CN115080905B - Remote sensing inversion method for chlorophyll a concentration in plateau lake - Google Patents
Remote sensing inversion method for chlorophyll a concentration in plateau lake Download PDFInfo
- Publication number
- CN115080905B CN115080905B CN202210639482.XA CN202210639482A CN115080905B CN 115080905 B CN115080905 B CN 115080905B CN 202210639482 A CN202210639482 A CN 202210639482A CN 115080905 B CN115080905 B CN 115080905B
- Authority
- CN
- China
- Prior art keywords
- chlorophyll
- index
- concentration
- inversion
- band
- 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.)
- Active
Links
- ATNHDLDRLWWWCB-AENOIHSZSA-M chlorophyll a Chemical compound C1([C@@H](C(=O)OC)C(=O)C2=C3C)=C2N2C3=CC(C(CC)=C3C)=[N+]4C3=CC3=C(C=C)C(C)=C5N3[Mg-2]42[N+]2=C1[C@@H](CCC(=O)OC\C=C(/C)CCC[C@H](C)CCC[C@H](C)CCCC(C)C)[C@H](C)C2=C5 ATNHDLDRLWWWCB-AENOIHSZSA-M 0.000 title claims abstract description 162
- 229930002868 chlorophyll a Natural products 0.000 title claims abstract description 161
- 238000000034 method Methods 0.000 title claims abstract description 31
- 238000002310 reflectometry Methods 0.000 claims abstract description 73
- 238000001228 spectrum Methods 0.000 claims abstract description 43
- 238000003066 decision tree Methods 0.000 claims abstract description 18
- 230000003287 optical effect Effects 0.000 claims abstract description 9
- 238000012417 linear regression Methods 0.000 claims abstract description 8
- 238000012937 correction Methods 0.000 claims abstract description 7
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims description 47
- 230000031700 light absorption Effects 0.000 claims description 32
- 230000003595 spectral effect Effects 0.000 claims description 32
- 238000010521 absorption reaction Methods 0.000 claims description 16
- 230000014509 gene expression Effects 0.000 claims description 16
- 210000000746 body region Anatomy 0.000 claims description 5
- 238000007637 random forest analysis Methods 0.000 claims description 5
- 238000004364 calculation method Methods 0.000 claims description 4
- 230000001419 dependent effect Effects 0.000 claims description 2
- 238000012544 monitoring process Methods 0.000 description 9
- 238000005259 measurement Methods 0.000 description 4
- 238000000605 extraction Methods 0.000 description 3
- 238000012625 in-situ measurement Methods 0.000 description 3
- CSCPPACGZOOCGX-UHFFFAOYSA-N Acetone Chemical compound CC(C)=O CSCPPACGZOOCGX-UHFFFAOYSA-N 0.000 description 2
- 238000012851 eutrophication Methods 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 235000016709 nutrition Nutrition 0.000 description 2
- 230000035764 nutrition Effects 0.000 description 2
- 230000000737 periodic effect Effects 0.000 description 2
- 238000005070 sampling Methods 0.000 description 2
- 238000012549 training Methods 0.000 description 2
- 239000002028 Biomass Substances 0.000 description 1
- 241000192700 Cyanobacteria Species 0.000 description 1
- 238000002835 absorbance Methods 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004422 calculation algorithm Methods 0.000 description 1
- 238000005119 centrifugation Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 229930002875 chlorophyll Natural products 0.000 description 1
- 235000019804 chlorophyll Nutrition 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000004880 explosion Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 239000000049 pigment Substances 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 230000001360 synchronised effect Effects 0.000 description 1
- 238000012876 topography Methods 0.000 description 1
- 238000001429 visible spectrum Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/18—Complex mathematical operations for evaluating statistical data, e.g. average values, frequency distributions, probability functions, regression analysis
-
- 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
- G01N21/17—Systems in which incident light is modified in accordance with the properties of the material investigated
- G01N21/55—Specular reflectivity
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
Landscapes
- Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Mathematical Physics (AREA)
- Pure & Applied Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Computational Mathematics (AREA)
- Mathematical Analysis (AREA)
- Mathematical Optimization (AREA)
- Health & Medical Sciences (AREA)
- Biochemistry (AREA)
- Pathology (AREA)
- Algebra (AREA)
- Immunology (AREA)
- Databases & Information Systems (AREA)
- Software Systems (AREA)
- General Engineering & Computer Science (AREA)
- General Health & Medical Sciences (AREA)
- Analytical Chemistry (AREA)
- Chemical & Material Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Spectroscopy & Molecular Physics (AREA)
- Probability & Statistics with Applications (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Operations Research (AREA)
- Investigating Or Analysing Materials By Optical Means (AREA)
Abstract
The invention discloses a remote sensing inversion method for chlorophyll a concentration in a plateau lake, which comprises the following steps: respectively obtaining the chlorophyll a concentration of the plateau lake and the remote sensing reflectivity data of the plateau lake: performing atmospheric correction on the remote sensing reflectivity data of the plateau lake to obtain the reflectivity data of the bottom layer of the atmosphere; respectively constructing chlorophyll a inversion spectrum indexes according to the inherent optical characteristics of the chlorophyll a and the reflectivity data of the bottom layer of the atmosphere; establishing a decision tree model according to the chlorophyll a inversion spectrum index, and determining the spectrum index with the greatest influence on the chlorophyll a concentration result as a preferred spectrum index; performing linear regression on the optimal spectrum index and the concentration of chlorophyll a in the plateau lake, and establishing a chlorophyll a inversion model; and according to the chlorophyll a inversion model and the atmospheric bottom layer reflectivity data, remote sensing inversion of the chlorophyll a concentration of the plateau lake is realized. The method can rapidly and optimally select the most effective model for inverting the chlorophyll a concentration, and further improves the chlorophyll a inversion accuracy.
