CN110516816B - All-weather earth surface temperature generation method and device based on machine learning - Google Patents
All-weather earth surface temperature generation method and device based on machine learning Download PDFInfo
- Publication number
- CN110516816B CN110516816B CN201910812200.XA CN201910812200A CN110516816B CN 110516816 B CN110516816 B CN 110516816B CN 201910812200 A CN201910812200 A CN 201910812200A CN 110516816 B CN110516816 B CN 110516816B
- Authority
- CN
- China
- Prior art keywords
- surface temperature
- data
- data set
- cloud
- formula
- 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
- 238000000034 method Methods 0.000 title claims abstract description 56
- 238000010801 machine learning Methods 0.000 title claims abstract description 54
- 230000005855 radiation Effects 0.000 claims abstract description 98
- 238000012549 training Methods 0.000 claims abstract description 31
- 238000007637 random forest analysis Methods 0.000 claims abstract description 17
- 230000002776 aggregation Effects 0.000 claims abstract description 7
- 238000004220 aggregation Methods 0.000 claims abstract description 7
- 230000003068 static effect Effects 0.000 claims abstract description 7
- 238000004364 calculation method Methods 0.000 claims description 47
- 238000004590 computer program Methods 0.000 claims description 22
- 230000005540 biological transmission Effects 0.000 claims description 13
- 238000012937 correction Methods 0.000 claims description 13
- 238000012545 processing Methods 0.000 claims description 11
- 230000006870 function Effects 0.000 claims description 10
- 238000002834 transmittance Methods 0.000 claims description 7
- 230000000007 visual effect Effects 0.000 claims description 6
- 238000012952 Resampling Methods 0.000 claims description 4
- 239000000126 substance Substances 0.000 claims description 4
- 230000004931 aggregating effect Effects 0.000 claims description 2
- 230000000379 polymerizing effect Effects 0.000 claims 1
- 238000003066 decision tree Methods 0.000 description 8
- 238000004422 calculation algorithm Methods 0.000 description 5
- 238000010586 diagram Methods 0.000 description 5
- 230000008569 process Effects 0.000 description 5
- 238000011160 research Methods 0.000 description 5
- 230000008901 benefit Effects 0.000 description 4
- 238000012544 monitoring process Methods 0.000 description 4
- 230000008859 change Effects 0.000 description 3
- 230000008878 coupling Effects 0.000 description 3
- 238000010168 coupling process Methods 0.000 description 3
- 238000005859 coupling reaction Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 238000004891 communication Methods 0.000 description 2
- 230000004927 fusion Effects 0.000 description 2
- 230000009467 reduction Effects 0.000 description 2
- 241000112598 Pseudoblennius percoides Species 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 230000001186 cumulative effect Effects 0.000 description 1
- 230000003247 decreasing effect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000012417 linear regression Methods 0.000 description 1
- 230000005012 migration Effects 0.000 description 1
- 238000013508 migration Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 238000003908 quality control method Methods 0.000 description 1
- 230000000630 rising effect Effects 0.000 description 1
- 239000002689 soil Substances 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01J—MEASUREMENT OF INTENSITY, VELOCITY, SPECTRAL CONTENT, POLARISATION, PHASE OR PULSE CHARACTERISTICS OF INFRARED, VISIBLE OR ULTRAVIOLET LIGHT; COLORIMETRY; RADIATION PYROMETRY
- G01J5/00—Radiation pyrometry, e.g. infrared or optical thermometry
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
Abstract
The invention discloses an all-weather earth surface temperature generation method and device based on machine learning, wherein the method adopts a tool MRT of MODIS to extract an MODIS data set subjected to remote sensing inversion; combining static meteorological satellite data with DEM topographic data of the ALOS satellite to estimate and obtain surface incident solar radiation; carrying out spatial aggregation on the data sets with the same spatial scale, and taking the data sets with the MODIS data sets as machine learning training data sets; constructing a surface temperature relation model through a random forest model; estimating the real surface temperature of the cloud covered pixel; and combining the real surface temperature with the cloud coverage pixel with the data set without the cloud coverage pixel to generate the all-weather surface temperature. The method solves the problems that the current thermal infrared remote sensing is easily influenced by cloud and fog and a large number of blank default areas exist in surface temperature products, achieves surface temperature estimation under cloud conditions, and provides an important basis for all-weather surface temperature product generation.
Description
Technical Field
The invention relates to a ground surface temperature monitoring technology, in particular to an all-weather ground surface temperature generation method and device based on machine learning.
Background
The surface temperature (LST) is an important parameter reflecting the interaction between earth and gas in the earth's surface system, is a key parameter influencing the processes of surface ecology, hydrology, meteorology and the like, and is a comprehensive result of energy exchange and substance migration and transformation between the atmosphere and the Land. Therefore, the method for quantitatively and accurately acquiring the space-time distribution characteristics of the surface temperature has important research significance and value for the energy balance of the ground gas system and the research of the ecological system. In addition, the dynamic monitoring of the resource environment on regional and global scales needs comprehensive, complete and continuous all-weather surface temperature and space-time distribution information, and is applied to the directions of agricultural drought prediction, farmland soil moisture management, crop yield prediction, numerical weather prediction, climate change and the like.
Among conventional surface temperature observations, surface site observation is the most direct way. However, due to the influence of high spatial heterogeneity of the earth surface temperature, the earth surface temperature measured by a single point has low spatial representativeness, and the earth surface temperature dynamic information of a large-scale area is difficult to accurately acquire through limited earth surface observation. In recent years, with the progress and development of satellite remote sensing space detection technology, remote sensing has become a main means for acquiring regional and global earth surface temperatures, wherein the most common monitoring modes comprise thermal infrared remote sensing inversion of earth surface temperatures and passive microwave remote sensing inversion of earth surface temperatures.
However, due to the fact that inversion accuracy of the passive microwave earth surface temperature is low, and large scale difference exists between passive microwave data and thermal infrared data, uncertainty of data fusion generated all-weather earth surface temperature products caused by errors of the passive microwave earth surface temperature and uncertainty of scale reduction exists; in addition, the problems of the existing inversion algorithm are as follows: the estimated surface temperature value is not the real surface temperature under the cloud at the satellite transit time, but the theoretical value of the surface temperature under the clear sky condition at the satellite transit time, so that the requirement of practical application is difficult to meet.
Therefore, how to overcome the problems in the method and effectively obtain the all-weather earth surface temperature of the cloud coverage area under the real condition so as to improve the spatial continuity of earth surface temperature products has important significance for improving the actual service and application level of earth surface temperature remote sensing products.
Disclosure of Invention
The present invention is directed to solving, at least to some extent, one of the technical problems in the related art. To this end, a first objective of the present invention is to provide an all-weather surface temperature generation method based on machine learning.
The second purpose of the invention is to provide an all-weather earth surface temperature generating device based on machine learning.
A third object of the invention is to propose a computer device.
A fourth object of the invention is to propose a computer storage medium.
