminal, and the intelligent mobile terminal is connected with the holder through a telescopic buckle; the holder fixes an observation angle through an adjusting nut, and the observation angle ensures that the angle between the long side of the three-probe hyperspectral imager and the horizontal plane is 0 degree.
3. The method for monitoring the water quality by the water-leaving type real-time intelligent remote sensing according to claim 1, wherein the hyperspectral remote sensing reflectivity data comprises: and (3) water body hyperspectral remote sensing reflectivity data of different types of water bodies, water bodies with different water quality conditions and water bodies with different weather conditions.
4. The method for monitoring the water quality by the offshore real-time intelligent remote sensing according to claim 3, wherein the different types of water bodies comprise inland water bodies and clear ocean water bodies with different nutrition levels and different clarity degrees; the water bodies with different water quality conditions comprise water bodies with different wind wave sizes, different algal bloom outbreak conditions and different turbidity degrees; the different weather conditions include sunny days, cloudy days and light rains.
5. The method for monitoring the water quality by the offline real-time intelligent remote sensing according to claim 1, wherein the optimal water quality intelligent inversion model is constructed by the following steps:
constructing an optimal water quality intelligent inversion model by adopting various machine learning algorithms based on the water quality hyperspectral remote sensing reflectivity data set;
the multiple machine learning algorithms include a random forest algorithm, a neural network algorithm and a gaussian process regression algorithm.
6. The method for monitoring the water quality by the offline real-time intelligent remote sensing according to claim 5, wherein the intelligent inversion model of the optimal water quality is written into application software suitable for an intelligent mobile terminal, and the application software can call a calculation function, a storage function, a data display function and a downloading function of the intelligent mobile terminal.
7. The method for monitoring the water quality by the offline real-time intelligent remote sensing according to claim 1, wherein the real-time water quality data comprises: water quality parameters and environmental condition parameters;
the water quality parameters comprise: total nitrogen, total phosphorus, chlorophyll, transparency, suspended matter, permanganate index, turbidity, extinction coefficient, ammonia nitrogen, colored soluble organic matter absorption coefficient, soluble organic carbon, granular organic carbon and eutrophication index;
the environmental condition parameters include: the observation point longitude and latitude, the observation time, the air temperature, the observation angle, the weather and the wind speed.
8. The method for monitoring the water quality by the offline real-time intelligent remote sensing according to claim 1, further comprising preprocessing the hyperspectral remote sensing reflectivity data and the water sample to be detected;
the preprocessing of the hyperspectral remote sensing reflectivity data comprises the following steps: non-water body spectrum rejection, abnormal water body spectrum rejection and spectrum data normalization;
the pretreatment of the water sample to be detected comprises the following steps: suction filtration, low-temperature sample storage, laboratory measurement and water sample data abnormal value elimination.
9. The water-leaving type real-time intelligent remote sensing water quality monitoring method according to claim 1, characterized in that when the water-leaving type real-time intelligent remote sensing water quality monitoring is carried out, when the water quality index exceeds a specific threshold value, alarm reminding is carried out by changing display colors and popping up windows;
the specific threshold is a national water quality evaluation standard or a self-defined threshold.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211135571.7A CN115524294A (en) | 2022-09-19 | 2022-09-19 | Water leaving type real-time intelligent remote sensing water quality monitoring method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211135571.7A CN115524294A (en) | 2022-09-19 | 2022-09-19 | Water leaving type real-time intelligent remote sensing water quality monitoring method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115524294A true CN115524294A (en) | 2022-12-27 |
Family
ID=84697532
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211135571.7A Pending CN115524294A (en) | 2022-09-19 | 2022-09-19 | Water leaving type real-time intelligent remote sensing water quality monitoring method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115524294A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116223756A (en) * | 2023-03-27 | 2023-06-06 | 北京智科远达数据技术有限公司 | Method for generating water body nitrogen content prediction model |
| CN116482317A (en) * | 2023-04-26 | 2023-07-25 | 大连理工大学 | Lake water nutrition state real-time monitoring method, system, equipment and medium |
| CN116893146A (en) * | 2023-06-12 | 2023-10-17 | 华能澜沧江水电股份有限公司 | Method, device, equipment and storage medium for determining phosphorus concentration of water body particles |
| CN117434034A (en) * | 2023-10-24 | 2024-01-23 | 上海普适导航科技股份有限公司 | Quick inversion method for water quality multisource remote sensing data based on spectrum library |
| CN117805109A (en) * | 2023-12-29 | 2024-04-02 | 江苏腾丰环保科技有限公司 | Water quality detection method and system based on texture feature recognition |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107036974A (en) * | 2016-11-18 | 2017-08-11 | 中国水利水电科学研究院 | Inversion method is cooperateed with based on the water quality parameter multi-model that certainty set is modeled |
| CN107677646A (en) * | 2017-10-13 | 2018-02-09 | 中国水利水电科学研究院 | A kind of improvement DBPSO water quality parameter monitoring method and device |