Description
Technical Field
The invention relates to the technical field of inversion of chlorophyll a concentration in lakes, in particular to a remote sensing inversion method of chlorophyll a concentration in a plateau lake.
Background
The eutrophication condition of water quality in plateau lakes is a topic of concern in the current society, wherein chlorophyll a is one of main components of phytoplankton, and is an important index for reflecting the nutrition condition of inland water bodies and monitoring the explosion of cyanobacteria bloom. Monitoring the distribution condition of chlorophyll a concentration in the lake is helpful for measuring phytoplankton biomass and evaluating the nutrition state of the water body. The plateau lake is affected by the topography factors, the traditional lake sampling monitoring method is not suitable, and along with the deep research on the spectral characteristics of the water body and the improvement of the lake chlorophyll a concentration inversion model, the lake chlorophyll a concentration information can be accurately obtained through a remote sensing means, and the eutrophication water body and the water quality change trend can be effectively found.
The method for effectively extracting the chlorophyll a concentration in the plateau lake mainly comprises two methods: the method is characterized in that a water body sample of the plateau lake is brought back to a laboratory for extraction through manual in-situ measurement, the method is time-consuming and high in cost, and the chlorophyll a concentration information of the lake in a large range can not be reflected by taking points and areas; another common chlorophyll a extraction method is to select a corresponding wave band of the reflectivity of a remote sensing satellite to fit based on the lake spectrum measurement according to the characteristics of the measured spectrum information, so as to achieve the chlorophyll a inversion effect.
Therefore, on the basis of the existing lake chlorophyll a concentration inversion technology, how to overcome the unilateral performance of manual field measurement, solve the difficulty of altimeter lake spectrum, improve the working efficiency and inversion accuracy, and become the problem to be solved by the technicians in the field.
Disclosure of Invention
In view of the problems, the invention provides a remote sensing inversion method for the chlorophyll a concentration of the plateau lake, which at least solves the technical problems of the part, and the method not only overcomes the unilateral property of manual in-situ measurement, but also solves the difficulty of actually measuring the spectrum of the plateau lake, and effectively improves the working efficiency and inversion precision.
The embodiment of the invention provides a remote sensing inversion method for chlorophyll a concentration in a plateau lake, which comprises the following steps:
S1, respectively obtaining the chlorophyll a concentration of the plateau lake and remote sensing reflectivity data of the plateau lake; the concentration of chlorophyll a in the plateau lake is obtained from the ground measuring point of the plateau lake;
S2, performing atmospheric correction on the remote sensing reflectivity data of the plateau lake to obtain the reflectivity data of the bottom layer of the atmosphere;
s3, respectively constructing chlorophyll a inversion spectrum indexes according to the inherent optical characteristics of the chlorophyll a and the reflectivity data of the bottom layer of the atmosphere; the chlorophyll a inversion spectral index includes: single band index, ratio index, normalized chlorophyll a index, and three band index;
S4, establishing a decision tree model according to the chlorophyll a inversion spectrum index, and determining the spectrum index with the greatest influence on the chlorophyll a concentration result as a preferred spectrum index;
s5, carrying out linear regression on the optimized spectral index and the concentration of chlorophyll a in the plateau lake, and establishing a chlorophyll a inversion model to obtain a correlation coefficient of a regression model;
and S6, according to the chlorophyll a inversion model and the atmospheric bottom layer reflectivity data, remote sensing inversion of the chlorophyll a concentration of the plateau lake is realized.
Further, the step S2 further includes: and extracting the plateau lake water body region by adopting an improved normalized difference water body index according to the green light wave band and the short wave infrared wave band in the atmospheric bottom reflectivity data.
Further, the plateau lake water body region is extracted by the following formula:
In the above formula MNDWI represents an improved normalized difference water index; ρ Green represents the green band reflectivity in the atmospheric bottom reflectivity data; ρ SWIR represents the short wave infrared band reflectivity in the atmospheric subsurface reflectivity data.
Further, in the step S3, a red light absorption peak band is selected according to the reflectivity data of the bottom layer of the atmosphere, and a single band index is constructed.
Further, in the step S3, according to the reflectivity data of the bottom layer of the atmosphere, a red light absorption peak band and a fluorescence peak band are selected, and a ratio index is constructed; by comparing the red light absorption peak wave band with the fluorescence peak wave band, the difference between chlorophyll a absorption trough and reflection peak is enlarged.
Further, in the step S3, a red light absorption peak band and a fluorescence peak band are selected according to the reflectivity data of the bottom layer of the atmosphere; the reflectivity contrast of the selected red light absorption peak wave band and the fluorescence peak wave band is further enhanced through a nonlinear stretching mode, and the normalized chlorophyll a index is constructed.