To achieve the above object, in a first aspect, an all-weather surface temperature generation method based on machine learning according to an embodiment of the present invention includes:
respectively extracting MODIS data sets corresponding to cloud coverage pixels and cloud coverage-free pixels which are subjected to remote sensing inversion in corresponding MODIS land products by adopting a data processing tool MRT of the MODIS; the MODIS data set comprises a normalized vegetation index, an enhanced vegetation index, a leaf area index, earth surface albedo data and earth surface temperature with optimal inversion precision;
combining static meteorological satellite data with DEM topographic data of an ALOS satellite, estimating and obtaining earth surface incident solar radiation;
according to the space scale of the earth surface temperature, the earth surface incident solar radiation with the same space scale is polymerized to obtain a space scale polymerized data set; taking the data set aggregated by the spatial scale and the data set without the cloud coverage pixel as a machine learning training data set;
training the machine learning training data set by selecting a random forest model, and constructing and obtaining a surface temperature relation model; applying the surface temperature relation model to the data set with the cloud coverage pixel, and estimating and obtaining the surface real temperature of the cloud coverage pixel;
and combining the real surface temperature of the cloud covered pixel with the data set of the cloud-free covered pixel to generate the all-weather surface temperature.
In a second aspect, an all-weather surface temperature generation device based on machine learning according to an embodiment of the present invention includes:
the MODIS data set acquisition module is used for respectively extracting MODIS data sets corresponding to cloud coverage pixels and cloud coverage-free pixels which are subjected to remote sensing inversion in corresponding MODIS land products by adopting a data processing tool MRT of the MODIS; the MODIS data set comprises a normalized vegetation index, an enhanced vegetation index, a leaf area index, earth surface albedo data and earth surface temperature with optimal inversion precision;
the earth surface incident solar radiation acquisition module is used for combining static meteorological satellite data with DEM topographic data of an ALOS satellite, estimating and obtaining earth surface incident solar radiation;
the machine learning training data set acquisition module is used for aggregating the earth surface incident solar radiation with the same spatial scale according to the spatial scale of the earth surface temperature to obtain a data set aggregated by the spatial scale; taking the data set aggregated by the spatial scale and the data set without the cloud coverage pixel as a machine learning training data set;
the earth surface real temperature estimation module is used for selecting a random forest model to train the machine learning training data set, and constructing and obtaining an earth surface temperature relation model; applying the surface temperature relation model to the data set with the cloud coverage pixel, and estimating and obtaining the surface real temperature of the cloud coverage pixel;
and the all-weather surface temperature generating module is used for combining the real surface temperature with the cloud coverage pixel with the data set without the cloud coverage pixel to generate the all-weather surface temperature.
In a third aspect, a computer device according to an embodiment of the present invention includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the all-weather surface temperature generation method based on machine learning as described above when executing the computer program.
In a fourth aspect, a computer storage medium according to an embodiment of the present invention, on which a computer program is stored, is characterized in that the program, when executed by a processor, implements the all-weather surface temperature generation method based on machine learning as described above.
According to the all-weather earth surface temperature generation method and device based on machine learning, provided by the embodiment of the invention, an MODIS data set subjected to remote sensing inversion is extracted by adopting an MRT tool of an MODIS; combining static meteorological satellite data with DEM topographic data of the ALOS satellite to estimate and obtain surface incident solar radiation; carrying out spatial aggregation on the data sets with the same spatial scale, and taking the data sets with the MODIS data sets as machine learning training data sets; constructing a surface temperature relation model through a random forest model; estimating the real surface temperature of the cloud covered pixel; and combining the real surface temperature with the cloud coverage pixel with the data set without the cloud coverage pixel to generate the all-weather surface temperature. The method effectively solves the problems that the current thermal infrared remote sensing is easily influenced by cloud and fog and a large number of blank default areas exist in surface temperature products, realizes all-weather surface temperature estimation under the cloud condition, and provides an important foundation for all-weather surface temperature product generation.
Additional aspects and advantages of the invention will be set forth in part in the description which follows and, in part, will be obvious from the description, or may be learned by practice of the invention.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the structures shown in the drawings without creative efforts.
FIG. 1 is a flow chart of an all-weather surface temperature generation method based on machine learning according to an embodiment of the present invention;
FIG. 2 is a diagram of raw MODIS daytime surface temperature product data for a MODIS surface temperature product display;
FIG. 3 is a schematic representation of surface temperature data reconstructed using an embodiment of the method of the present invention;
FIG. 4 is a block diagram of an all-weather surface temperature generating device based on machine learning according to an embodiment of the present invention;
FIG. 5 is a schematic diagram of the structure of one embodiment of the computer device of the present invention.
The implementation, functional features and advantages of the objects of the present invention will be further explained with reference to the accompanying drawings.
Detailed Description
Reference will now be made in detail to embodiments of the present invention, examples of which are illustrated in the accompanying drawings, wherein like or similar reference numerals refer to the same or similar elements or elements having the same or similar function throughout. The embodiments described below with reference to the accompanying drawings are illustrative only for the purpose of explaining the present invention, and are not to be construed as limiting the present invention.
Currently, remote sensing has become a main means for acquiring regional and global earth surface temperatures, wherein the most common monitoring methods include thermal infrared remote sensing inversion of earth surface temperatures and passive microwave remote sensing inversion of earth surface temperatures. However, due to the fact that inversion accuracy of the passive microwave earth surface temperature is low, and large scale difference exists between passive microwave data and thermal infrared data, uncertainty of data fusion generated all-weather earth surface temperature products caused by errors of the passive microwave earth surface temperature and uncertainty of scale reduction exists; in addition, the problems of the existing inversion algorithm are as follows: the estimated surface temperature value is not the real surface temperature under the cloud at the satellite transit time, but the theoretical value of the surface temperature under the clear sky condition at the satellite transit time, so that the requirement of practical application is difficult to meet.
Referring to fig. 1, fig. 1 is a flowchart illustrating an embodiment of an all-weather surface temperature generation method based on machine learning according to an embodiment of the present invention, and for convenience of description, only the portions related to the embodiment of the present invention are shown. Specifically, the all-weather surface temperature generation method based on machine learning is executed by a computer.
When the invention is implemented, the all-weather earth surface temperature generation method based on machine learning specifically comprises the following steps:
s101, adopting an MODIS data processing tool MRT, and respectively extracting MODIS data sets corresponding to cloud coverage pixels and cloud coverage-free pixels which are subjected to remote sensing inversion from corresponding MODIS land products. The MODIS data set comprises a normalized vegetation index, an enhanced vegetation index, a leaf area index, earth surface albedo data and earth surface temperature with optimal inversion accuracy.
And S102, combining the data of the stationary meteorological satellite with the DEM topographic data of the ALOS satellite, and estimating and obtaining the earth surface incident solar radiation.
S103, according to the space scale of the earth surface temperature, earth surface incident solar radiation with the same space scale is aggregated to obtain a space scale aggregated data set. And taking the data set aggregated by the spatial scale and the data set without the cloud coverage pixel as a machine learning training data set.