| CN110672805A (en) * | 2019-10-08 | 2020-01-10 | 核工业北京地质研究院 | Reservoir water quality parameter quantitative inversion method based on aviation hyperspectral data |
| CN111504915A (en) * | 2020-04-27 | 2020-08-07 | 中国科学技术大学先进技术研究院 | Method, device and equipment for inverting chlorophyll concentration of water body and storage medium |
| CN114112941A (en) * | 2021-12-14 | 2022-03-01 | 江苏省地质勘查技术院 | Aviation hyperspectral water eutrophication evaluation method based on support vector regression |
| CN114758218A (en) * | 2022-04-25 | 2022-07-15 | 中交上海航道勘察设计研究院有限公司 | High-turbidity underwater topography inversion method suitable for hyperspectral satellite images |
-
2022
- 2022-09-19 CN CN202211135571.7A patent/CN115524294A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107036974A (en) * | 2016-11-18 | 2017-08-11 | 中国水利水电科学研究院 | Inversion method is cooperateed with based on the water quality parameter multi-model that certainty set is modeled |
| CN107677646A (en) * | 2017-10-13 | 2018-02-09 | 中国水利水电科学研究院 | A kind of improvement DBPSO water quality parameter monitoring method and device |
| CN110672805A (en) * | 2019-10-08 | 2020-01-10 | 核工业北京地质研究院 | Reservoir water quality parameter quantitative inversion method based on aviation hyperspectral data |
| CN111504915A (en) * | 2020-04-27 | 2020-08-07 | 中国科学技术大学先进技术研究院 | Method, device and equipment for inverting chlorophyll concentration of water body and storage medium |
| CN114112941A (en) * | 2021-12-14 | 2022-03-01 | 江苏省地质勘查技术院 | Aviation hyperspectral water eutrophication evaluation method based on support vector regression |
| CN114758218A (en) * | 2022-04-25 | 2022-07-15 | 中交上海航道勘察设计研究院有限公司 | High-turbidity underwater topography inversion method suitable for hyperspectral satellite images |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116223756A (en) * | 2023-03-27 | 2023-06-06 | 北京智科远达数据技术有限公司 | Method for generating water body nitrogen content prediction model |
| CN116223756B (en) * | 2023-03-27 | 2023-09-08 | 北京智科远达数据技术有限公司 | Method for generating water body nitrogen content prediction model |
| CN116482317A (en) * | 2023-04-26 | 2023-07-25 | 大连理工大学 | Lake water nutrition state real-time monitoring method, system, equipment and medium |
| CN116482317B (en) * | 2023-04-26 | 2023-10-27 | 大连理工大学 | Lake water nutrition state real-time monitoring method, system, equipment and medium |
| CN116893146A (en) * | 2023-06-12 | 2023-10-17 | 华能澜沧江水电股份有限公司 | Method, device, equipment and storage medium for determining phosphorus concentration of water body particles |
| CN116893146B (en) * | 2023-06-12 | 2024-03-29 | 华能澜沧江水电股份有限公司 | Method, device, equipment and storage medium for determining phosphorus concentration of water body particles |
| CN117434034A (en) * | 2023-10-24 | 2024-01-23 | 上海普适导航科技股份有限公司 | Quick inversion method for water quality multisource remote sensing data based on spectrum library |
| CN117805109A (en) * | 2023-12-29 | 2024-04-02 | 江苏腾丰环保科技有限公司 | Water quality detection method and system based on texture feature recognition |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN115524294A (en) | Water leaving type real-time intelligent remote sensing water quality monitoring method | |
| Zheng et al. | Remote sensing of diffuse attenuation coefficient patterns from Landsat 8 OLI imagery of turbid inland waters: A case study of Dongting Lake | |
| Clark et al. | Validation of atmospheric correction over the oceans | |
| Brewin et al. | Underway spectrophotometry along the Atlantic Meridional Transect reveals high performance in satellite chlorophyll retrievals | |
| Vos et al. | Multiplatform optical monitoring of eutrophication in temporally and spatially variable lakes | |
| CN114239422B (en) | Method for improving marine chlorophyll a concentration prediction accuracy based on machine learning | |
| CN103698750A (en) | HY-2 satellite scatterometer sea surface wind field retrieval method and device | |
| CN112070234A (en) | Ground-based remote sensing machine learning algorithm for chlorophyll and phycocyanin in water body under complex scene | |
| CN113420497B (en) | Remote sensing estimation method for total phosphorus concentration in turbid lake | |
| CN111079835B (en) | Himapari-8 atmospheric aerosol inversion method based on deep full-connection network | |
| Ye et al. | Atmospheric correction of Landsat-8/OLI imagery in turbid estuarine waters: A case study for the Pearl River estuary | |
| Mélin | Uncertainties in ocean colour remote sensing | |
| CN115078262A (en) | MODIS data-based long-term remote sensing monitoring method for chlorophyll a concentration in inland lake | |
| Guo et al. | An enhanced deep learning approach to assessing inland lake water quality and its response to climate and anthropogenic factors | |
| CN114705632A (en) | Method for estimating reservoir nutrition state index by satellite remote sensing reflectivity | |
| CN113780177A (en) | Non-contact real-time in-situ water quality monitoring method | |
| Men et al. | OC_3S: An optical classification and spectral scoring system for global waters using UV–visible remote sensing reflectance | |
| CN116563733A (en) | Water quality monitoring method based on hyperspectral and thermal infrared of unmanned aerial vehicle | |
| Li et al. | Monitoring Chlorophyll-a concentration in karst plateau lakes using Sentinel 2 imagery from a case study of pingzhai reservoir in Guizhou, China | |
| CN115950832A (en) | Method for inverting near-surface nitrogen dioxide concentration based on satellite data | |
| CN110988286B (en) | Intelligent water resource long-term detection system | |
| Wang et al. | Ammonia nitrogen monitoring of urban rivers with UAV-borne hyperspectral remote sensing imagery | |
| CN118090615A (en) | Inversion model comprehensive dynamic detection method for cross-medium nutrient elements in region | |
| LU500691B1 (en) | Non-touching Real-time In-situ Water Quality Detection Method | |
| Brando et al. | Reef Rescue Marine Monitoring Program: Using Remote Sensing for GBR Wide Water Quality: Final Report for 2008/09 Activities |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |