CN115495988A - Soil remote sensing inversion method based on optimal time window selection - Google Patents
Soil remote sensing inversion method based on optimal time window selection Download PDFInfo
- Publication number
- CN115495988A CN115495988A CN202211189137.7A CN202211189137A CN115495988A CN 115495988 A CN115495988 A CN 115495988A CN 202211189137 A CN202211189137 A CN 202211189137A CN 115495988 A CN115495988 A CN 115495988A
- Authority
- CN
- China
- Prior art keywords
- remote sensing
- soil
- target
- value
- month
- 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.)
- Granted
Links
- 239000002689 soil Substances 0.000 title claims abstract description 138
- 238000000034 method Methods 0.000 title claims abstract description 28
- 238000011160 research Methods 0.000 claims abstract description 53
- 238000011156 evaluation Methods 0.000 claims description 36
- 238000005070 sampling Methods 0.000 claims description 34
- 238000012545 processing Methods 0.000 claims description 15
- 238000012937 correction Methods 0.000 claims description 12
- 238000010801 machine learning Methods 0.000 claims description 11
- 238000002310 reflectometry Methods 0.000 claims description 6
- 238000012549 training Methods 0.000 claims description 5
- 230000008878 coupling Effects 0.000 claims description 4
- 238000010168 coupling process Methods 0.000 claims description 4
- 238000005859 coupling reaction Methods 0.000 claims description 4
- 238000006243 chemical reaction Methods 0.000 claims description 3
- 238000005516 engineering process Methods 0.000 abstract description 5
- 239000010902 straw Substances 0.000 abstract description 5
- 239000004016 soil organic matter Substances 0.000 description 8
- 238000001228 spectrum Methods 0.000 description 4
- 238000013461 design Methods 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 238000013507 mapping Methods 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000004181 pedogenesis Methods 0.000 description 2
- 230000015572 biosynthetic process Effects 0.000 description 1
- 238000002790 cross-validation Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
- 238000003786 synthesis reaction Methods 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Images
Classifications
-
- 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
- G01N21/17—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
- G01N21/31—Investigating relative effect of material at wavelengths characteristic of specific elements or molecules, e.g. atomic absorption spectrometry
- G01N21/35—Investigating relative effect of material at wavelengths characteristic of specific elements or molecules, e.g. atomic absorption spectrometry using infrared light
- G01N21/359—Investigating relative effect of material at wavelengths characteristic of specific elements or molecules, e.g. atomic absorption spectrometry using infrared light using near infrared light
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9537—Spatial or temporal dependent retrieval, e.g. spatiotemporal queries
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
- G06N20/10—Machine learning using kernel methods, e.g. support vector machines [SVM]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/77—Processing image or video features in feature spaces; using data integration or data reduction, e.g. principal component analysis [PCA] or independent component analysis [ICA] or self-organising maps [SOM]; Blind source separation
- G06V10/774—Generating sets of training patterns; Bootstrap methods, e.g. bagging or boosting
-
- Y—GENERAL 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
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Databases & Information Systems (AREA)
- Software Systems (AREA)
- Evolutionary Computation (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Medical Informatics (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Data Mining & Analysis (AREA)
- Spectroscopy & Molecular Physics (AREA)
- Computing Systems (AREA)
- Analytical Chemistry (AREA)
- Chemical & Material Sciences (AREA)
- Geometry (AREA)
- Biochemistry (AREA)
- Life Sciences & Earth Sciences (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Computer Hardware Design (AREA)
- Multimedia (AREA)
- Mathematical Physics (AREA)
- Image Processing (AREA)
Abstract
The invention relates to a soil remote sensing inversion method based on optimal time window selection, which comprises the steps of firstly, selecting optimal years, and reducing possible influence of factors such as climate among different years on soil attribute remote sensing inversion accuracy; secondly, determining an optimal month jointly by combining the crop growth condition reflected by the regional vegetation index and the inverted precision result of the soil attributes in different months, and reducing the influence of crop and crop straw coverage; finally, an optimal time window for regional soil attribute inversion is determined by integrating the optimal year and the optimal month, and a target prediction model in the optimal time window is used for obtaining the target soil attribute in the target research region; the remote sensing image in the optimal time window is designed and utilized, so that the influence of factors such as climate, crop coverage and the like can be effectively overcome, the precision of soil attribute remote sensing inversion is effectively improved, and the practical application of the soil attribute remote sensing inversion technology is promoted.
Description
Technical Field
The invention relates to a soil remote sensing inversion method based on optimal time window selection, and belongs to the technical field of soil attribute detection.
Background
The soil has multiple functions, has important ecological and environmental functions besides the production function, and is vital to national food safety in terms of soil resource protection. Currently, some unreasonable links and measures exist in the utilization of soil resources, and effective monitoring of the soil resources is a premise for realizing sustainable utilization and protection of the soil resources. However, the soil has strong space-time variability and is comprehensively influenced by factors such as mother substances, climate, terrain, organisms and the like in the long forming process, so that the soil has strong regional characteristics.