And S104, training the machine learning training data set by selecting a random forest model, and constructing and obtaining a surface temperature relation model. And applying the surface temperature relation model to the data set with the cloud coverage pixel, and estimating and obtaining the surface real temperature of the cloud coverage pixel.
And S105, combining the real surface temperature of the cloud covered pixel with the data set of the cloud-free covered pixel to generate the all-weather surface temperature.
The all-weather earth surface temperature generation method based on machine learning provided by the embodiment of the invention overcomes the existing problems, effectively solves the problems that the current thermal infrared remote sensing is easily influenced by cloud and fog and earth surface temperature products have a large number of blank default areas, realizes the effective acquisition of all-weather earth surface temperature of a cloud coverage area under real conditions, further improves the spatial continuity of the earth surface temperature products, effectively improves the application level of the earth surface temperature products in the aspects of regional hydrology, ecology, agriculture, meteorology and the like, and has important significance in improving the actual service and application level of the earth surface temperature remote sensing products.
In specific implementation, the thermal environment condition of the earth surface is related to a plurality of factors such as earth surface vegetation coverage conditions, earth surface albedo, earth surface atmospheric forcing conditions and the like, and for accurately constructing a relation model between the earth surface temperature and other parameters of the earth surface, it is critical to select appropriate parameters capable of representing the earth surface temperature, wherein earth surface parameters subjected to remote sensing inversion are important data sources.
In step S101, by using an MODIS data processing tool MRT, MODIS data sets corresponding to cloud-covered pixels and cloud-free pixels that are inverted by remote sensing are extracted from corresponding MODIS land products. The MODIS data set comprises a normalized vegetation index, an enhanced vegetation index, a leaf area index, earth surface albedo data and earth surface temperature with optimal inversion accuracy.
The method selects MODIS surface temperature reconstruction as an example, and mainly carries out related parameter extraction work based on MODIS land products; the specific parameters include MOD11A1 ground surface temperature product, MOD13A2 vegetation index product, MCD15A3 leaf area index product, and MCD43A3 ground surface albedo product, and the description of each product can be referred to for the inversion algorithm of the related parameters.
Based on the land product, the invention adopts a MODIS Reprojection Tool (MRT), which is a processing tool for MODIS data. The method can help a user to re-project the MODIS image to a more standard map projection, and can select a space subset and a waveband subset in the image to perform projection conversion. And respectively extracting earth surface temperature, normalized vegetation index (NDVI), Enhanced Vegetation Index (EVI), Leaf Area Index (LAI) and earth surface albedo data by adopting a data processing tool MRT of MODIS.
Further, the data sets corresponding to the cloud-covered pixels and the cloud-free covered pixels respectively adopt MODIS data with a spatial scale of 1 kilometer. And resampling the leaf area index and the earth surface albedo data of 300 m spatial scale to a MODIS data set of 1km spatial scale through spatial aggregation according to the earth surface temperature of 1km spatial scale. That is, the leaf area index and the surface albedo data (500m) are resampled to 1km spatial resolution by spatial aggregation based on the surface temperature spatial resolution (1 km). In addition, according to the quality control file of the corresponding earth surface temperature product, the earth surface temperature data with the optimal inversion precision is selected as a data source for later modeling. Accordingly, the MODIS data set includes surface temperatures that select the best inversion accuracy from the MODIS products.
Furthermore, the surface incident solar radiation is an important energy source for surface temperature increase, and has very important significance on the daily cycle change process of the surface temperature. Therefore, accurate surface incident solar radiation information is required to develop surface temperature relationship model modeling. In the present invention, the solar radiation factor is expressed by the cumulative incident solar radiation, i.e. the total amount of solar radiation received when the sun rises to the satellite observation time. However, in order to accurately represent the situation that the earth surface receives the solar radiation in the process, the invention adopts a cooperative mode of the data of the stationary meteorological satellite and utilizes the characteristic of high time resolution to accurately represent the cloud coverage situation of each pixel from the rising of the sun to the observation time of the satellite, thereby effectively estimating the effective solar radiation of each pixel. Meanwhile, in order to accurately describe the influence of different terrain conditions on solar incident radiation, high-spatial-resolution DEM data, namely 30-meter DEM data of an ALOS satellite, is adopted, and terrain factors such as gradient and slope are introduced to provide terrain data support for estimation of surface incident solar radiation.
Specifically, in step 102, the geostationary weather satellite data is combined with the DEM terrain data of the ALOS satellites to estimate and derive the surface incident solar radiation. The data of the stationary meteorological satellite is data with a spatial scale of 3-5 kilometers, and the DEM topographic data of the ALOS satellite is topographic data with a scale of 30 m.
Further, the surface incident solar radiation fRComprising direct solar radiation RbSky scattered radiation RdAnd adjacent terrain radiation Rr。
The direct solar radiation RbCalculated according to the following formula:
wherein E isoIs the solar constant.
dr is a day-to-ground distance correction coefficient of the day, wherein the calculation formula of dr is as follows:
τbIs the transmission rate of the direct radiation atmosphere.
Theta is an included angle between the sun straight line light and the surface slope surface normal, wherein the calculation formula of cos theta is as follows:
cosθ＝cosZscosS+cosZssinScos(As-A), in which formula ZsIs the zenith angle of the sun, AsThe azimuth angle of the sun, S the slope of the terrain and A the slope of the terrain.
Further, sky scattered radiation R under complex terraindIs to scattered radiation R of sky under flat terraind,flatAll right (1)And (4) obtaining the sky visibility factor SVF through correction. The sky scattered radiation RdCalculated according to the following formula:
Rd＝Rd,flat×SVF。
wherein R isd,flatScattering radiation for the sky under flat terrain, where Rd,flatThe calculation formula of (2) is as follows: rd,flat＝EO×dr×cos(Zs)×τdIn the formula EoAnd d is a solar constant, and dr is a day-to-day distance correction coefficient, wherein the calculation formula of dr is as follows:
τd＝0.271-0.294×τbin the formula τbIs the transmission rate of the direct radiation atmosphere.
SVF is sky visual factor, defines as dividing hemisphere 2 pi space into n equal parts, the ratio of hemisphere visible part area above the target point to hemisphere area, wherein the formula of SVF is:
Further, for the additional radiation generated by the reflection of the surrounding slope, a simplified approximate calculation method of Dozier can be adopted, and only the terrain gradient, the sky view factor and the average reflection effect of the surrounding terrain are considered. Thus, the adjacent terrain radiation RrThe method adopts a Dozier simplified approximate calculation method, and concretely comprises the following steps:
Rr＝ρ×Ct×(Rb+Rd)。
where ρ is the average albedo of the adjacent terrain.
CtAs a topographic structural parameter, CtThe reflecting radiation anisotropic property is included, and the anisotropic property exists because the slope surface around the reflecting radiation anisotropic property has different slopes and slope directions. The device also comprises a slope element which is a slope surface grid unit with a certain slope direction and a certain slope gradient; because only the visible slope elements around a certain slope element have influence on the reflected radiation of the slope element, the geometrical effect between the slope element and the visible slope elements around the slope element, C, assuming that the underlying surface is a lambertian bodytThe calculation formula of (2) is as follows:
Ct(1+ cosS)/2-SVF, where S is slope, SVF is sky visibility factor, and SVF is calculated as described above, where context calculation is consistent.
Specifically, it can be found from the ground observation data size that the ground incident solar radiation is mainly affected by direct radiation and scattered radiation, and the proportion of the reflected radiation of the adjacent terrain is small. Therefore, in the present invention, the adjacent terrain reflected radiation term may be disregarded. To carefully consider differences in surface incident radiation under different terrain conditions, the present invention first uses the ALOS satellite 30m high spatial resolution DEM data for estimating surface incident solar radiation.
Further, based on the above relation, after eliminating the common term, the surface incident solar radiation fRThe calculation can be simplified by the following equation:
wherein, taubIs the transmission rate of the direct radiation atmosphere.
τdIs the transmission of scattered radiation, wheredThe calculation formula of (2) is as follows:
τd＝0.271-0.294×τbin the formula τbIs the transmission rate of the direct radiation atmosphere.
ZsThe zenith angle of the sun.
Theta is an included angle between the sun straight line light and the surface slope surface normal, wherein the calculation formula of cos theta is as follows:
cosθ＝cosZscosS+cosZssinScos(As-A), in which formula ZsIs the zenith angle of the sun, AsThe azimuth angle of the sun, S the slope of the terrain and A the slope of the terrain.
SVF is sky visual factor, defines as dividing hemisphere 2 pi space into n equal parts, the ratio of hemisphere visible part area above the target point to hemisphere area, wherein the formula of SVF is:
VtAnd whether the cloud exists at the moment or not is represented, the value is 1 under the cloud-free coverage condition, and the value is 0 under the cloud-covered condition. The parameter is mainly obtained based on observation data of a stationary meteorological satellite. When the static meteorological data obtain effective earth surface temperature data, the pixel can be considered to be clear sky, otherwise, the pixel is covered by cloud.
Further, the direct radiation atmospheric transmittance τbSpecifically, the calculation is as follows:
τb＝0.56·(e-0.65M+e-0.095M)。
where M is the air mass ratio, i.e. the ratio of the mass of the atmosphere passing through the solar radiation direction to the mass of the atmosphere passing directly through the zenith of the sun, where τdThe calculation formula of (2) is as follows:
p/p0exp (-z/8434.5), where z is altitude.
Wherein the content of the first and second substances,
the solar altitude after the correction of the solar refraction,
the calculation formula of (2) is as follows:
further, after the surface solar incident radiation is obtained, in step S103, the surface incident solar radiation of the same spatial scale is aggregated according to the spatial scale of the surface temperature, so as to obtain a spatial scale aggregated data set. And taking the data set aggregated by the spatial scale and the data set without the cloud coverage pixel as a machine learning training data set.
In specific implementation, as can be seen from the estimation process, the data sets of different spatial scales are adopted in the method, and the data sets comprise MODIS data of 1km scale, geostationary satellite data of 3-5km scale and topographic data of 30m scale. In order to accurately characterize the change process of the earth surface temperature with the 1km scale, an important step is the problem of spatial matching of data with different spatial scales. The method mainly establishes a spatial correspondence relationship between the coarse resolution pixel and the fine resolution pixel through longitude and latitude information of each pixel in each spatial data set, and aggregates solar radiation, topographic factors and the like estimated on a 30m scale to the same spatial scale as the surface temperature of the MODIS by adopting a spatial aggregation mode on the basis to form a data set with consistent space. Meanwhile, extracting clear sky cloud-free pixels and extracting all surface parameters to form a machine learning training data set by combining MODIS surface temperature data quality information; and the residual cloud coverage pixel is an application data set and is used for estimating the subsurface temperature data under the cloud coverage of the pixel.
Further, in steps S104 and S105, a random forest model is selected to train the machine learning training data set, and a surface temperature relationship model is constructed and obtained. And applying the surface temperature relation model to the data set with the cloud coverage pixel, and estimating and obtaining the surface real temperature of the cloud coverage pixel. And combining the real surface temperature of the cloud covered pixel with the data set of the cloud-free covered pixel to generate the all-weather surface temperature.
In specific implementation, the high-precision modeling of the earth surface temperature relation model is a premise for realizing accurate estimation of the earth surface temperature under the cloud coverage condition. According to research and early-stage research results, the invention selects a Random Forest Regression (Random Forest Regression) method to realize the establishment of the surface temperature relationship model.
Random Forest (RF) is a machine learning model proposed by Breiman, the essence of which is a decision tree algorithm, but the result of combining multiple decision trees is changed from a single decision tree. Thus, a random forest is made up of a plurality of decision trees, and there is no correlation between each decision tree in the forest, and the final output of the model is jointly determined by each decision tree in the forest. According to the method, through a bootstrap resampling technology, a plurality of samples extracted from original training samples are combined by self to generate a training sample collection; multiple decision trees are then generated from the bootstrap sample set and composed into RFs, with their classification or regression model results being based on decision tree voting scores. In the invention, the relationship between the surface temperature and various parameters is not linear, and the RF model is not sensitive to the multivariate collinearity, thereby effectively preventing overfitting.
Therefore, the earth surface temperature relation model is firstly constructed on the basis of the training set without cloud pixels in clear sky, the prediction result is more stable to the missing data and the unbalanced data, and a large number of input variables can be processed and better estimation accuracy can be obtained for various observation data adopted by the method. Therefore, the method does not obviously improve the operation amount while improving the prediction precision, and has more obvious advantages compared with the traditional least square linear regression fitting.
Finally, the surface temperature relation model generated by training of the random forest method is applied to the cloud coverage image metadata set, the real surface temperature of the cloud coverage image element is estimated, and is combined with the clear sky image element to form an all-weather surface temperature data set, specific application comparison effects are shown in fig. 2 and fig. 3, the fig. 2 shows original MODIS daytime surface temperature product data, wherein due to cloud coverage, most of north and south regions of a research area are subjected to null values. FIG. 3 shows that the earth surface temperature data reconstructed by the method of the present invention has the advantage that the earth surface temperature information of most of the cloud coverage pixels is effectively recovered, and the earth surface temperature data has a reasonable spatial distribution pattern.
In summary, the all-weather earth surface temperature generation method based on machine learning provided by the embodiment of the invention overcomes the existing problems, effectively solves the problems that the current thermal infrared remote sensing is easily affected by cloud and fog and earth surface temperature products have a large number of blank default areas, realizes effective acquisition of all-weather earth surface temperature of a cloud coverage area under real conditions, further improves the spatial continuity of the earth surface temperature products, effectively improves the application level of the earth surface temperature products in the aspects of regional hydrology, ecology, agriculture, meteorology and the like, and has important significance in improving the actual service and application level of the earth surface temperature remote sensing products.