Further, in the step S3, according to the inherent optical characteristics of chlorophyll a and the reflectivity data of the bottom layer of the atmosphere, a red light absorption peak wave band, a fluorescence peak wave band and an absorption peak wave band of the pure water body under ideal conditions are respectively selected, so as to construct a three-wave-band index; the selected fluorescence peak band is adjacent to the selected red light absorption peak band.
Further, the step S5 further includes: and constructing a linear expression formed by the chlorophyll a inversion spectral indexes and the chlorophyll a concentration of the plateau lake according to the water reflection characteristics of the plateau lake and the reflectivity data of the bottom layer of the atmosphere.
Further, the linear expression formed by the single-band index and the concentration of chlorophyll a in the plateau lake is as follows:
Cchl-a=A+B·ρa
In the above formula, C chl-a represents the chlorophyll a concentration in the plateau lake; A. b is a first correlation coefficient of the regression model; ρ a represents the water body reflectivity of the red light absorption peak band.
Further, the linear expression formed by the ratio index and the concentration of chlorophyll a in the plateau lake is as follows:
in the above formula, C chl-a represents the chlorophyll a concentration in the plateau lake; C. d is a second correlation coefficient of the regression model; ρ a and ρ b represent the water body reflectivities of the red absorption peak band and the fluorescence peak band, respectively.
Further, the linear expression formed by the normalized chlorophyll a index and the concentration of chlorophyll a in the plateau lake is as follows:
In the above formula, C chl-a represents the chlorophyll a concentration in the plateau lake; J. k is a third correlation coefficient of the regression model; ρ a and ρ b represent the water body reflectivities of the red absorption peak band and the fluorescence peak band, respectively.
Further, the linear expression formed by the three-band index and the concentration of chlorophyll a in the plateau lake is as follows:
Cchl-a=M+N·(ρa -1-ρb -1)·ρc
In the above formula, C chl-a represents the chlorophyll a concentration in the plateau lake; m, N is the fourth correlation coefficient of the regression model; ρ a -1 and ρ b -1 represent the inverse of the water body reflectivity of the red light absorption peak band and the fluorescence peak band, respectively; ρ c represents the water body reflectivity of the absorption peak band.
Further, the step S4 includes:
establishing a decision tree model according to the chlorophyll a inversion spectrum index, selecting corresponding out-of-bag data to calculate out-of-bag data errors, and generating a first error;
Adding random noise, respectively changing the value of the data outside the bag corresponding to each chlorophyll a inversion spectrum index, and calculating the data error outside the bag again to generate a second error;
According to the first error and the second error, respectively calculating the characteristic importance of each chlorophyll-a inversion spectral index, and determining the spectral index corresponding to the highest characteristic importance;
And taking the spectral index with the highest characteristic importance as the spectral index with the greatest influence on the chlorophyll a concentration result as the optimal spectral index.
Further, the feature importance of each chlorophyll a inversion spectral index is calculated by the following formula:
In the above formula, V represents feature importance; err 1 represents the first error; err 2 represents the second error; n represents the number of decision trees.
The technical scheme provided by the embodiment of the invention has the beneficial effects that at least:
The remote sensing inversion method for chlorophyll a concentration in the plateau lake provided by the embodiment of the invention comprises the following steps: respectively obtaining the chlorophyll a concentration of the plateau lake and the remote sensing reflectivity data of the plateau lake: performing atmospheric correction on the remote sensing reflectivity data of the plateau lake to obtain the reflectivity data of the bottom layer of the atmosphere; respectively constructing chlorophyll a inversion spectrum indexes according to the inherent optical characteristics of the chlorophyll a and the reflectivity data of the bottom layer of the atmosphere; establishing a decision tree model according to the chlorophyll a inversion spectrum index, and determining the spectrum index with the greatest influence on the chlorophyll a concentration result as a preferred spectrum index; performing linear regression on the optimal spectrum index and the concentration of chlorophyll a in the plateau lake, and establishing a chlorophyll a inversion model; and according to the chlorophyll a inversion model and the atmospheric bottom layer reflectivity data, remote sensing inversion of the chlorophyll a concentration of the plateau lake is realized. The method can effectively overcome the unilateral property of manual in-situ measurement, solves the difficulty of actually measuring the spectrum of the plateau lake, and can realize rapid, accurate and large-area-range monitoring of the plateau lake. Based on feature selection, various inversion spectral indexes can be compared, a most effective model for inverting chlorophyll a concentration is rapidly optimized, and chlorophyll a inversion accuracy is further improved.
Additional features and advantages of the invention will be set forth in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. The objectives and other advantages of the invention will be realized and attained by the structure particularly pointed out in the written description and claims thereof as well as the appended drawings.
The technical scheme of the invention is further described in detail through the drawings and the embodiments.
Drawings
The accompanying drawings are included to provide a further understanding of the invention and are incorporated in and constitute a part of this specification, illustrate the invention and together with the embodiments of the invention, serve to explain the invention. In the drawings:
FIG. 1 is a flow chart of a remote sensing inversion method for chlorophyll a concentration in a plateau lake provided by the embodiment of the invention;
fig. 2 is a general flow chart of an inversion method according to an embodiment of the present invention.
Detailed Description
Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be embodied in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.
