CN114993268A - Water depth inversion method and device combined with Catboost and storage medium - Google Patents

Water depth inversion method and device combined with Catboost and storage medium Download PDF

Info

Publication number
CN114993268A
CN114993268A CN202210385441.2A CN202210385441A CN114993268A CN 114993268 A CN114993268 A CN 114993268A CN 202210385441 A CN202210385441 A CN 202210385441A CN 114993268 A CN114993268 A CN 114993268A
Authority
CN
China
Prior art keywords
water depth
data
inversion
model
catboost
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
Application number
CN202210385441.2A
Other languages
Chinese (zh)
Inventor
谢涛
孔瑞瑶
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Nanjing University of Information Science and Technology
Original Assignee
Nanjing University of Information Science and Technology
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Nanjing University of Information Science and Technology filed Critical Nanjing University of Information Science and Technology
Priority to CN202210385441.2A priority Critical patent/CN114993268A/en
Publication of CN114993268A publication Critical patent/CN114993268A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • GPHYSICS
    • G01MEASURING; TESTING
    • G01CMEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
    • G01C13/00Surveying specially adapted to open water, e.g. sea, lake, river or canal
    • G01C13/008Surveying specially adapted to open water, e.g. sea, lake, river or canal measuring depth of open water
    • GPHYSICS
    • G01MEASURING; TESTING
    • G01BMEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
    • G01B11/00Measuring arrangements characterised by the use of optical techniques
    • G01B11/22Measuring arrangements characterised by the use of optical techniques for measuring depth
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F17/00Digital computing or data processing equipment or methods, specially adapted for specific functions
    • G06F17/10Complex mathematical operations
    • G06F17/18Complex mathematical operations for evaluating statistical data, e.g. average values, frequency distributions, probability functions, regression analysis
    • YGENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02ATECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
    • Y02A90/00Technologies having an indirect contribution to adaptation to climate change
    • Y02A90/30Assessment of water resources

Landscapes

  • Physics & Mathematics (AREA)
  • Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • Mathematical Physics (AREA)
  • Mathematical Optimization (AREA)
  • Theoretical Computer Science (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Pure & Applied Mathematics (AREA)
  • Computational Mathematics (AREA)
  • Mathematical Analysis (AREA)
  • Algebra (AREA)
  • Hydrology & Water Resources (AREA)
  • Radar, Positioning & Navigation (AREA)
  • Operations Research (AREA)
  • Probability & Statistics with Applications (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Evolutionary Biology (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Databases & Information Systems (AREA)
  • Software Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • Remote Sensing (AREA)
  • Image Processing (AREA)

Abstract

The invention discloses a water depth inversion method, a water depth inversion device and a water depth inversion storage medium which are combined with Catboost, belongs to the technical field of optical remote sensing, and comprises the following steps: acquiring multispectral remote sensing image data of a position to be detected and preprocessing the multispectral remote sensing image data; based on the preprocessed multispectral remote sensing image data, acquiring radiance data of different wave bands of a position to be detected by adopting a multi-value extraction method; inputting the radiance data into a prefabricated traditional water depth inversion model and a Catboost water depth inversion model, and respectively solving the water depth data of the position to be detected; and comparing, evaluating and analyzing the water depth data solved by the traditional water depth inversion model and the Catboost water depth inversion model, and solving the inversion value of the final water depth data. According to the invention, the traditional water depth inversion model and the Catboost water depth inversion model are combined, and water depth inversion is carried out at the same time, so that more accurate water depth data values can be obtained after comparison, evaluation and analysis.