Referring to fig. 4, fig. 4 is a schematic structural diagram illustrating an embodiment of an all-weather surface temperature generation apparatus based on machine learning according to an embodiment of the present invention, and for convenience of description, only the parts related to the embodiment of the present invention are shown. Specifically, the all-weather surface temperature generation device 10 based on machine learning includes:
the MODIS data set obtaining module 11 is configured to use a data processing tool MRT of the MODIS to respectively extract, from corresponding MODIS land products, MODIS data sets corresponding to the cloud-covered pixels and the cloud-free pixels that are inverted by remote sensing. The MODIS data set comprises a normalized vegetation index, an enhanced vegetation index, a leaf area index, earth surface albedo data and earth surface temperature with optimal inversion accuracy.
And the earth surface incident solar radiation acquisition module 12 is used for combining the data of the stationary meteorological satellite and the DEM topographic data of the ALOS satellite to estimate and obtain the earth surface incident solar radiation.
And the machine learning training data set acquisition module 13 is configured to aggregate the surface incident solar radiation with the same spatial scale according to the spatial scale of the surface temperature to obtain a spatial scale aggregated data set. And taking the data set aggregated by the spatial scale and the data set without the cloud coverage pixel as a machine learning training data set.
And the surface real temperature estimation module 14 is used for selecting a random forest model to train the machine learning training data set, and constructing and obtaining a surface temperature relation model. And applying the surface temperature relation model to the data set with the cloud coverage pixel, and estimating and obtaining the surface real temperature of the cloud coverage pixel.
And the all-weather ground surface temperature generating module 15 is used for combining the real ground surface temperature with the cloud coverage pixel with the data set without the cloud coverage pixel to generate the all-weather ground surface temperature.
Further, in the device, the data sets corresponding to the cloud coverage pixels and the cloud-free coverage pixels respectively adopt MODIS data with a spatial scale of 1 kilometer, the data of the geostationary weather satellite adopts data with a spatial scale of 3-5 kilometers, and the DEM topographic data of the ALOS satellite adopts topographic data with a scale of 30 m.
And resampling the leaf area index and the earth surface albedo data of 300 m spatial scale to a MODIS data set of 1km spatial scale through spatial aggregation according to the earth surface temperature of 1km spatial scale.
Further, in the apparatus, the surface incident solar radiation fRComprising direct solar radiation RbSky scattered radiation RdAnd adjacent terrain radiation Rr。
The direct solar radiation RbCalculated according to the following formula:
wherein E isoIs the solar constant.
dr is a day-to-ground distance correction coefficient of the day, wherein the calculation formula of dr is as follows:
τbIs the transmission rate of the direct radiation atmosphere.
Theta is an included angle between the sun straight line light and the surface slope surface normal, wherein the calculation formula of cos theta is as follows:
cosθ＝cosZscosS+cosZssinScos(As-A), in which formula ZsIs the zenith angle of the sun, AsThe azimuth angle of the sun, S the slope of the terrain and A the slope of the terrain.
Further, in the apparatus, the sky scattered radiation RdCalculated according to the following formula:
Rd＝Rd,flat×SVF。
wherein R isd,flatScattering radiation for the sky under flat terrain, where Rd,flatThe calculation formula of (2) is as follows: rd,flat＝EO×dr×cos(Zs)×τdIn the formula EoAnd d is a solar constant, and dr is a day-to-day distance correction coefficient, wherein the calculation formula of dr is as follows:
τd＝0.271-0.294×τbin the formula τbIs the transmission rate of the direct radiation atmosphere.
SVF is sky visual factor, defines as dividing hemisphere 2 pi space into n equal parts, the ratio of hemisphere visible part area above the target point to hemisphere area, wherein the formula of SVF is:
Further, in the apparatus, the adjacent terrain radiation RrThe method adopts a Dozier simplified approximate calculation method, and concretely comprises the following steps:
Rr＝ρ×Ct×(Rb+Rd)。
where ρ is the average albedo of the adjacent terrain.
CtIs a topographic structural parameter, wherein CtThe calculation formula of (2) is as follows:
Ct(1+ cosS)/2-SVF, where S is slope and SVF is sky visibility factor.
Further, in the apparatus, the surface incident solar radiation fRThe calculation can be simplified by the following equation:
wherein, taubIs the transmission rate of the direct radiation atmosphere.
τdIs the transmission of scattered radiation, wheredThe calculation formula of (2) is as follows:
τd＝0.271-0.294×τbin the formula τbIs the transmission rate of the direct radiation atmosphere.
ZsThe zenith angle of the sun.
Theta is an included angle between the sun straight line light and the surface slope surface normal, wherein the calculation formula of cos theta is as follows:
cosθ＝cosZscosS+cosZssinScos(As-A), in which formula ZsIs the zenith angle of the sun, AsThe azimuth angle of the sun, S the slope of the terrain and A the slope of the terrain.
SVF is sky visual factor, defines as dividing hemisphere 2 pi space into n equal parts, the ratio of hemisphere visible part area above the target point to hemisphere area, wherein the formula of SVF is:
VtAnd whether the cloud exists at the moment or not is represented, the value is 1 under the cloud-free coverage condition, and the value is 0 under the cloud-covered condition.
Further, in the apparatus, the direct radiation atmospheric transmittance τ isbSpecifically, the calculation is as follows:
τb＝0.56·(e-0.65M+e-0.095M)。
where M is the air mass ratio, i.e. the ratio of the mass of the atmosphere passing through the solar radiation direction to the mass of the atmosphere passing directly through the zenith of the sun, where τdThe calculation formula of (2) is as follows:
p/p0exp (-z/8434.5), where z is altitude.
Wherein the content of the first and second substances,
the solar altitude after the correction of the solar refraction,
the calculation formula of (2) is as follows:
it should be noted that, in the present specification, the embodiments are all described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same and similar parts among the embodiments may be referred to each other. For the device or system type embodiment, since it is basically similar to the method embodiment, the description is simple, and for the relevant points, refer to the partial description of the method embodiment.
Referring to fig. 5, fig. 5 is a schematic structural diagram of an embodiment of a computer device according to an embodiment of the present invention, and for convenience of description, only portions related to the embodiment of the present invention are shown. Specifically, the computer device 500 includes a memory 502, a processor 501 and a computer program 5021 stored in the memory 502 and operable on the processor 501, and when the processor 501 executes the computer program, the steps of the method according to the above embodiment, such as the steps S101 to S105 shown in fig. 1, are implemented. Alternatively, the processor 501, when executing the computer program, implements the functions of each module/unit in the apparatus according to the above embodiment, for example, the functions of the modules 11 to 15 shown in fig. 2.