The embodiment of the invention provides a remote sensing inversion method for chlorophyll a concentration in a plateau lake, which is shown by referring to FIG. 1 and comprises the following steps:
S1, respectively obtaining the chlorophyll a concentration of the plateau lake and remote sensing reflectivity data of the plateau lake; the chlorophyll a concentration of the plateau lake is obtained from the ground measuring point of the plateau lake; the remote sensing reflectivity data of the plateau lake are obtained from remote sensing satellites;
s2, performing atmospheric correction on the remote sensing reflectivity data of the plateau lake to obtain the reflectivity data of the bottom layer of the atmosphere;
S3, respectively constructing chlorophyll a inversion spectral indexes according to the inherent optical characteristics of the chlorophyll a and the reflectivity data of the bottom layer of the atmosphere; chlorophyll a inversion spectral index includes: single band index, ratio index, normalized chlorophyll a index, and three band index;
S4, inverting the spectrum index according to chlorophyll a, establishing a decision tree model, and determining the spectrum index with the greatest influence on the concentration result of chlorophyll a as a preferred spectrum index;
S5, carrying out linear regression on the optimized spectral index and the concentration of chlorophyll a in the plateau lake, and establishing a chlorophyll a inversion model to obtain a correlation coefficient of the regression model;
and S6, according to the chlorophyll a inversion model and the reflectivity data of the bottom layer of the atmosphere, remote sensing inversion of the chlorophyll a concentration of the plateau lake is realized.
According to the remote sensing inversion method for the chlorophyll a concentration of the plateau lake, the method is applicable to the plateau lake, field measurement is not needed under the condition that the spectrum data of the plateau lake is difficult to obtain, an optimal model for inverting the chlorophyll a of the plateau lake can be obtained rapidly only through comparison of characteristic wave bands, large-scale and periodic observation of the plateau lake is achieved, material resources and financial resources are saved, and the method has important significance for intelligent monitoring and management of the plateau lake.
The remote sensing inversion method for chlorophyll a concentration in the plateau lake is specifically described below, and reference may be made to fig. 2, where fig. 2 is a general flow diagram of the inversion method provided in this embodiment:
Step one, acquiring chlorophyll a concentration data of a ground measuring point plateau lake and synchronous Sentinel-2MSI remote sensing reflectivity data:
1) Extracting pigments from the acquired plateau lake water body sample of the ground measuring point by using acetone; measuring absorbance at a specific wavelength by using a laboratory spectrophotometer according to absorption of chlorophyll extraction solution to visible spectrum by centrifugation;
2) Extracting the chlorophyll a content of the plateau lake in the lake sample according to an empirical formula provided by the existing experimental principle;
3) Sentinel-2MSI remote sensing reflectance data (satellite data) is downloaded through the United States Geological Survey (USGS) official website.
And secondly, performing atmospheric correction on the Sentinel-2MSI remote sensing reflectivity data to obtain the atmospheric bottom reflectivity data of the Sentinel-2MSI after the atmospheric correction.
Step three, extracting the plateau lake region through the atmospheric corrected Sentinel-2MSI atmospheric bottom layer reflectivity data:
Since the Sentinel-2MSI atmospheric bottom reflectance data, including "water" and "non-water portions" (buildings, land, etc.), is required to extract the plateau lake "water" region, i.e., the plateau lake region.
For the water body information of the plateau and the lake, green light wave bands (with the center wavelength of 560 nm) and short wave infrared wave bands (with the center wavelength of 1610 nm) in satellite data are utilized for calculation, and an improved normalized difference water body index (MNDWI) is adopted for extracting the water body region of the plateau and the lake. MNDWI, the calculation result is larger than 0, and the result is a plateau lake water body region; less than or equal to 0 is vegetation, land, and other non-water areas. The formula is expressed as follows:
Where MNDWI denotes the modified normalized difference water index, ρ Green denotes the green band reflectivity, ρ SWIR denotes the short wave infrared band reflectivity.
Constructing a chlorophyll a inversion spectrum index (comprising a single-band index, a ratio index, a normalized chlorophyll a index and a three-band index) based on the reflection characteristics of the water body of the plateau lake, the inherent optical characteristics of the chlorophyll a and the reflectivity data of the bottom layer of the Sentinel-2MSI atmosphere, and a linear expression formed by each chlorophyll a inversion spectrum index and the concentration of chlorophyll a:
1) Constructing a single-band index:
The position of the reflection peak or absorption valley of the chlorophyll a reflection spectrum is mainly related to the red light absorption peak wave band of Sentinel-2MSI, the selected wave band is marked as a wave band (the red light absorption peak wave band, the center wavelength is 665 nm), and a single wave band index is constructed. The linear expression consisting of the single band index and chlorophyll a concentration is:
Cchl-a=A+B·ρa (2)
wherein C chl-a represents the chlorophyll a concentration in the plateau lake; A. b is a first correlation coefficient of the regression model; ρ a is the water body reflectance of the selected a-band (selected red light absorption peak band).