Description

Water depth inversion method and device combined with Catboost and storage medium
Technical Field
The invention relates to a water depth inversion method and device combined with Catboost and a storage medium, and belongs to the technical field of optical remote sensing.
Background
Accurate water depth information provides powerful support for guaranteeing the safety of ships, protecting marine ecological environment and developing and utilizing marine resources.
The traditional water depth measuring method consumes time, material resources, manpower and financial resources, is not easy to carry out in remote water areas and shallow sea areas, and has high requirements on sea surface conditions.
In recent decades, the development of remote sensing technology has been rapid, and many researchers have proposed a water depth inversion method based on remote sensing images by using the relationship between the remote sensing images and the water depth.
In the multispectral remote sensing water depth inversion, the water depth inversion model is roughly divided into a theoretical analytical model and a semi-theoretical semi-empirical model.
When the theoretical analytical model and the semi-theoretical semi-empirical model are calculated, certain real physical data (such as a plurality of water body optical parameters) need to be acquired, so that the inversion process is complicated, the generalization capability of the model is weak, and the inversion accuracy is low.
Disclosure of Invention
The invention aims to overcome the defects of the prior art, provides a water depth inversion method, a water depth inversion device and a water depth inversion storage medium which are combined with a Catboost, and solves the problems that a certain practical physical data needs to be acquired during calculation by adopting a theoretical analytical model and a semi-theoretical semi-empirical model, so that the inversion process is complicated, the generalization capability of the model is weak, and the inversion accuracy is low in the prior art.
In order to achieve the above object, the present invention provides a water depth inversion method combining the castboost, comprising the following steps:
acquiring multispectral remote sensing image data of a position to be detected and preprocessing the multispectral remote sensing image data;
based on the preprocessed multispectral remote sensing image data, acquiring radiance data of different wave bands of a position to be detected by adopting a multi-value extraction method;
inputting the radiance data into a pre-trained traditional water depth inversion model and a pre-trained Catboost water depth inversion model, and respectively solving water depth data of a position to be detected;
and comparing, evaluating and analyzing the water depth data solved by the traditional water depth inversion model and the Catboost water depth inversion model, and solving the inversion value of the final water depth data.
Further, training the Catboost water depth inversion model comprises the following steps:
acquiring multispectral remote sensing image data of a plurality of sample positions and preprocessing the multispectral remote sensing image data;
acquiring radiance data of different wave bands of the position of a sample point by adopting a multi-value extraction method based on the preprocessed multispectral remote sensing image data of the sample position;
acquiring an actual measurement water depth value of each sample position, and constructing a data set corresponding to the actual measurement water depth value and the radiance data;
and selecting a Catboost model, taking the radiance data in the data set as an input value, taking the actually measured water depth value in the data set as an output value, training and adjusting parameters of the Catboost model, and obtaining a Catboost water depth inversion model for inverting the water depth data.
Further, the step of obtaining the measured water depth value of each sample position includes the following steps:
and acquiring water depth data of each sample position and tide data at the imaging moment of the remote sensing image data, carrying out tide correction on the water depth data according to the tide data, and taking the corrected water depth data as an actually measured water depth value.
Further, the conventional water depth inversion model includes: log-transformed ratio models, dual-band linear regression models, and multi-band linear regression models.
Further, the pre-processing comprises:
atmospheric correction is carried out on the multispectral remote sensing image data;
resampling the multispectral remote sensing image data after atmospheric correction;
and the atmospheric correction is used for acquiring the reflectivity of each wave band of the water body.
In a second aspect, the present invention provides a water depth inversion apparatus incorporating Catboost, the apparatus comprising:
the preprocessing module is used for preprocessing the multispectral remote sensing image data;
the radiation brightness acquisition module is used for acquiring radiation brightness data of a position to be detected;
the water depth inversion module is used for solving water depth data in an inversion mode and comprises a traditional model water depth inversion module and a Catboost model water depth inversion module;
and the precision correction module is used for carrying out precision correction on the solved water depth data.
Further, the traditional model water depth inversion module comprises a logarithmic conversion ratio model water depth inversion module, a dual-band linear regression model water depth inversion module and a multi-band linear regression model water depth inversion module.
In a third aspect, the invention provides a storage medium having stored thereon a computer program which, when executed by a processor, performs the steps of the method of any one of the first aspect.
The invention achieves the following beneficial effects:
compared with the traditional water depth inversion model, the water depth inversion is carried out by combining the CaBoost model, the water depth can be inverted by acquiring the multispectral remote sensing image data of the position to be detected, additional real physical data are not involved, the water depth inversion process can be effectively simplified, and the generalization capability and the inversion accuracy are improved.
Meanwhile, compared with the existing machine learning water depth inversion model, the CaBoost model solves the problems of gradient deviation and prediction deviation of the model, effectively prevents the occurrence of an over-fitting phenomenon, improves the algorithm accuracy and enhances the generalization capability effect;
according to the invention, the water depth inversion module adopts the Catboost and the traditional inversion water depth model to carry out inversion, and various models are subjected to precision comparison evaluation, so that more accurate and reliable water depth data values can be obtained.
Drawings
FIG. 1 is a flowchart of a method for water depth inversion in combination with Catboost according to an embodiment of the present invention;