Illustratively, the computer program may be partitioned into one or more modules/units that are stored in the memory 502 and executed by the processor 501 to implement the present invention. The one or more modules/units may be a series of computer program instruction segments capable of performing specific functions, which are used to describe the execution of the computer program in the computer device 500. For example, the computer program may be partitioned into a MODIS dataset acquisition module 11, a surface incident solar radiation acquisition module 12, a machine learning training dataset acquisition module 13, a surface true temperature estimation module 14, an all-weather surface temperature generation module 15.
The MODIS data set obtaining module 11 is configured to use a data processing tool MRT of the MODIS to respectively extract, from corresponding MODIS land products, MODIS data sets corresponding to the cloud-covered pixels and the cloud-free pixels that are inverted by remote sensing. The MODIS data set comprises a normalized vegetation index, an enhanced vegetation index, a leaf area index, earth surface albedo data and earth surface temperature with optimal inversion accuracy.
And the earth surface incident solar radiation acquisition module 12 is used for combining the data of the stationary meteorological satellite and the DEM topographic data of the ALOS satellite to estimate and obtain the earth surface incident solar radiation.
And the machine learning training data set acquisition module 13 is configured to aggregate the surface incident solar radiation with the same spatial scale according to the spatial scale of the surface temperature to obtain a spatial scale aggregated data set. And taking the data set aggregated by the spatial scale and the data set without the cloud coverage pixel as a machine learning training data set.
And the surface real temperature estimation module 14 is used for selecting a random forest model to train the machine learning training data set, and constructing and obtaining a surface temperature relation model. And applying the surface temperature relation model to the data set with the cloud coverage pixel, and estimating and obtaining the surface real temperature of the cloud coverage pixel.
And the all-weather ground surface temperature generating module 15 is used for combining the real ground surface temperature with the cloud coverage pixel with the data set without the cloud coverage pixel to generate the all-weather ground surface temperature.
The computer device 500 may include, but is not limited to, a processor 501, a memory 502. Those skilled in the art will appreciate that the figure is merely an example of a computer device 500 and is not intended to limit the computer device 500 and that the computer device 500 may include more or less components than those shown, or some of the components may be combined, or different components, for example, the computer device 500 may also include input output devices, network access devices, buses, and the like.
The Processor 501 may be a Central Processing Unit (CPU), other general-purpose Processor 501, a Digital Signal Processor 501 (DSP), an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA) or other programmable logic device, discrete Gate or transistor logic, discrete default hardware components, and so on. The general purpose processor 501 may be a microprocessor 501 or the processor 501 may be any conventional processor 501 or the like.
The memory 502 may be an internal storage unit of the computer device 500, such as a hard disk or a memory of the computer device 500. The memory 502 may also be an external storage device of the computer device 500, such as a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash memory Card (Flash Card), etc. provided on the computer device 500. Further, the memory 502 may also include both internal and external storage for the computer device 500. The memory 502 is used for storing the computer program 5021 as well as other programs and data required by the computer device 500. The memory 502 may also be used to temporarily store data that has been output or is to be output.
Embodiments of the present invention further provide a computer-readable storage medium, which stores a computer program, and when the computer program is executed by the processor 501, the computer program implements steps in the method described in the above embodiments, such as steps S101 to S105 shown in fig. 1. Alternatively, the computer program realizes the functions of the modules/units in the apparatus described in the above embodiments, such as the functions of the modules 11 to 15 shown in fig. 2, when being executed by the processor 501.
The computer program may be stored in a computer readable storage medium, which when executed by the processor 501, may implement the steps of the various method embodiments described above. Wherein the computer program comprises computer program code, which may be in the form of source code, object code, an executable file or some intermediate form, etc. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, usb disk, removable hard disk, magnetic disk, optical disk, computer Memory, Read-Only Memory (ROM), Random Access Memory (RAM), electrical carrier wave signals, telecommunications signals, software distribution medium, and the like.
It should be noted that the computer readable medium may contain other components which may be suitably increased or decreased as required by legislation and patent practice in jurisdictions, for example, in some jurisdictions, computer readable media which may not include electrical carrier signals and telecommunications signals in accordance with legislation and patent practice.
In the above embodiments, the descriptions of the respective embodiments have respective emphasis, and reference may be made to the related descriptions of other embodiments for parts that are not described or illustrated in a certain embodiment.
The steps in the method of the embodiment of the invention can be sequentially adjusted, combined and deleted according to actual needs.
The modules or units in the system of the embodiment of the invention can be combined, divided and deleted according to actual needs.
Those of ordinary skill in the art would appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic pre-set hardware or in a combination of computer software and electronic pre-set hardware. Whether these functions are performed by pre-determined hardware or software depends on the particular application and design constraints imposed on the solution. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present invention.
In the embodiments provided by the present invention, it should be understood that the disclosed apparatus/computer device 500 and method may be implemented in other ways. For example, the above-described embodiment of apparatus/computer device 500 is merely illustrative, and for example, the division of the modules or units is only one logical division, and there may be other divisions when actually implemented, e.g., multiple units or components may be combined or integrated into another system, or some features may be omitted, or not implemented. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection through some interfaces, devices or units, and may be in an electrical, mechanical or other form.
The above-mentioned embodiments are only used for illustrating the technical solutions of the present invention, and not for limiting the same; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; such modifications and substitutions do not substantially depart from the spirit and scope of the embodiments of the present invention, and are intended to be included within the scope of the present invention.
Claims (9)
1. An all-weather surface temperature generation method based on machine learning, characterized in that the method comprises:
respectively extracting MODIS data sets corresponding to cloud coverage pixels and cloud coverage-free pixels which are subjected to remote sensing inversion in corresponding MODIS land products by adopting a data processing tool MRT of the MODIS; the MODIS data set comprises a normalized vegetation index, an enhanced vegetation index, a leaf area index, earth surface albedo data and earth surface temperature with optimal inversion precision;
combining static meteorological satellite data with DEM topographic data of an ALOS satellite, estimating and obtaining earth surface incident solar radiation;
according to the spatial scale of the surface temperature, polymerizing surface incident solar radiation with the same spatial scale to obtain a data set polymerized with the spatial scale, wherein MODIS data with the spatial scale of 1 kilometer are selected for data sets corresponding to the cloud coverage pixels and the cloud coverage pixels, data with the spatial scale of 3-5 kilometers are selected for stationary meteorological satellite data, and topographic data with the scale of 30m is selected for DEM topographic data of the ALOS satellite; resampling leaf area index and earth surface albedo data of 500m spatial scale to MODIS data set of 1km spatial scale through spatial aggregation according to earth surface temperature of 1km spatial scale; taking the data set aggregated by the spatial scale and the data set without the cloud coverage pixel as a machine learning training data set;
training the machine learning training data set by selecting a random forest model, and constructing and obtaining a surface temperature relation model; applying the surface temperature relation model to the data set with the cloud coverage pixel, and estimating and obtaining the surface real temperature of the cloud coverage pixel;
and combining the real surface temperature of the cloud covered pixel with the data set of the cloud-free covered pixel to generate the all-weather surface temperature.