In recent years, the rapid development of digital soil mapping technology for indirectly reflecting soil properties by comprehensively utilizing relevant parameters in the soil formation process has become an important method for regional soil property mapping. With the progress of relevant research, certain challenges are faced in mechanism explanation and application in areas with small variation of relevant parameters in the soil formation process. The remote sensing technology has the characteristics of strong timeliness and large coverage range, and is applied to soil attribute inversion and mapping to a certain extent in recent years.
Compared with the spectrums of green crops and crop straws, the soil has unique spectrum curve characteristics, and soil attribute information is reflected by remote sensing by utilizing soil spectrum information. The farmland soil and the soil surface are inevitably influenced by coverage factors such as crops, crop straws and the like, so that the precision of soil remote sensing inversion is possibly reduced; on the other hand, the state of the soil itself, such as its water content difference, may also affect the inversion accuracy. Eliminating or reducing the influence of the factors on the inversion accuracy is a key link of the soil remote sensing application.
In recent years, many soil remote sensing inversion applications only select a remote sensing image time window according to regional crop growth characteristics, and the method has a certain effect on selection of months, but often lacks objective basis on selection of years, so that the inversion accuracy has larger uncertainty. The regional soil remote sensing inversion application represented by image synthesis has large prediction precision difference in different regions, and also faces certain challenges in the regional application with short bare soil regions and time windows.
Disclosure of Invention
The invention aims to solve the technical problem of providing a soil remote sensing inversion method based on optimal time window selection, which can effectively and accurately realize the rapid acquisition of regional soil attributes and improve the actual working efficiency.
The invention adopts the following technical scheme for solving the technical problems: the invention designs a soil remote sensing inversion method based on optimal time window selection, which comprises the following steps of A to E, obtaining a target prediction model of target soil attributes in a target research area, and obtaining the target soil attributes in the target research area by applying the target prediction model;
step A, obtaining remote sensing images covering a target research area respectively corresponding to at least two months in each year, wherein each month should have at least one remote sensing image, and presetting data values of target soil attributes corresponding to each sampling position in the target research area, then carrying out atmospheric correction processing on each remote sensing image, and entering step B;
b, respectively aiming at each scene remote sensing image, based on the remote sensing image and a data value preset in a target research area, of each sampling position corresponding to the target soil attribute, applying a preset machine learning algorithm to train and obtain a first prediction model taking the remote sensing image value of each sampling position in the target research area as input and the data value of the target soil attribute corresponding to each sampling position as output; further obtaining a first prediction model corresponding to each scene remote sensing image, and then entering the step C;
step C, carrying out statistical summary on the first prediction models corresponding to different years based on the evaluation results of the preset target evaluation indexes corresponding to the first prediction models, determining the optimal year of remote sensing inversion of soil attributes, and then entering step D;
step D, coupling normalized vegetation indexes NDVI based on the evaluation results of the preset target evaluation indexes corresponding to the first prediction models, carrying out statistical summary on the first prediction models corresponding to different months, determining the optimal month of remote sensing inversion of soil attributes, and then entering step E;
and E, performing remote sensing inversion on the optimal year and the optimal month by using the soil attributes to serve as an optimal time window for soil attribute remote sensing inversion, performing remote sensing inversion on each scene remote sensing image in the target research area in the optimal time window based on the soil attributes, and training to obtain a target prediction model by using the remote sensing image value of each sampling position in the target research area as input and the data value of the target soil attribute corresponding to each sampling position in the target research area as output by applying a preset machine learning algorithm.
As a preferred technical scheme of the invention: in the step A, the following operations are executed respectively for each scene remote sensing image, and atmospheric correction processing updating is carried out;
the operation is as follows: performing radiometric calibration processing on each pixel of the remote sensing image according to L = DN α + β to obtain radiance data L, and performing atmospheric correction processing on the remote sensing image by using a FLAASH model according to the radiance data L; further realizing the atmospheric correction processing updating of the remote sensing image; wherein DN represents the pixel value of the remote sensing image, and alpha and beta are respectively the conversion coefficients attached to the remote sensing image.
As a preferred technical scheme of the invention: in the step C, R of the coefficient evaluation index is determined based on the correspondence of each first prediction model 2 The value and the RMSE value of the root mean square error evaluation index, wherein the maximum R is counted 2 And taking the year corresponding to the first prediction model to which the value and the minimum RMSE value belong together as the optimal year of the soil attribute remote sensing inversion.