FIG. 2 is a distribution of positions of a selected region to be measured and a sample in a water depth inversion method combined with Catboost according to an embodiment of the present invention;
FIG. 3 is a comparison diagram of the accuracy evaluation indexes of the models;
FIG. 4 is a scatter plot of inversion results of a log transformed model versus actual water depth values;
FIG. 5 is a scatter plot of the inversion results of a two-band linear regression model versus actual water depth values;
FIG. 6 is a scatter plot of inversion results of a multi-band linear regression model versus actual water depth values;
FIG. 7 is a scatter plot of inversion results of a Catboost water depth inversion model versus actual water depth values;
FIG. 8 is a residual normal distribution plot of the inversion results of the log-transformed ratio model;
FIG. 9 is a residual normal distribution plot of the inversion results of the dual band linear regression model;
FIG. 10 is a residual normal distribution plot of the inversion results of the multiband linear regression model;
FIG. 11 is a residual normal distribution diagram of inversion results of the Catboost water depth inversion model.
Detailed Description
The invention is further described below on the basis of the drawings. The following examples are only for illustrating the technical solutions of the present invention more clearly, and the protection scope of the present invention is not limited thereby.
The first embodiment is as follows:
the embodiment of the invention provides a water depth inversion method combined with Catboost, which is characterized in that the water depth data of a position to be detected is solved by inversion and subjected to precision comparison evaluation on the basis of the traditional water depth inversion method, so that more accurate water depth data values can be obtained, the problems of low inversion precision and weak model generalization capability can be effectively solved, and meanwhile, compared with the traditional water depth inversion model, the water depth inversion model of the Catboost solves the problems of gradient deviation and prediction deviation of the model, the generation of overfitting phenomenon can be effectively prevented, the algorithm accuracy can be effectively improved, and the generalization capability can be enhanced.
The embodiment of the invention provides a water depth inversion method combined with a Catboost model, which comprises the following steps as shown in figures 1 to 2:
the method comprises the following steps: acquiring multispectral remote sensing image data of a position to be detected and preprocessing the multispectral remote sensing image data:
in the embodiment of the invention, an area near the Wasabia is selected as an area to be measured to carry out inversion of water depth data, and multispectral remote sensing image data of a research area of the Wasabia is obtained through Sentinel-2;
atmospheric correction and resampling are carried out on collected multispectral remote sensing image data of the region to be detected: remote sensing data of a region to be measured is downloaded from an European Space Agency (ESA) to multi-spectral remote sensing image data of a sentinel level 2L 1C, the product is an atmospheric apparent reflectivity product which is only subjected to orthorectification and geometric fine rectification, and the L2A level data needs to be produced automatically. In the embodiment of the invention, a Sen2cor plug-in unit issued by the European Space Agency (ESA) is adopted to carry out atmospheric correction on L1C product data to obtain the real reflectivity of each wave band of a water body. In an embodiment of the present invention, because there are three different resolutions between the bands of Sentinel-2, the SNAP software is used to resample all bands to 10 m.
Step two: based on the preprocessed multispectral remote sensing image data, acquiring radiance data of different wave bands of a position to be detected by adopting a multi-value extraction method:
in the embodiment of the invention, based on preprocessed multispectral remote sensing image data, a multi-value extraction endpoint method is utilized in ArcGIS, and longitude and latitude coordinates are used for positioning to acquire the radiance data of each wave band of a position to be detected;
step three: and inputting the radiance data into a prefabricated traditional water depth inversion model and a Catboost water depth inversion model, and solving the water depth data of the position to be measured respectively.
Step four: and comparing, evaluating and analyzing the water depth data solved by the traditional water depth inversion model and the Catboost water depth inversion model, and solving the inversion value of the final water depth data.
In the third step of the invention, prefabricating a Catboost water depth inversion model comprises the following steps:
selecting sample positions in a research area of the Wa lake island, selecting the sample positions in a shallow water range of 0-20m in the embodiment of the invention, selecting 2000 sample positions for modeling, acquiring multispectral remote sensing image data of each sample position and preprocessing the multispectral remote sensing image data;
the process of preprocessing each sample location includes: atmospheric correction and resampling are carried out on the collected multispectral remote sensing image data of each sample position: remote sensing data of each sample position is downloaded from the European Space Agency (ESA) to the sentinel 2 # L1C level multispectral remote sensing image data, the product is an atmospheric apparent reflectivity product which is only subjected to orthorectification and geometric precise rectification, and L2A level data need to be produced by self. In the embodiment of the invention, a Sen2cor plug-in unit issued by the European Space Agency (ESA) is adopted to carry out atmospheric correction on L1C product data to obtain the real reflectivity of each wave band of a water body. In the embodiment of the invention, because three different resolutions exist among the bands of the Sentinel-2, the SNAP software is used for resampling all the bands to 10 m;
after preprocessing the multispectral remote sensing image of each sample position, positioning by longitude and latitude coordinates by using a multi-value extraction to point method in ArcGIS, and acquiring radiance data of each wave band of each sample position, wherein in the embodiment of the invention, the radiance values of 12 wave bands of each sample position are acquired;
acquiring actual measurement water depth data of each sample position, and constructing a corresponding data set based on the actual measurement water depth data of each sample position and the radiation brightness data of each sample position;
selecting a Catboost model, taking radiance data in a data set as input values, recording radiance values of 12 wave bands of each sample position as B1, B2, B3, B4, B5, B6, B7, B8, B8A, B9, B11 and B12 as input values, taking actual measurement water depth values in the data set as output values, training the Catboost model under default parameters, traversing a given parameter range by using a GridSearchCV function, and finding out parameters enabling the model to have the highest precision, wherein the parameter selection combination is as follows: the model is trained by using the group of parameters, and the obtained model is a Catboost water depth inversion model;