2) Constructing a ratio index:
The positions of the reflection peaks or absorption valleys of the chlorophyll a reflection spectrum are mainly related to red light absorption peak wave bands and fluorescence peak wave bands of Sentinel-2MSI, two wave bands are selected to be a wave bands and b wave bands (respectively, the red light absorption peak wave bands and the fluorescence peak wave bands, the center wavelengths are 665nm and 708 nm), and the difference between the chlorophyll a absorption valleys and the reflection peaks is enlarged by rho a and rho b, so that a ratio index is constructed. The linear expression consisting of the ratio index and chlorophyll a concentration is:
Wherein C chl-a represents the chlorophyll a concentration in the plateau lake; C. d is a second correlation coefficient of the regression model; ρ a、ρb is the water body reflectivity of the selected a and b wave bands (the red light absorption peak wave band and the fluorescence peak wave band).
3) Constructing normalized chlorophyll a index:
The positions of reflection peaks or absorption valleys of the chlorophyll a reflection spectrum are mainly related to red light absorption peak wave bands and fluorescence peak wave bands of Sentinel-2MSI, two wave bands are selected to be a wave bands and b wave bands (respectively, the red light absorption peak wave bands and the fluorescence peak wave bands, and the center wavelengths are 665nm and 708nm respectively), and the reflectivity contrast of the two wave bands is further enhanced through a nonlinear stretching mode (the difference between rho a、ρb and rho a、ρb is compared), so that a normalized chlorophyll a index is constructed. The linear expression consisting of normalized chlorophyll a index and chlorophyll a concentration is:
Wherein C chl-a represents the chlorophyll a concentration in the plateau lake; J. k is a third correlation coefficient of the regression model; ρ a、ρb is the water body reflectivity of the selected a and b wave bands (the red light absorption peak wave band and the fluorescence peak wave band).
4) Constructing three-band indexes:
Based on a chlorophyll a biological optical model, red light absorption peak wave bands, fluorescence peak wave bands adjacent to the red light absorption peak wave bands and absorption peak wave bands of pure water under ideal conditions are respectively selected according to model requirements, the 3 wave bands are respectively selected to be a, b and c wave bands (respectively red light absorption peak wave bands, fluorescence peak wave bands and absorption peak wave bands, and central wavelengths are 665nm, 708nm and 730 nm) by combining with the wave band setting of Sentinel-2MSI, and a three-wave band index is constructed, wherein the linear expression formed by the three-wave band index and the concentration of chlorophyll a is as follows:
Cchl-a=M+N·(ρa -1-ρb -1)·ρc (5)
Wherein C chl-a represents the chlorophyll a concentration in the plateau lake; m, N is the fourth correlation coefficient of the regression model; ρ a -1、ρb -1 is the inverse of the water body reflectivity of the selected a, b bands (the selected red light absorption peak band and fluorescence peak band), respectively; ρ c is the water reflectance of the selected c-band (selected absorption peak band).
Fifthly, measuring the feature importance of the random forest algorithm:
1) Random forest refers to a classifier which uses a plurality of trees to train and predict samples, and for each decision tree, corresponding out-of-bag data are selected to calculate out-of-bag data errors, which are denoted as err 1. When the decision tree model is built, the spectral index (chlorophyll a inversion spectral index) built by the wave band combination is used as the characteristic of the decision tree to be input, simulation training is carried out by actually measuring the concentration of chlorophyll a, and the spectral index with high correlation with the concentration of chlorophyll a is output. The data outside the bag means that when the decision tree is built every time, one data is obtained through repeated sampling and used for training the decision tree, and at the moment, about 1/3 of the data is not utilized and does not participate in the building of the decision tree. This portion of the data may be used to evaluate the performance of the decision tree, calculating the prediction error rate of the model, referred to as the out-of-bag data error.
2) The value of the data outside the bag at a certain characteristic is changed randomly, namely random noise is added, the value of the data outside the bag under different spectral indexes is changed, and the error of the data outside the bag is calculated again and is recorded as err 2. Wherein the spectral index comprises: single band index, ratio index, normalized chlorophyll a index, and three band index.
3) Adding random noise, if the accuracy of the data outside the bag is greatly reduced, namely err 1 is reduced, err 2 is increased, the characteristic importance degree is high, and further, the characteristic (particularly referred to as a spectrum index) has great influence on chlorophyll a concentration results. The characteristic importance is calculated respectively on the assumption that N decision trees exist in the forest, and the spectrum index with great influence on chlorophyll a concentration results is obtained, and the general formula is as follows:
Where V represents feature importance, err 1 represents the out-of-bag data error, err 2 represents the out-of-bag data error after changing the value at a feature, and N represents the number of decision trees in the random forest.
4) And arranging the feature importance calculation results from high to low, namely arranging the influence results of the spectrum indexes on the chlorophyll a concentration, and obtaining the spectrum index with the highest feature importance.
Step six, inverting the chlorophyll a concentration of the plateau lake:
1) obtaining a spectral index with highest random forest feature importance as a preferred spectral index, and carrying out linear regression on the preferred spectral index serving as an independent variable and the concentration of chlorophyll a as a dependent variable according to formulas (2), (3), (4) and (5) to obtain a correlation coefficient (such as A, B in formula (2) of a regression model, wherein the regression model is a chlorophyll a inversion model. Wherein the above formulas (2) (3) (4) (5) are linear expressions of the regression model.