As a preferred technical scheme of the invention: the step D comprises the following steps D1 to D4;
step D1, determining R of coefficient evaluation index based on each first prediction model 2 The value and the RMSE value of the root mean square error evaluation index, wherein the maximum R is counted 2 The month corresponding to the first prediction model to which the value and the minimum RMSE value belong together is taken as the month to be matched, and then the step D2 is carried out;
and D2, aiming at each scene remote sensing image, respectively, according to the following formula:
NDVI=(NIR-Red)/(NIR+Red)
obtaining the NDVI value of each pixel in the target research area in the remote sensing image, obtaining an average value as the NDVI value corresponding to the target research area in the remote sensing image, further obtaining the NDVI value corresponding to each scene of the remote sensing image, and then entering the step D3; the NIR represents the reflectivity of a near infrared band in the remote sensing image, and Red represents the reflectivity of a Red band in the remote sensing image;
step D3, respectively aiming at different months, calculating and obtaining an average value of NDVI values corresponding to all scene remote sensing images in the month, taking the average value as the NDVI value corresponding to the month, further obtaining the NDVI values corresponding to different months, and then entering the step D4;
step D4, if the month corresponding to the minimum NDVI value is the same month as the month to be matched, taking the month as the optimal month for remote sensing inversion of the soil attribute; and if the month corresponding to the minimum NDVI value is different from the month to be matched, selecting the month corresponding to the minimum NDVI value and the next adjacent month as the optimal remote sensing inversion month of the soil attributes together based on the closed loop sequence of the head month and the tail month.
As a preferred technical scheme of the invention: the step E comprises the following steps E1 to E4;
step E1, performing remote sensing inversion on the optimal year and the optimal month by using the soil attributes as an optimal time window for the remote sensing inversion of the soil attributes, and if the number of remote sensing images contained in the optimal time window for the remote sensing inversion of the soil attributes is one scene, using a first prediction model corresponding to the scene remote sensing image in the step B as a target prediction model; if the number of the remote sensing images contained in the soil attribute remote sensing inversion optimal time window is larger than one scene, entering the step E2;
e2, inverting each scene remote sensing image in the optimal time window based on soil attribute remote sensing to obtain each combination respectively comprising at least two scenes remote sensing images, and then entering the step E3;
step E3, respectively aiming at each combination, based on each scene remote sensing image in the combination and the data value of the target soil attribute corresponding to each sampling position in the target research area, a preset machine learning algorithm is applied, and a second prediction model which takes the remote sensing image value of each sampling position in the target research area as input and the data value of the target soil attribute corresponding to each sampling position as output is obtained through training; then, second prediction models corresponding to all the combinations are obtained, and then the step E4 is carried out;
and E4, determining the optimal second prediction model as the target prediction model based on the evaluation results of the preset target evaluation indexes corresponding to the second prediction models respectively.
As a preferred technical scheme of the invention: in the step E4, the determination coefficient evaluation index R is associated with each of the second prediction models 2 The RMSE value of the value and the root mean square error evaluation index, and the maximum R in the RMSE value 2 And a second prediction model to which the value and the minimum RMSE value belong together is used as a target prediction model.
As a preferred technical scheme of the invention: the preset machine learning algorithm is an SVM algorithm.
Compared with the prior art, the soil remote sensing inversion method based on the optimal time window selection has the following technical effects by adopting the technical scheme:
according to the soil remote sensing inversion method based on the optimal time window selection, the possible influence of factors such as climate among different years on the soil attribute remote sensing inversion accuracy is reduced by selecting the optimal year; secondly, determining an optimal month jointly by combining the crop growth condition reflected by the regional vegetation index and the inverted precision result of the soil attributes in different months, and reducing the influence of crop and crop straw coverage; finally, an optimal time window for regional soil attribute inversion is determined by integrating the optimal year and the optimal month, and a target prediction model in the optimal time window is used for obtaining target soil attributes in a target research region; the remote sensing image in the optimal time window is designed and utilized, so that the influence of factors such as climate, crop coverage and the like can be effectively overcome, the precision of soil attribute remote sensing inversion is effectively improved, and the practical application of the soil attribute remote sensing inversion technology is promoted.
Drawings
FIG. 1 is a flow chart of a soil remote sensing inversion method based on optimal time window selection according to the invention;
FIG. 2 is a spatial distribution diagram of predetermined sampling locations within a target study area according to an exemplary embodiment of the present invention;
FIG. 3 is a scatter diagram of soil organic matter prediction results obtained in the design example of the present invention;
fig. 4 is a soil organic matter space prediction map obtained in the design example of the present invention.
Detailed Description
The following description will explain embodiments of the present invention in further detail with reference to the accompanying drawings.
Factors such as soil surface coverings in different areas, soil self states and the like have a certain time variation rule, the rule is on the aspect of annual variation, namely, the difference between months exists, meanwhile, certain variation also exists between years, and the influence of interference factors on soil spectrums can be reduced by selecting effective year and month time windows, so that the aim of improving inversion accuracy is fulfilled. Therefore, the soil attribute information can be rapidly and accurately obtained by the method for realizing the remote sensing inversion of the soil attribute by searching the image in the optimal time window.