the method for acquiring the actually measured water depth value of each sample position comprises the following steps:
and acquiring water depth data of each sample position and tide data of the remote sensing image data imaging moment of each sample position, carrying out tide correction on the water depth value according to the tide data, and taking the corrected water depth value as an actually measured water depth value.
In the third step of the invention, prefabricating the traditional water depth inversion model comprises the following steps:
firstly, the traditional water depth inversion model comprises the following steps: a log-conversion ratio model, a dual-band linear regression model, and a multi-band linear regression model;
the prefabricated logarithm conversion ratio model comprises:
and (3) using SPSS software, and carrying out nonlinear fitting by introducing a blue-green waveband radiation brightness value and an actually measured water depth value and utilizing an L-M algorithm to obtain unknown parameters. The model formula is as follows:
Figure BDA0003594822510000061
wherein Z is water depth data obtained by inversion, m 1 、m 0 Is a regression coefficient; b (lambda) i )、R(λ j ) The reflectivity of blue and green wave bands respectively; n is a fixed coefficient of blue and green wave bands, and the blue and green wave bands are adopted because visible light (such as blue light and green light) with shorter wavelength has stronger penetrating power to a water body, so that the underwater topography condition can be reflected.
The prefabricated dual band linear regression model comprises:
presetting reflectivity ratios of 2 wave bands on different substrate types to be unchanged, establishing a dual-wave band linear regression model, using a linear fitting function in SPSS software, taking the radiation brightness value of a blue-green wave band of a modeling point as an independent variable, taking water depth data as a dependent variable, and using the linear fitting function to iterate to obtain unknown parameters of the model, wherein the model formula is as follows:
Figure BDA0003594822510000062
wherein Z is water depth data obtained by inversion, L (lambda) n ) Receiving λ for the sensor n Radiation value of band, n is band number, L sn ) Is λ n Radiation value of band deepwater zone, wherein 1 、λ 2 The wave bands are blue wave band and green wave band. Regression analysis calculation is carried out through part of actually measured water depth data, pixel values of the images and the selected average pixel value of the deep water area to obtain experience parameters a and b corresponding to a single wave band and multiple wave bands, and then the experience parameters are applied to other water body areas to invert the water depth.
The pre-formed multiband linear regression model comprises:
by utilizing a linear fitting function in SPSS software, the radiation brightness values of 12 wave bands of each sample position used for modeling are used as independent variables, the actually measured water depth value is used as a dependent variable, iteration is carried out by utilizing the linear fitting function, and unknown parameters of the model can be obtained, wherein the model formula is as follows:
Figure BDA0003594822510000071
wherein Z is water depth data obtained by inversion, a 1 、a 2 …a N B is a model coefficient; i is 1, … N, N is the number of wave bands participating in calculation; l (lambda) i ) Receiving λ for the sensor i Radiation values of the bands; l is si ) Is λ i The radiation value of the wave band deep water area.
Example two:
the second embodiment of the invention provides a water depth inversion device combined with Catboost, which comprises:
the preprocessing module is used for preprocessing the multispectral remote sensing image data;
the radiation brightness acquisition module is used for acquiring radiation brightness data of a position to be detected;
the water depth inversion module is used for inverting and solving water depth data and comprises a traditional model water depth inversion module and a Catboost model water depth inversion module;
and the precision correction module is used for carrying out precision correction on the solved water depth data.
The traditional model water depth inversion module comprises a logarithmic conversion ratio model water depth inversion module, a dual-waveband linear regression model water depth inversion module and a multiband linear regression model water depth inversion module.
Example three:
a third embodiment of the present invention provides a storage medium, and the program implements the steps of the method according to any one of the embodiments when executed by a processor.
In order to verify the accuracy of the Catboost water depth inversion model in the invention, as shown in fig. 3 to 11, the Catboost water depth inversion model and the traditional water depth inversion model are compared and analyzed, the accuracy evaluation is carried out by using a decision coefficient (R ^2), a Root Mean Square Error (RMSE), a Mean Absolute Error (MAE) and a Mean Relative Error (MRE) as accuracy evaluation indexes, fig. 3 is the accuracy evaluation result of the Catboost water depth inversion model and the traditional water depth inversion model, a distribution scatter diagram of the water depth inversion values obtained by the models shown in fig. 4 and fig. 7 and the actually measured water depth values obtained by actual measurement is drawn, and a residual distribution histogram shown in fig. 8 to fig. 11 is drawn.
As shown in fig. 3 to 11, compared with the existing water depth inversion method, the water depth inversion method using the castboost water depth inversion model can effectively solve the problems of poor fitting effect and low precision of the traditional water depth inversion model, and the castboost water depth inversion model has significant effect and excellent advantages for solving the problem of high nonlinearity between water depth data and radiation brightness value in water depth inversion; therefore, the traditional water depth inversion model and the Catboost water depth inversion model are combined, water depth inversion is carried out simultaneously, more accurate water depth data values can be obtained after comparison, evaluation and analysis, and meanwhile, the reliability of inversion results can be effectively improved by adopting a mode of combining various inversion methods to carry out water depth inversion.
The above description is only a preferred embodiment of the present invention, and it should be noted that, for those skilled in the art, several modifications and variations can be made without departing from the technical principle of the present invention, and these modifications and variations should also be regarded as the protection scope of the present invention.