2) And applying the chlorophyll a inversion model to the atmospheric corrected Sentinel-2MSI atmospheric bottom reflectivity data, taking the optimal spectral index as an independent variable, obtaining a correlation coefficient of the model after linear regression, and realizing remote sensing inversion of the chlorophyll a concentration of the plateau lake. For example, regarding Yunnan province plateau lakes, taking Yunnan province as an example, a linear expression formula of a regression model is finally obtained as follows:
In the above formula, C chl-a represents the chlorophyll a concentration in the plateau lake; ρ a and ρ b represent the water body reflectivities of the red absorption peak band and the fluorescence peak band, respectively.
Compared with the traditional monitoring method, the remote sensing quantitative inversion method for the chlorophyll a concentration of the plateau lake can realize rapid, accurate and large-area-range monitoring of the plateau lake, and provides a more convenient means for monitoring the water quality of the plateau lake with complex and various geological features. The method is simple and practical, based on feature selection, multiple inversion indexes can be compared, and aiming at a certain plateau lake, the most effective model for inverting the chlorophyll a concentration (establishing a chlorophyll a inversion model) can be rapidly optimized, so that the chlorophyll a inversion accuracy is further improved. Aiming at the application of the plateau lake, under the condition that the spectrum data of the plateau lake is difficult to acquire, the method does not need to carry out field measurement, and can quickly obtain the optimal model for inverting the chlorophyll a of the plateau lake only through the comparison of characteristic wave bands, so that the method realizes the large-scale and periodic observation of the plateau lake, saves material resources and financial resources, and has important significance for intelligent monitoring and management of the plateau lake.
It will be apparent to those skilled in the art that various modifications and variations can be made to the present invention without departing from the spirit or scope of the invention. Thus, it is intended that the present invention also include such modifications and alterations insofar as they come within the scope of the appended claims or the equivalents thereof.
Claims (3)
1. A remote sensing inversion method for chlorophyll a concentration in a plateau lake is characterized by comprising the following steps:
S1, respectively obtaining the chlorophyll a concentration of the plateau lake and remote sensing reflectivity data of the plateau lake; the concentration of chlorophyll a in the plateau lake is obtained from a plateau lake ground measuring point, wherein the remote sensing reflectivity data of the plateau lake is Sentinel-2MSI remote sensing reflectivity data;
S2, performing atmospheric correction on the remote sensing reflectivity data of the plateau lake to obtain the reflectivity data of the bottom layer of the atmosphere;
s3, respectively constructing chlorophyll a inversion spectrum indexes according to the inherent optical characteristics of the chlorophyll a and the reflectivity data of the bottom layer of the atmosphere; the chlorophyll a inversion spectral index includes: single band index, ratio index, normalized chlorophyll a index, and three band index;
S4, establishing a decision tree model according to the chlorophyll a inversion spectrum index, and determining the spectrum index with the greatest influence on the chlorophyll a concentration result as a preferred spectrum index;
s5, carrying out linear regression on the optimized spectral index and the concentration of chlorophyll a in the plateau lake, and establishing a chlorophyll a inversion model to obtain a correlation coefficient of a regression model;
s6, according to the chlorophyll a inversion model and the atmospheric bottom layer reflectivity data, remote sensing inversion of the chlorophyll a concentration of the plateau lake is realized;
In step 3, selecting a red light absorption peak wave band according to the reflectivity data of the bottom layer of the atmosphere, and constructing a single-band index, wherein a linear expression formed by the single-band index and the concentration of chlorophyll a in the plateau lake is as follows:
Cchl-a=A+B·ρa (2);
Wherein C chl-a represents the chlorophyll a concentration in the plateau lake; A. b is a first correlation coefficient of the regression model; ρ a is the water body reflectivity of the selected red light absorption peak band;
selecting the red light absorption peak wave band and the fluorescence peak wave band according to the reflectivity data of the atmosphere bottom layer, and constructing a ratio index, wherein a linear expression formed by the ratio index and the concentration of chlorophyll a in the plateau lake is as follows:
Wherein C chl-a represents the chlorophyll a concentration in the plateau lake; C. d is a second correlation coefficient of the regression model; ρ a、ρb is the water body reflectivity of the red light absorption peak wave band and the fluorescence peak wave band respectively;
selecting the red light absorption peak wave band and the fluorescence peak wave band according to the reflectivity data of the bottom layer of the atmosphere; the contrast of the reflectivities of the red light absorption peak wave band and the fluorescence peak wave band is further enhanced through a nonlinear stretching mode, and a normalized chlorophyll a index is constructed; the linear expression formed by the normalized chlorophyll a index and the concentration of chlorophyll a in the plateau lake is as follows:
Wherein C chl-a represents the chlorophyll a concentration in the plateau lake; J. k is a third correlation coefficient of the regression model; ρ a、ρb is the water body reflectivity of the red light absorption peak wave band and the fluorescence peak wave band respectively;
according to the inherent optical characteristics of the chlorophyll a and the reflectivity data of the bottom layer of the atmosphere, respectively selecting the red light absorption peak wave band, the fluorescence peak wave band and the absorption peak wave band of the pure water body under ideal conditions to construct a three-wave-band index; wherein the fluorescence peak wave band is selected to be close to the red light absorption peak wave band; the linear expression formed by the three-band index and the concentration of chlorophyll a in the plateau lake is as follows:
Cchl-a=M+N·(ρa -1-ρb -1)·ρc (5);
Wherein C chl-a represents the chlorophyll a concentration in the plateau lake; m, N is the fourth correlation coefficient of the regression model; ρ a -1、ρb -1 is the inverse of the water body reflectivity of the selected red light absorption peak band and fluorescence peak band respectively; ρ c is the water body reflectivity of the selected absorption peak band;
in step 4, the step of obtaining the preferred spectral index is:
Establishing the decision tree model according to the chlorophyll a inversion spectrum index, selecting corresponding out-of-bag data to calculate out-of-bag data errors, and generating a first error;
Adding random noise, respectively changing the value of the data outside the bag corresponding to each chlorophyll a inversion spectrum index, and calculating the data error outside the bag again to generate a second error;
According to the first error and the second error, respectively calculating the characteristic importance of each chlorophyll-a inversion spectral index, and determining the spectral index corresponding to the highest characteristic importance;
Taking the spectral index with the highest characteristic importance as the spectral index with the greatest influence on the chlorophyll a concentration result, and taking the spectral index as the optimal spectral index;
The calculation formula of the feature importance is as follows:
Wherein V represents feature importance, err 1 represents an out-of-bag data error, err 2 represents an out-of-bag data error after changing a value at a certain feature, and N represents the number of decision trees in a random forest;
in step 5, a chlorophyll a inversion model is established: and obtaining the optimal spectrum index, carrying out linear regression on the optimal spectrum index serving as an independent variable and the concentration of chlorophyll a in the plateau lake as a dependent variable to obtain a correlation coefficient of a regression model, and taking the regression model as the chlorophyll a inversion model.