2. The all-weather surface temperature generation method based on machine learning of claim 1, wherein the surface incident solar radiation fRComprising direct solar radiation RbSky scattered radiation RdAnd adjacent terrain radiation Rr；
The direct solar radiation RbCalculated according to the following formula:
wherein E isoIs the solar constant;
dr is a day-to-ground distance correction coefficient of the day, wherein the calculation formula of dr is as follows:
theta is an included angle between the sun straight line light and the surface slope surface normal, wherein the calculation formula of cos theta is as follows:
cosθ＝cosZscosS+cosZssinScos(As-A), in which formula ZsIs the zenith angle of the sun, AsThe azimuth angle of the sun, S the slope of the terrain and A the slope of the terrain.
3. The machine-learning-based all-weather surface temperature generation method of claim 2, wherein the sky scattered radiation RdCalculated according to the following formula:
Rd＝Rd,flat×SVF；
wherein R isd,flatScattering radiation for the sky under flat terrain, where Rd,flatThe calculation formula of (2) is as follows: rd,flat＝EO×dr×cos(Zs)×τdIn the formula EoAnd d is a solar constant, and dr is a day-to-day distance correction coefficient, wherein the calculation formula of dr is as follows:
τd＝0.271-0.294×τbin the formula τbThe atmospheric transmittance is directly radiated;
SVF is sky visual factor, defines as dividing hemisphere 2 pi space into n equal parts, the ratio of hemisphere visible part area above the target point to hemisphere area, wherein the formula of SVF is:
4. The machine learning-based all-weather surface temperature generation method of claim 3, wherein the adjacent terrain radiation RrThe method adopts a Dozier simplified approximate calculation method, and concretely comprises the following steps:
Rr＝ρ×Ct×(Rb+Rd)；
wherein ρ is the average albedo of adjacent terrain;
Ctis a topographic structural parameter, wherein CtThe calculation formula of (2) is as follows:
Ct(1+ cosS)/2-SVF, where S is slope and SVF is sky visibility factor.
5. The all-weather surface temperature generation method based on machine learning of claim 4, wherein the surface incident solar radiation fRThe calculation can be simplified by the following equation:
wherein, taub,tThe direct radiation atmospheric transmittance at time t;
τd,tis the transmission of scattered radiation at time t, whered,tThe calculation formula of (2) is as follows:
τd,t＝0.271-0.294×τb,tin the formula τb,tThe direct radiation atmospheric transmittance at time t;
Zsis the solar zenith angle;
theta is an included angle between the sun straight line light and the surface slope surface normal, wherein the calculation formula of cos theta is as follows:
cosθ＝cosZscosS+cosZssinScos(As-A), in which formula ZsIs the zenith angle of the sun, AsThe azimuth angle of the sun, S the terrain slope and A the terrain slope direction;
SVF is sky visual factor, defines as dividing hemisphere 2 pi space into n equal parts, the ratio of hemisphere visible part area above the target point to hemisphere area, wherein the formula of SVF is:
Vtand whether the cloud exists at the moment or not is represented, the value is 1 under the cloud-free coverage condition, and the value is 0 under the cloud-covered condition.
6. The all-weather surface temperature generation method based on machine learning of claim 5, wherein the direct radiation atmospheric transmittance τ isbSpecifically, the calculation is as follows:
τb＝0.56·(e-0.65M+e-0.095M)；
wherein, M is the air mass ratio, namely the ratio of the mass of the atmosphere passing through the solar radiation direction to the mass of the atmosphere passing through the direct sunlight at the zenith, wherein the calculation formula of M is as follows:
p/p0exp (-z/8434.5), where z is altitude;
wherein the content of the first and second substances,
the solar altitude after the correction of the solar refraction,
the calculation formula of (2) is as follows:
7. an all-weather surface temperature generation apparatus based on machine learning, the apparatus comprising:
the MODIS data set acquisition module is used for respectively extracting MODIS data sets corresponding to cloud coverage pixels and cloud coverage-free pixels which are subjected to remote sensing inversion in corresponding MODIS land products by adopting a data processing tool MRT of the MODIS; the MODIS data set comprises a normalized vegetation index, an enhanced vegetation index, a leaf area index, earth surface albedo data and earth surface temperature with optimal inversion precision;
the earth surface incident solar radiation acquisition module is used for combining static meteorological satellite data with DEM topographic data of an ALOS satellite, estimating and obtaining earth surface incident solar radiation;
the machine learning training data set acquisition module is used for aggregating the earth surface incident solar radiation with the same spatial scale according to the spatial scale of the earth surface temperature to obtain a data set aggregated by the spatial scale; taking the data set aggregated by the spatial scale and the data set without the cloud coverage pixel as a machine learning training data set;
the earth surface real temperature estimation module is used for selecting a random forest model to train the machine learning training data set, and constructing and obtaining an earth surface temperature relation model; applying the surface temperature relation model to the data set with the cloud coverage pixel, and estimating and obtaining the surface real temperature of the cloud coverage pixel;
and the all-weather surface temperature generating module is used for combining the real surface temperature with the cloud coverage pixel with the data set without the cloud coverage pixel to generate the all-weather surface temperature.
8. A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor when executing the computer program implements the all-weather surface temperature generation method based on machine learning of any one of claims 1 to 6.