Claims (8)

1. A water depth inversion method combined with Catboost is characterized in that: the method comprises the following steps:
acquiring multispectral remote sensing image data of a position to be detected and preprocessing the multispectral remote sensing image data;
based on the preprocessed multispectral remote sensing image data, acquiring radiance data of different wave bands of a position to be detected by adopting a multi-value extraction method;
inputting the radiance data into a pre-trained traditional water depth inversion model and a pre-trained Catboost water depth inversion model, and respectively solving water depth data of a position to be detected;
and comparing, evaluating and analyzing the water depth data solved by the traditional water depth inversion model and the Catboost water depth inversion model, and solving the inversion value of the final water depth data.
2. The water depth inversion method combining Catboost as claimed in claim 1, wherein: the method for training the water depth inversion model of the Catboost comprises the following steps:
acquiring multispectral remote sensing image data of a plurality of sample positions and preprocessing the multispectral remote sensing image data;
acquiring radiance data of different wave bands of the position of a sample point by adopting a multi-value extraction method based on the preprocessed multispectral remote sensing image data of the sample position;
acquiring an actual measurement water depth value of each sample position, and constructing a data set corresponding to the actual measurement water depth value and the radiance data;
and selecting a Catboost model, taking the radiance data in the data set as an input value, taking the actually measured water depth value in the data set as an output value, training and adjusting parameters of the Catboost model, and obtaining a Catboost water depth inversion model for inverting the water depth data.
3. The water depth inversion method combining Catboost as claimed in claim 2, wherein: the method for acquiring the actually measured water depth value of each sample position comprises the following steps:
and acquiring water depth data of each sample position and tide data at the imaging moment of the remote sensing image data, carrying out tide correction on the water depth data according to the tide data, and taking the corrected water depth data as an actually measured water depth value.
4. The water depth inversion method combining Catboost as claimed in claim 1, wherein: the conventional water depth inversion model comprises: log-transformed ratio models, dual-band linear regression models, and multi-band linear regression models.
5. The water depth inversion method combining Catboost as claimed in claim 2, wherein: the pretreatment comprises the following steps:
atmospheric correction is carried out on the multispectral remote sensing image data;
resampling the multispectral remote sensing image data after atmospheric correction;
the atmospheric correction is used for acquiring the reflectivity of each wave band of the water body.
6. A water depth inversion device combined with Catboost is characterized in that: the device comprises:
the preprocessing module is used for preprocessing the multispectral remote sensing image data;
the radiation brightness acquisition module is used for acquiring radiation brightness data of a position to be detected;
the water depth inversion module is used for solving water depth data in an inversion mode and comprises a traditional model water depth inversion module and a Catboost model water depth inversion module;
and the precision correction module is used for carrying out precision correction on the solved water depth data.
7. The water depth inversion device combined with Catboost according to claim 6, wherein: the traditional model water depth inversion module comprises a logarithmic conversion ratio model water depth inversion module, a dual-waveband linear regression model water depth inversion module and a multiband linear regression model water depth inversion module.
8. A storage medium on which a computer program is stored which, when being executed by a processor, carries out the steps of the method of any one of claims 1 to 5.
CN202210385441.2A 2022-04-13 2022-04-13 Water depth inversion method and device combined with Catboost and storage medium Pending CN114993268A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202210385441.2A CN114993268A (en) 2022-04-13 2022-04-13 Water depth inversion method and device combined with Catboost and storage medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202210385441.2A CN114993268A (en) 2022-04-13 2022-04-13 Water depth inversion method and device combined with Catboost and storage medium