In practical application, as shown in fig. 1, a target prediction model of target soil attributes in a target research area is obtained according to the following steps a to E, and the target prediction model is applied to obtain the target soil attributes in the target research area.
Step A, obtaining remote sensing images covering a target research area respectively corresponding to at least two months in each year, wherein each month should have at least one remote sensing image, and a data value of a target soil attribute corresponding to each sampling position is preset in the target research area, then respectively aiming at each remote sensing image, executing the following operation, carrying out atmospheric correction processing updating, and then entering step B.
The operation is as follows: performing radiometric calibration processing on each pixel of the remote sensing image according to L = DN α + β to obtain radiance data L, and performing atmospheric correction processing on the remote sensing image by using a FLAASH model according to the radiance data L; further realizing the atmospheric correction processing updating of the remote sensing image; wherein DN represents the pixel value of the remote sensing image, and alpha and beta are respectively conversion coefficients attached to the remote sensing image.
In the practical implementation of the step A, for example, remote sensing image sources with a reentry interval of not less than 16 days, such as MODIS, landsat, sentinel-2 and other image sources, are selected for obtaining remote sensing images, so as to ensure that the remote sensing images can be obtained every month.
B, respectively aiming at each scene remote sensing image, based on the remote sensing image and a data value preset in a target research area and corresponding to the target soil attribute of each sampling position, applying a machine learning algorithm such as an SVM algorithm to train and obtain a first prediction model taking the remote sensing image value of each sampling position in the target research area as input and the data value corresponding to the target soil attribute of each sampling position as output; and further obtaining first prediction models corresponding to the remote sensing images of all scenes respectively, and then entering the step C.
C, determining R of coefficient evaluation index based on each first prediction model 2 The value and the RMSE value of the root mean square error evaluation index, wherein the maximum R is counted 2 And D, taking the year corresponding to the first prediction model to which the value and the minimum RMSE value belong together as the optimal year of the remote sensing inversion of the soil property, and then entering the step D.
And D, coupling normalized vegetation indexes NDVI based on the evaluation results of the first prediction models corresponding to the preset target evaluation indexes according to the following steps D1 to D4, carrying out statistical summary on the first prediction models corresponding to different months, determining the optimal month of remote sensing inversion of soil attributes, and entering the step E.
Step D1, determining R of coefficient evaluation index based on correspondence of each first prediction model 2 The value and the RMSE value of the root mean square error evaluation index, wherein the maximum R is counted 2 The month corresponding to the first prediction model to which the value and the minimum RMSE value belong together is taken as the month to be matched, and then the step D2 is carried out;
and D2, aiming at each scene remote sensing image, respectively, according to the following formula:
NDVI=(NIR-Red)/(NIR+Red)
obtaining the NDVI value of each pixel in the target research area in the remote sensing image, obtaining an average value as the NDVI value corresponding to the target research area in the remote sensing image, further obtaining the NDVI value corresponding to each scene remote sensing image, and then entering the step D3; the NIR represents the reflectivity of a near infrared band in the remote sensing image, and Red represents the reflectivity of a Red band in the remote sensing image;
step D3, respectively aiming at different months, calculating and obtaining an average value of NDVI values corresponding to all scene remote sensing images in the month, taking the average value as the NDVI value corresponding to the month, further obtaining the NDVI values corresponding to different months, and then entering the step D4;
step D4, if the month corresponding to the minimum NDVI value is the same month as the month to be matched, taking the month as the optimal month for remote sensing inversion of the soil attribute; and if the month corresponding to the minimum NDVI value is not the same as the month to be matched, selecting the month corresponding to the minimum NDVI value and the next adjacent month as the optimal month of the soil attribute remote sensing inversion together based on the closed loop sequence of the head month and the tail month.
And E, performing remote sensing inversion on the optimal year and the optimal month by using the soil attributes as an optimal time window for remote sensing inversion of the soil attributes, and applying a preset machine learning algorithm such as an SVM algorithm to train and obtain a target prediction model by using the remote sensing image values of the sampling positions in the target research area as input and the data values of the target soil attributes corresponding to the sampling positions in the target research area as output based on the remote sensing images of the scenes in the target research area in the optimal time window for remote sensing inversion of the soil attributes and the data values of the target soil attributes corresponding to the sampling positions in the target research area.
In practical application, the step E is specifically designed and executed as the following steps E1 to E4, so as to obtain the target prediction model.