2. The remote sensing inversion method for chlorophyll a concentration in a plateau lake of claim 1, wherein the step S2 further comprises: and extracting the plateau lake water body region by adopting an improved normalized difference water body index according to the green light wave band and the short wave infrared wave band in the atmospheric bottom reflectivity data.
3. The remote sensing inversion method of chlorophyll a concentration in a plateau lake according to claim 2, wherein the water body area in the plateau lake is extracted by the following formula:
In the above formula MNDWI represents an improved normalized difference water index; ρ Green represents the green band reflectivity in the atmospheric bottom reflectivity data; ρ SWIR represents the short wave infrared band reflectivity in the atmospheric subsurface reflectivity data.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210639482.XA CN115080905B (en) | 2022-06-07 | 2022-06-07 | Remote sensing inversion method for chlorophyll a concentration in plateau lake |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210639482.XA CN115080905B (en) | 2022-06-07 | 2022-06-07 | Remote sensing inversion method for chlorophyll a concentration in plateau lake |
Publications (2)
Publication Number | Publication Date |
---|---|
CN115080905A CN115080905A (en) | 2022-09-20 |
CN115080905B true CN115080905B (en) | 2024-05-17 |
Family
ID=83251466
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202210639482.XA Active CN115080905B (en) | 2022-06-07 | 2022-06-07 | Remote sensing inversion method for chlorophyll a concentration in plateau lake |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN115080905B (en) |
Families Citing this family (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN116067897B (en) * | 2022-12-27 | 2024-02-09 | 二十一世纪空间技术应用股份有限公司 | Chlorophyll a concentration inversion method and system based on integrated factor system |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108956505A (en) * | 2018-09-18 | 2018-12-07 | 航天信德智图(北京)科技有限公司 | The detection method and device of small water Determination of Chlorophyll a concentration based on Sentinel-2 image |
CN111781147A (en) * | 2020-07-06 | 2020-10-16 | 淮阴师范学院 | Remote sensing inversion model and method for chlorophyll a concentration of inland lake water body based on two wave bands |
CN112131746A (en) * | 2020-09-24 | 2020-12-25 | 中国科学院空天信息创新研究院 | Chlorophyll a concentration inversion method and system |
CN112131708A (en) * | 2020-08-24 | 2020-12-25 | 无锡德林海环保科技股份有限公司 | Plateau lake chlorophyll a concentration remote sensing inversion method based on HY-1C data |
CN112881293A (en) * | 2021-01-08 | 2021-06-01 | 浙江工商大学 | Inland lake clean water body chlorophyll a concentration inversion method based on high-grade first satellite |
CN112989692A (en) * | 2021-02-10 | 2021-06-18 | 中国科学院南京地理与湖泊研究所 | Lake eutrophication inversion method based on remote sensing data |
CN114330530A (en) * | 2021-12-24 | 2022-04-12 | 安徽大学 | Chlorophyll a concentration inversion method based on space-spectrum fusion and model learning coupling |
-
2022
- 2022-06-07 CN CN202210639482.XA patent/CN115080905B/en active Active
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108956505A (en) * | 2018-09-18 | 2018-12-07 | 航天信德智图(北京)科技有限公司 | The detection method and device of small water Determination of Chlorophyll a concentration based on Sentinel-2 image |
CN111781147A (en) * | 2020-07-06 | 2020-10-16 | 淮阴师范学院 | Remote sensing inversion model and method for chlorophyll a concentration of inland lake water body based on two wave bands |
CN112131708A (en) * | 2020-08-24 | 2020-12-25 | 无锡德林海环保科技股份有限公司 | Plateau lake chlorophyll a concentration remote sensing inversion method based on HY-1C data |
CN112131746A (en) * | 2020-09-24 | 2020-12-25 | 中国科学院空天信息创新研究院 | Chlorophyll a concentration inversion method and system |
CN112881293A (en) * | 2021-01-08 | 2021-06-01 | 浙江工商大学 | Inland lake clean water body chlorophyll a concentration inversion method based on high-grade first satellite |
CN112989692A (en) * | 2021-02-10 | 2021-06-18 | 中国科学院南京地理与湖泊研究所 | Lake eutrophication inversion method based on remote sensing data |