Publications (1)

Publication Number Publication Date
CN114993268A true CN114993268A (en) 2022-09-02

Family

ID=83023811

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202210385441.2A Pending CN114993268A (en) 2022-04-13 2022-04-13 Water depth inversion method and device combined with Catboost and storage medium

Country Status (1)

Country Link
CN (1) CN114993268A (en)

Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112213287A (en) * 2020-12-07 2021-01-12 速度时空信息科技股份有限公司 Coastal beach salinity inversion method based on remote sensing satellite image
CN112525831A (en) * 2020-11-23 2021-03-19 淮阴师范学院 Remote sensing inversion model and method for protein nitrogen accumulation of rice overground part based on Catboost regression algorithm
CN113051817A (en) * 2021-03-19 2021-06-29 上海海洋大学 Sea wave height prediction method based on deep learning and application thereof
CN113203694A (en) * 2021-04-26 2021-08-03 中国科学院东北地理与农业生态研究所 MSI lake eutrophication index remote sensing estimation method
CN113514833A (en) * 2021-04-25 2021-10-19 南京信息工程大学 Sea surface arbitrary point wave direction inversion method based on sea wave image
CN113793374A (en) * 2021-09-01 2021-12-14 自然资源部第二海洋研究所 Method for inverting water depth based on water quality inversion result by using improved four-waveband remote sensing image QAA algorithm
CN114117886A (en) * 2021-10-28 2022-03-01 南京信息工程大学 Water depth inversion method for multispectral remote sensing
CN114139437A (en) * 2021-10-19 2022-03-04 中国海洋大学 Method and system for inverting submarine topography by using satellite height measurement data
CN114201732A (en) * 2021-11-24 2022-03-18 中国人民解放军92859部队 Sentinel-2A image-based shallow sea water depth inversion method