Step E1, performing remote sensing inversion on the optimal year and the optimal month by using the soil attributes as an optimal time window for the remote sensing inversion of the soil attributes, and if the number of remote sensing images contained in the optimal time window for the remote sensing inversion of the soil attributes is one scene, using a first prediction model corresponding to the scene remote sensing image in the step B as a target prediction model; if the number of the remote sensing images contained in the soil attribute remote sensing inversion optimal time window is larger than one scene, entering the step E2;
e2, inverting each scene remote sensing image in the optimal time window based on soil attribute remote sensing to obtain each combination respectively comprising at least two scenes remote sensing images, and then entering the step E3;
step E3, respectively aiming at each combination, based on each scene remote sensing image in the combination and the data value of the target soil attribute corresponding to each sampling position in the target research area, a preset machine learning algorithm such as an SVM algorithm is applied, and a second prediction model which takes the remote sensing image value of each sampling position in the target research area as input and the data value of the target soil attribute corresponding to each sampling position as output is obtained through training; then, a second prediction model corresponding to each combination is obtained, and then the step E4 is carried out;
step E4, based on the R of each second prediction model corresponding to the coefficient evaluation index 2 The value and the RMSE value of the root mean square error evaluation index, wherein the maximum R is counted 2 And a second prediction model to which the value and the minimum RMSE value belong together is used as a target prediction model.
The soil remote sensing inversion method based on the optimal time window selection is applied to practice, and further detailed description is given by taking a district in Henan province as a target research area, taking farmland soil organic matters as inversion objects, taking long-time MODIS remote sensing images as image sources, and realizing regional farmland soil organic matter remote sensing inversion as an example, but not limiting the invention, and specifically, steps A to E are executed according to the method shown in FIG. 1.
Step A, selecting an MODIS image (MOD 09A 1) which is synthesized for 8 days and has a spatial resolution of 500m as a remote sensing image source to ensure that a remote sensing image can be obtained every month, wherein the image has 7 wave bands (wave band 1: 620-670 nm, wave band 2: 841-876 nm, wave band 3: 459-479 nm, wave band 4: 545-565 nm, wave band 5: 1230-1250 nm, wave band 6: 1628-1652 nm and wave band 7: 2105-2155 nm), acquiring 644 year-round remote sensing images in a covered hillock from 1 month to 12 months in 2021 year based on the remote sensing image source to ensure that a certain number of remote sensing images can be obtained in different parts and months, presetting sampling positions in a target research area shown in figure 2 to obtain data values of soil organic matters corresponding to the preset sampling positions in the target research area, then carrying out atmospheric correction processing on the remote sensing images, and then entering step B.
B, regarding 644 scene MODIS remote sensing images, respectively aiming at each remote sensing image, based on the remote sensing image and a data value of soil organic matter corresponding to each sampling position preset in a target research area, applying an SVM algorithm to train and obtain a first prediction model taking the remote sensing image value of each sampling position in the target research area as input and the data value of target soil attribute corresponding to each sampling position as output; and further obtaining first prediction models respectively corresponding to 644 scene MODIS remote sensing images, and then entering the step C.
And C, counting and summarizing the first prediction models corresponding to different years based on the evaluation results of the preset target evaluation indexes of the first prediction models, determining the optimal year of remote sensing inversion of the soil attributes, and entering the step D.
The step C is specifically as follows in the examples:
step C1, the soil organic matter prediction result of the 644 scene remote sensing image obtained by calculation, namely R 2 The values of RMSE are summarized according to different years from 2008 to 2021, and then the step C2 is carried out;
step C2. R in statistical results of different years 2 The maximum value is 0.31, the minimum value of RMSE is 0.25%, the year is 2010, so that the optimal year is predicted by remote sensing of 2010 as the soil attribute, and then the step D is carried out.
And D, coupling normalized vegetation indexes NDVI based on the evaluation results of the preset target evaluation indexes corresponding to the first prediction models, performing statistical summary on the first prediction models corresponding to different months, determining the optimal month for remote sensing inversion of the soil organic matters, and entering the step E.
The step D is specifically as follows in the example:
step D1, summarizing and counting the soil organic matter prediction result of the 644 scene remote sensing image obtained by calculation according to different months 2 With the RMSE results, the maximum R in the above was counted 2 Value associated with minimum RMSE valueThe month corresponding to the first prediction model is taken as the month to be matched, and then the step D2 is carried out;
and D2, aiming at all 644 scenes of remote sensing images, according to the following formula:
NDVI=(NIR-Red)/(NIR+Red)
calculating to obtain the NDVI value of each pixel in the mound county in the remote sensing image, obtaining an average value as the NDVI value corresponding to the target research area in the remote sensing image, further obtaining the NDVI value corresponding to each scene remote sensing image, and then entering the step D3;
step D3, calculating and obtaining the average value of the corresponding NDVI values of all scene remote sensing images in the month as the NDVI value corresponding to the month respectively aiming at different months, further obtaining the corresponding NDVI values of different months respectively, and then entering the step D4;
step D4, if the month corresponding to the minimum NDVI value is the same month as the month to be matched, taking the month as the optimal month for remote sensing inversion of the soil attribute; and if the month corresponding to the minimum NDVI value is not the same as the month to be matched, selecting the month corresponding to the minimum NDVI value and the next adjacent month as the optimal month of the soil attribute remote sensing inversion together based on the closed loop sequence of the head month and the tail month, and then entering the step E.