CN114330530A (en) * | 2021-12-24 | 2022-04-12 | 安徽大学 | Chlorophyll a concentration inversion method based on space-spectrum fusion and model learning coupling |
Non-Patent Citations (8)
Title |
---|
Quantification of chlorophyll-a in typical lakes across China using Sentinel-2 MSI imagery with machine learning algorithm;Sijia Li 等;《Science of The Total Environment》;20210715;第778卷;第1-6页 * |
一种湿地河流叶绿素a遥感反演方法;刘卫华;王思远;马元旭;申明;游永发;海凯;吴林霖;;地球信息科学学报;20201022(第10期);第2062-2077页 * |
农作物叶绿素含量遥感估算的研究进展与展望;徐晋;蒙继华;;遥感技术与应用;20160215(第01期);第74-85页 * |
利用光谱指数反演植被叶绿素含量的精度及稳定性研究;姜海玲;杨杭;陈小平;王树东;李雪轲;刘凯;岑奕;;光谱学与光谱分析;20150415(第04期);第975-981页 * |
基于Sentinel-2的平寨水库叶绿素a浓度反演;但雨生;周忠发;李韶慧;张昊天;蒋翼;;环境工程;20200315(第03期);第180-186页 * |
太湖水体叶绿素浓度反演模型适宜性分析;王珊珊;李云梅;王永波;王帅;杜成功;;湖泊科学;20150106(第01期);第150-162页 * |
太阳诱导叶绿素荧光的卫星遥感反演方法研究进展;纪梦豪 等;《中国优秀硕士学位论文全文数据库-工程科技Ⅰ辑》;20190620;第34卷(第03期);第455-466页 * |
鄱阳湖叶绿素a浓度遥感定量模型研究;江辉;;测绘科学;20121120(第06期);第49-52页 * |
Also Published As
Publication number | Publication date |
---|---|
CN115080905A (en) | 2022-09-20 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Craig et al. | Deriving optical metrics of coastal phytoplankton biomass from ocean colour | |
Li et al. | Multi-LUTs method for canopy nitrogen density estimation in winter wheat by field and UAV hyperspectral | |
Zhao et al. | Hyperspectral remote sensing of plant biochemistry using Bayesian model averaging with variable and band selection | |
Verrelst et al. | Retrieval of vegetation biophysical parameters using Gaussian process techniques | |
CN107271382A (en) | A kind of different growing rape leaf SPAD value remote sensing estimation methods | |
Ismail et al. | Discriminating the early stages of Sirex noctilio infestation using classification tree ensembles and shortwave infrared bands | |
CN105608296B (en) | A kind of blade potassium concn inversion method based on lichee canopy spectra | |
CN110569605A (en) | Non-glutinous rice leaf nitrogen content inversion model method based on NSGA2-ELM | |
CN115080905B (en) | Remote sensing inversion method for chlorophyll a concentration in plateau lake | |
Li et al. | Estimation of potato chlorophyll content using composite hyperspectral index parameters collected by an unmanned aerial vehicle | |
CN113420497A (en) | Remote sensing estimation method for total phosphorus concentration of turbid lake | |
CN103278467A (en) | Rapid nondestructive high-accuracy method with for identifying abundance degree of nitrogen element in plant leaf | |
CN112666328A (en) | Chlorophyll a remote sensing inversion method and device based on genetic neural network model | |
Liu et al. | Detection of chlorophyll content in growth potato based on spectral variable analysis | |
Pozdnyakov et al. | An advanced algorithm for operational retrieval of water quality from satellite data in the visible | |
CN113065230B (en) | High-spectrum inversion model for establishing rice leaf SPAD based on optimized spectral index | |
Paavel et al. | Parameterization of chlorophyll-specific phytoplankton absorption coefficients for productive lake waters | |
CN116501925A (en) | Method, system, equipment and medium for predicting leaf area index under multi-source remote sensing data | |
CN116448687A (en) | Gray neural network-based inversion method for nitrogen and phosphorus content of vegetation canopy in falling area | |
Tang et al. | Hyperspectral prediction of mangrove leaf stoichiometries in different restoration areas based on machine learning models | |
CN112595267A (en) | Summer corn LAI inversion method fusing factor analysis and ant colony wavelet neural network model | |
Kong et al. | An integrated field and hyperspectral remote sensing method for the estimation of pigments content of Stipa Purpurea in Shenzha, Tibet | |
Riha et al. | Development of a remote sensing algorithm for cyanobacterial phycocyanin pigment in the Baltic Sea using neural network approach | |
CN111766224B (en) | Method for inverting chlorophyll fluorescence spectrum by only using radiance data | |
Sawut et al. | Applicability of optimized hyperspectral indices for estimating Betalain content in Suaeda salsa |
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 | ||
GR01 | Patent grant | ||
GR01 | Patent grant |