Patent Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112525831A (en) * 2020-11-23 2021-03-19 淮阴师范学院 Remote sensing inversion model and method for protein nitrogen accumulation of rice overground part based on Catboost regression algorithm
CN112213287A (en) * 2020-12-07 2021-01-12 速度时空信息科技股份有限公司 Coastal beach salinity inversion method based on remote sensing satellite image
CN113051817A (en) * 2021-03-19 2021-06-29 上海海洋大学 Sea wave height prediction method based on deep learning and application thereof
CN113514833A (en) * 2021-04-25 2021-10-19 南京信息工程大学 Sea surface arbitrary point wave direction inversion method based on sea wave image
CN113203694A (en) * 2021-04-26 2021-08-03 中国科学院东北地理与农业生态研究所 MSI lake eutrophication index remote sensing estimation method
CN113793374A (en) * 2021-09-01 2021-12-14 自然资源部第二海洋研究所 Method for inverting water depth based on water quality inversion result by using improved four-waveband remote sensing image QAA algorithm
CN114139437A (en) * 2021-10-19 2022-03-04 中国海洋大学 Method and system for inverting submarine topography by using satellite height measurement data
CN114117886A (en) * 2021-10-28 2022-03-01 南京信息工程大学 Water depth inversion method for multispectral remote sensing
CN114201732A (en) * 2021-11-24 2022-03-18 中国人民解放军92859部队 Sentinel-2A image-based shallow sea water depth inversion method

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
肖桂林: "基于Sentinel卫星的福建省近海岸岸叶绿素a浓度遥感反演研究", pages 027 - 590 *
谢佳标: "Keras深度学习:入门、实践与进阶", vol. 1, 机械工业出版社, pages: 2 - 5 *

Similar Documents

Publication Publication Date Title
CN109059796B (en) Shallow sea water depth multispectral satellite remote sensing inversion method for water depth control point-free area
CN110749568A (en) MODIS remote sensing inversion method for phytoplankton absorption characteristics of high-turbidity eutrophic lakes
CN109781073B (en) Shallow sea water depth remote sensing extraction method integrating sea wave characteristics and spectral characteristics
CN112712071A (en) Crop growth detection method and related equipment thereof
CN111781146A (en) Wave parameter inversion method using high-resolution satellite optical image
Xia et al. A bathymetry mapping approach combining log-ratio and semianalytical models using four-band multispectral imagery without ground data
CN112013822A (en) Multispectral remote sensing water depth inversion method based on improved GWR model
CN111008664B (en) Hyperspectral sea ice detection method based on space-spectrum combined characteristics
CN114201732A (en) Sentinel-2A image-based shallow sea water depth inversion method
CN111561916B (en) Shallow sea water depth uncontrolled extraction method based on four-waveband multispectral remote sensing image
CN115856925A (en) Multispectral remote sensing image water depth inversion method, medium and equipment based on chart data
CN110186851A (en) It is a kind of based on the semi-supervised Hyperspectral imaging heavy metal-polluted soil concentration evaluation method from Coded Analysis
CN116295285A (en) Shallow sea water depth remote sensing inversion method based on region self-adaption
Wang et al. Sea ice classification with convolutional neural networks using Sentinel-L scansar images
CN115761463A (en) Shallow sea water depth inversion method, system, equipment and medium
Li et al. Satellite-derived bathymetry integrating spatial and spectral information of multispectral images
CN115979972B (en) Real-time monitoring method and system for hyperspectral of crude oil film on sea surface
CN114993268A (en) Water depth inversion method and device combined with Catboost and storage medium
Cissé et al. A new deep learning method for multispectral image time series completion using hyperspectral data
Alboody et al. DeepSen3: Deep multi-scale learning model for spatial-spectral fusion of Sentinel-2 and Sentinel-3 remote sensing images
Aitken et al. Prelude to CZMIL: seafloor imaging and classification results achieved with CHARTS and the Rapid Environmental Assessment (REA) Processor
CN112668670A (en) Mulch film identification method and related equipment
CN111368261A (en) Quantitative and qualitative description method for impervious surface index based on atmospheric correction
Bachmann et al. Modeling coastal waters from hyperspectral imagery using manifold coordinates
Wozencraft et al. Use of SHOALS data to produce spectrally-derived depths in Kaneohe Bay, Hawaii

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