And E, determining an optimal time window of remote sensing inversion by comprehensively determining the optimal year and the optimal month, constructing a target prediction model of the image in the optimal time window, and generating a soil attribute parameter map in the target research area.
In the embodiment of the step E, according to the step E1, comprehensively judging the optimal year 2010 and the optimal months of 10 months and 11 months, determining the optimal time windows of the remote sensing inversion of the soil attributes as the months of 10 months and 11 months of 2010, because the number of remote sensing images of the internal scenes of the windows is greater than one scene; in the embodiment, two remote sensing images are selected and integrated to establish a target prediction model, namely, each combination respectively comprising at least two remote sensing images is obtained according to the step E2; obtaining a second prediction model corresponding to each combination according to the step E3, and finally adopting a Leave-one-out cross validation method according to the step E4, using the method R 2 The model accuracy is evaluated together with two indexes of RMSE, wherein R 2 The larger the value is, the smaller the RMSE value is, the higher the model precision is, and the optimal result of the model is R 2 = 0.45, rmse = 0.23%, scatter plot of model prediction results is shown in fig. 3, where maximum R is finally counted 2 And the second prediction model to which the value and the minimum RMSE value jointly belong is taken as a target prediction model.
In practical application, after the target prediction model is obtained, the target prediction model is applied to obtain the soil organic matter in the target research area, and a soil organic matter prediction drawing of the target research area is performed, wherein the result is shown in fig. 4.
In the application of the soil remote sensing inversion method based on the optimal time window selection, the possible influence of factors such as climate among different years on the soil attribute remote sensing inversion accuracy is reduced by selecting the optimal year; secondly, determining an optimal month jointly by combining the crop growth condition reflected by the regional vegetation index and the inverted precision result of the soil attributes in different months, and reducing the influence of crop and crop straw coverage; finally, an optimal time window for regional soil attribute inversion is determined by integrating the optimal year and the optimal month, and a target prediction model in the optimal time window is used for obtaining the target soil attribute in the target research region; the remote sensing image in the optimal time window is designed and utilized, so that the influence of factors such as climate, crop coverage and the like can be effectively overcome, the precision of soil attribute remote sensing inversion is effectively improved, and the practical application of the soil attribute remote sensing inversion technology is promoted.
The embodiments of the present invention have been described in detail with reference to the drawings, but the present invention is not limited to the above embodiments, and various changes can be made within the knowledge of those skilled in the art without departing from the gist of the present invention.
Claims (7)
1. A soil remote sensing inversion method based on optimal time window selection is characterized by comprising the following steps: according to the following steps A to E, obtaining a target prediction model of the target soil attribute of the target research area, and obtaining the target soil attribute of the target research area by applying the target prediction model;
step A, obtaining remote sensing images covering a target research area respectively corresponding to at least two months in each year, wherein each month should have at least one remote sensing image and a data value of a target soil attribute corresponding to each sampling position preset in the target research area, then carrying out atmospheric correction processing on each remote sensing image, and entering step B;
b, respectively aiming at each scene remote sensing image, based on the remote sensing image and a data value preset in a target research area and corresponding to the target soil attribute of each sampling position, applying a preset machine learning algorithm to train and obtain a first prediction model taking the remote sensing image value of each sampling position in the target research area as input and the data value corresponding to the target soil attribute of each sampling position as output; further obtaining a first prediction model corresponding to each scene remote sensing image, and then entering the step C;
step C, carrying out statistical summary on the first prediction models corresponding to different years based on the evaluation results of the preset target evaluation indexes corresponding to the first prediction models, determining the optimal year of remote sensing inversion of soil attributes, and then entering step D;
step D, coupling normalized vegetation indexes NDVI based on the evaluation results of the preset target evaluation indexes corresponding to the first prediction models, carrying out statistical summary on the first prediction models corresponding to different months, determining the optimal month of remote sensing inversion of soil attributes, and then entering step E;
and E, performing remote sensing inversion on the optimal year and the optimal month by using the soil attributes to serve as an optimal time window for soil attribute remote sensing inversion, performing remote sensing inversion on each scene remote sensing image in the target research area in the optimal time window based on the soil attributes, and training to obtain a target prediction model by using the remote sensing image value of each sampling position in the target research area as input and the data value of the target soil attribute corresponding to each sampling position in the target research area as output by applying a preset machine learning algorithm.
2. The soil remote sensing inversion method based on optimal time window selection as claimed in claim 1, wherein: in the step A, the following operations are executed for each scene remote sensing image respectively, and atmosphere correction processing updating is carried out;
the operation is as follows: performing radiometric calibration processing on each pixel of the remote sensing image according to L = DN α + β to obtain radiance data L, and performing atmospheric correction processing on the remote sensing image by using a FLAASH model according to the radiance data L; further realizing the atmospheric correction processing updating of the remote sensing image; wherein DN represents the pixel value of the remote sensing image, and alpha and beta are respectively the conversion coefficients attached to the remote sensing image.
3. The soil remote sensing inversion method based on the optimal time window selection as claimed in claim 1, characterized in that: in the step C, R of the coefficient evaluation index is determined based on the correspondence of each first prediction model 2 The RMSE value of the value and the root mean square error evaluation index, and the maximum R in the RMSE value 2 And taking the year corresponding to the first prediction model to which the value and the minimum RMSE value belong together as the optimal year for remote sensing inversion of the soil property.
4. The soil remote sensing inversion method based on optimal time window selection as claimed in claim 1, wherein: the step D comprises the following steps D1 to D4;
step D1, determining R of coefficient evaluation index based on correspondence of each first prediction model 2 The RMSE value of the value and the root mean square error evaluation index, and the maximum R in the RMSE value 2 The month corresponding to the first prediction model to which the value and the minimum RMSE value belong together is taken as the month to be matched, and then the step D2 is carried out;
and D2, aiming at each scene remote sensing image, respectively, according to the following formula:
NDVI=(NIR-Red)/(NIR+Red)
obtaining the NDVI value of each pixel in the target research area in the remote sensing image, obtaining an average value as the NDVI value corresponding to the target research area in the remote sensing image, further obtaining the NDVI value corresponding to each scene remote sensing image, and then entering the step D3; the NIR represents the reflectivity of a near infrared band in the remote sensing image, and Red represents the reflectivity of a Red band in the remote sensing image;
step D3, respectively aiming at different months, calculating and obtaining an average value of NDVI values corresponding to all scene remote sensing images in the month, taking the average value as the NDVI value corresponding to the month, further obtaining the NDVI values corresponding to different months, and then entering the step D4;
step D4, if the month corresponding to the minimum NDVI value is the same as the month to be matched, the month is used as the optimal month for remote sensing inversion of the soil attributes; and if the month corresponding to the minimum NDVI value is not the same as the month to be matched, selecting the month corresponding to the minimum NDVI value and the next adjacent month as the optimal month of the soil attribute remote sensing inversion together based on the closed loop sequence of the head month and the tail month.
5. The soil remote sensing inversion method based on the optimal time window selection as claimed in claim 1, characterized in that: the step E comprises the following steps E1 to E4;
step E1, performing remote sensing inversion on the optimal year and the optimal month by using the soil attributes as an optimal time window for the remote sensing inversion of the soil attributes, and if the number of remote sensing images contained in the optimal time window for the remote sensing inversion of the soil attributes is one scene, using a first prediction model corresponding to the scene remote sensing images in the step B as a target prediction model; if the number of the remote sensing images contained in the soil attribute remote sensing inversion optimal time window is larger than one scene, entering the step E2;
e2, inverting each scene remote sensing image in the optimal time window based on soil attribute remote sensing to obtain each combination respectively comprising at least two scenes remote sensing images, and then entering the step E3;
step E3, respectively aiming at each combination, based on each scene remote sensing image in the combination and the data value of the target soil attribute corresponding to each sampling position in the target research area, a preset machine learning algorithm is applied, and a second prediction model which takes the remote sensing image value of each sampling position in the target research area as input and the data value of the target soil attribute corresponding to each sampling position as output is obtained through training; then, second prediction models corresponding to all the combinations are obtained, and then the step E4 is carried out;
and E4, determining the optimal second prediction model as the target prediction model based on the evaluation results of the preset target evaluation indexes corresponding to the second prediction models respectively.
6. The soil remote sensing inversion method based on the optimal time window selection as claimed in claim 5, wherein the method comprises the following steps: in the above step E4, R of the coefficient evaluation index is determined based on each of the second prediction models 2 The value and the RMSE value of the root mean square error evaluation index, wherein the maximum R is counted 2 And the second prediction model to which the value and the minimum RMSE value jointly belong is taken as a target prediction model.
7. The soil remote sensing inversion method based on the optimal time window selection as claimed in any one of claims 1 to 6, characterized in that: the preset machine learning algorithm is an SVM algorithm.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211189137.7A CN115495988B (en) | 2022-09-28 | 2022-09-28 | Soil remote sensing inversion method based on optimal time window selection |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211189137.7A CN115495988B (en) | 2022-09-28 | 2022-09-28 | Soil remote sensing inversion method based on optimal time window selection |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115495988A true CN115495988A (en) | 2022-12-20 |
| CN115495988B CN115495988B (en) | 2023-06-13 |
Family
ID=84472812
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211189137.7A Active CN115495988B (en) | 2022-09-28 | 2022-09-28 | Soil remote sensing inversion method based on optimal time window selection |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115495988B (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20190041377A1 (en) * | 2017-02-22 | 2019-02-07 | Hohai University | Method for measuring a mudflat elevation by remotely sensed water content |
| CN110046415A (en) * | 2019-04-08 | 2019-07-23 | 中国科学院南京地理与湖泊研究所 | A kind of soil organic matter content remote sensing dynamic playback method of space-time fining |
| CN110222656A (en) * | 2019-06-11 | 2019-09-10 | 成都理工大学 | Aboveground vegetation ecological water quantitative inversion method based on remote sensing technology |
| CN111678866A (en) * | 2020-05-28 | 2020-09-18 | 电子科技大学 | Soil water content inversion method for multi-model ensemble learning |
-
2022
- 2022-09-28 CN CN202211189137.7A patent/CN115495988B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20190041377A1 (en) * | 2017-02-22 | 2019-02-07 | Hohai University | Method for measuring a mudflat elevation by remotely sensed water content |
| CN110046415A (en) * | 2019-04-08 | 2019-07-23 | 中国科学院南京地理与湖泊研究所 | A kind of soil organic matter content remote sensing dynamic playback method of space-time fining |
| CN110222656A (en) * | 2019-06-11 | 2019-09-10 | 成都理工大学 | Aboveground vegetation ecological water quantitative inversion method based on remote sensing technology |
| CN111678866A (en) * | 2020-05-28 | 2020-09-18 | 电子科技大学 | Soil water content inversion method for multi-model ensemble learning |
Non-Patent Citations (4)
| Title |
|---|
| CHANGKUN WANG ET AL.: "The simultaneous prediction of soil properties and vegetation coverage from Vis-NIR Hyperspectral data with a one-dimensional convolutional neural network:a labortatory simulation study", REMOTE SENSING * |
| 徐佳等: "基于人工神经网络和随机森林学习模型从土壤 属性推测关键成土环境要素的研究" * |
| 李新尧等: "基于遥感的广东省陆河县水土流失敏感性分析", 亚热带水土保持 * |
| 葛玲玲: "基于机器学习的土壤湿度和气温遥感反演方法研究" * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115495988B (en) | 2023-06-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Tenreiro et al. | Using NDVI for the assessment of canopy cover in agricultural crops within modelling research | |
| US20220012385A1 (en) | Computer-implemented methods, computer readable medium and systems for a precision agriculture platform that detects healthy conditions | |
| Alqurashi et al. | An assessment of the impact of urbanization and land use changes in the fast-growing cities of Saudi Arabia | |
| CN108985959B (en) | Wheat powdery mildew remote sensing monitoring method based on surface temperature inversion technology | |
| Wu et al. | Land-cover classification using GF-2 images and airborne lidar data based on Random Forest | |
| CN106226260A (en) | A kind of combination microwave and the Soil Moisture Inversion method of infrared remote sensing image | |
| US10748081B2 (en) | Computer-implemented methods, computer readable medium and systems for a precision agriculture platform that identifies generic anomalies in crops | |
| US10410334B2 (en) | Computer-implemented methods, computer readable medium and systems for a precision agriculture platform with a satellite data model | |
| CN108986040B (en) | NDVI shadow influence removing method based on remote sensing multispectral image | |
| Kong et al. | phenofit: An R package for extracting vegetation phenology from time series remote sensing | |
| US10943173B2 (en) | Computer-implemented methods, computer readable medium and systems for generating a satellite data model for a precision agriculture platform | |
| CN114821362A (en) | Multi-source data-based rice planting area extraction method | |
| CN114663489A (en) | Crop leaf area index remote sensing inversion method and system under constraint of space-time characteristics of land blocks | |
| CN115631419B (en) | Rice planting area and spatial distribution extraction method and device based on change detection | |
| CN114120101A (en) | Soil moisture multi-scale comprehensive sensing method | |
| CN107273797B (en) | Rice sub-pixel identification method based on water body index variation coefficient | |
| CN115344815A (en) | Natural runoff change attribution analysis method and system considering vegetation space change | |
| Cheng et al. | Crop type classification with combined spectral, texture, and radar features of time-series Sentinel-1 and Sentinel-2 data | |
| Kim et al. | Growth monitoring of field-grown onion and garlic by CIE L* a* b* color space and region-based crop segmentation of UAV RGB images | |
| CN117274798B (en) | Remote sensing rice identification method based on regularized time sequence variation model | |
| Zhang et al. | Crop type mapping with temporal sample migration | |
| CN115495988A (en) | Soil remote sensing inversion method based on optimal time window selection | |
| Zou et al. | Combining spectral and texture feature of UAV image with plant height to improve LAI estimation of winter wheat at jointing stage | |
| Käfer et al. | Modeling Pinus elliottii growth with multitemporal Landsat data: a study case in southern Brazil | |
| Zhou et al. | Superpixel-based time-series reconstruction for optical images incorporating SAR data using autoencoder networks |
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 |