9. A computer storage medium having a computer program stored thereon, wherein the program, when executed by a processor, implements the all-weather surface temperature generation method based on machine learning of any one of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910812200.XA CN110516816B (en) | 2019-08-30 | 2019-08-30 | All-weather earth surface temperature generation method and device based on machine learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910812200.XA CN110516816B (en) | 2019-08-30 | 2019-08-30 | All-weather earth surface temperature generation method and device based on machine learning |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110516816A CN110516816A (en) | 2019-11-29 |
| CN110516816B true CN110516816B (en) | 2021-05-28 |
Family
ID=68629222
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910812200.XA Active CN110516816B (en) | 2019-08-30 | 2019-08-30 | All-weather earth surface temperature generation method and device based on machine learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110516816B (en) |
Families Citing this family (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111652404B (en) * | 2020-02-06 | 2023-04-28 | 北京师范大学 | All-weather earth surface temperature inversion method and system |
| CN111982294B (en) * | 2020-07-21 | 2022-06-03 | 电子科技大学 | All-weather earth surface temperature generation method integrating thermal infrared and reanalysis data |
| CN112560570B (en) * | 2020-09-29 | 2023-07-25 | 中国科学院大学 | High-resolution earth surface temperature estimation method for collaborative downscaling and data fusion |
| CN112487346B (en) * | 2020-10-26 | 2021-07-23 | 中国农业科学院农业资源与农业区划研究所 | Mountain land surface temperature remote sensing retrieval method |
| CN112818605B (en) * | 2021-02-07 | 2022-04-26 | 武汉大学 | Method and system for rapidly estimating earth surface albedo |
| CN113158570A (en) * | 2021-04-26 | 2021-07-23 | 电子科技大学 | All-weather surface temperature near-real-time inversion method fusing multi-source satellite remote sensing |
| CN113255148A (en) * | 2021-06-04 | 2021-08-13 | 中国科学院地理科学与资源研究所 | Method for estimating all-weather air temperature and space-time distribution thereof based on MODIS product data |
| CN113408742B (en) * | 2021-06-24 | 2023-06-02 | 桂林理工大学 | High-precision sea surface temperature inversion method based on machine learning |
| CN114169215B (en) * | 2021-08-06 | 2022-11-15 | 中国矿业大学(北京) | Surface temperature inversion method coupling remote sensing and regional meteorological model |
| CN113743000B (en) * | 2021-08-13 | 2023-04-18 | 电子科技大学 | Method for generating all-weather surface temperature with high time resolution |
| CN114781148A (en) * | 2022-04-14 | 2022-07-22 | 河北地质大学 | Surface temperature inversion method and system for thermal infrared remote sensing cloud coverage pixel |
| CN117390969B (en) * | 2023-12-05 | 2024-03-12 | 易智瑞信息技术有限公司 | Method for generating surface temperature, electronic equipment and storage medium |
| CN117554300B (en) * | 2024-01-10 | 2024-03-19 | 中国科学院、水利部成都山地灾害与环境研究所 | Remote sensing space downscaling method for mountain land surface albedo site observation |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103293522A (en) * | 2013-05-08 | 2013-09-11 | 中国科学院光电研究院 | Intermediate infrared two-channel remote sensing data surface temperature inversion method and device |
| CN107065036A (en) * | 2017-04-19 | 2017-08-18 | 中国农业科学院农业资源与农业区划研究所 | A kind of method that joint remote sensing and meteorological data obtain round-the-clock evapotranspiration |
| CN107748736A (en) * | 2017-10-13 | 2018-03-02 | 河海大学 | A kind of multiple-factor Remote Sensing temperature space NO emissions reduction method based on random forest |
| CN108388956A (en) * | 2018-01-18 | 2018-08-10 | 华北电力大学 | Consider the photovoltaic power prediction technique of attenuation |
| CN109635309A (en) * | 2018-10-17 | 2019-04-16 | 广州地理研究所 | A kind of surface temperature space NO emissions reduction method |
| CN109885959A (en) * | 2019-03-05 | 2019-06-14 | 中国科学院地理科学与资源研究所 | A kind of surface temperature robust NO emissions reduction method |
| CN110175375A (en) * | 2019-05-13 | 2019-08-27 | 中国科学院遥感与数字地球研究所 | A kind of earth's surface Calculation method for solar radiation based on deep learning |
-
2019
- 2019-08-30 CN CN201910812200.XA patent/CN110516816B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103293522A (en) * | 2013-05-08 | 2013-09-11 | 中国科学院光电研究院 | Intermediate infrared two-channel remote sensing data surface temperature inversion method and device |
| CN107065036A (en) * | 2017-04-19 | 2017-08-18 | 中国农业科学院农业资源与农业区划研究所 | A kind of method that joint remote sensing and meteorological data obtain round-the-clock evapotranspiration |
| CN107748736A (en) * | 2017-10-13 | 2018-03-02 | 河海大学 | A kind of multiple-factor Remote Sensing temperature space NO emissions reduction method based on random forest |
| CN108388956A (en) * | 2018-01-18 | 2018-08-10 | 华北电力大学 | Consider the photovoltaic power prediction technique of attenuation |
| CN109635309A (en) * | 2018-10-17 | 2019-04-16 | 广州地理研究所 | A kind of surface temperature space NO emissions reduction method |
| CN109885959A (en) * | 2019-03-05 | 2019-06-14 | 中国科学院地理科学与资源研究所 | A kind of surface temperature robust NO emissions reduction method |
| CN110175375A (en) * | 2019-05-13 | 2019-08-27 | 中国科学院遥感与数字地球研究所 | A kind of earth's surface Calculation method for solar radiation based on deep learning |
Non-Patent Citations (2)
| Title |
|---|
| The Preliminary Investigation on the Uncertainties Associated with surface solar radiation estimation in mountainous areas;Pan Huang，Wei Zhao;《IEEE Geoscience and remote sensing letters》;20170731;第14卷(第7期);第1071-1075页 * |
| 基于随机森林算法的近地表气温遥感反演研究;白琳等;《地球信息科学》;20170331;第19卷(第3期);第390-397页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110516816A (en) | 2019-11-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110516816B (en) | All-weather earth surface temperature generation method and device based on machine learning | |
| CN102539336B (en) | Method and system for estimating inhalable particles based on HJ-1 satellite | |
| CN108268735A (en) | Soil moisture NO emissions reduction method based on multi-source remote sensing satellite fused data | |
| Shi et al. | Spatial downscaling of monthly TRMM precipitation based on EVI and other geospatial variables over the Tibetan Plateau from 2001 to 2012 | |
| CN102298150B (en) | Global land cover broadband emissivity inversion method and system | |
| Meng et al. | Modeling alpine grassland above ground biomass based on remote sensing data and machine learning algorithm: A case study in east of the Tibetan Plateau, China | |
| CN103994976A (en) | MODIS data-based agricultural drought remote sensing monitoring method | |
| CN108982548A (en) | A kind of soil moisture inversion method based on passive microwave remote sensing data | |
| CN108647401A (en) | A kind of basin nitrogen and phosphorus pollution appraisal procedure based on space remote sensing technology | |
| Zhang et al. | Development of the direct-estimation albedo algorithm for snow-free Landsat TM albedo retrievals using field flux measurements | |
| Lou et al. | An effective method for canopy chlorophyll content estimation of marsh vegetation based on multiscale remote sensing data | |
| Zhou et al. | Assessment of bidirectional reflectance effects on desert and forest for radiometric cross-calibration of satellite sensors | |
| Qiu et al. | Estimation of European terrestrial ecosystem NEP based on an improved CASA model | |
| CN112285808B (en) | Method for reducing scale of APHRODITE precipitation data | |
| Lou et al. | Recognition of thaw slumps based on machine learning and UAVs: A case study in the Qilian Mountains, northeastern Qinghai-Tibet Plateau | |
| Han et al. | Refined land-cover classification mapping using a multi-scale transformation method from remote sensing, unmanned aerial vehicle, and field surveys in Sanjiangyuan National Park, China | |
| CN111199092A (en) | Solar radiation remote sensing estimation method and system and data processing device | |
| Ning et al. | Soil moisture at 30 m from multiple satellite datasets fused by random forest | |
| Gueymard et al. | Surface albedo spatial variability in North America: Gridded data vs. local measurements | |
| CN111914396B (en) | Sub-grid terrain three-dimensional earth surface solar radiation forced effect rapid parameterization method based on high-resolution DEM data | |
| Yang et al. | Validation of the MODIS LAI product in Qinghai Lake Basin combined with field measurements using Landsat 8 OLI data | |
| Yue et al. | Soil moisture assessment through the SSMMI and GSSIM algorithm based on SPOT, WorldView-2, and Sentinel-2 images in the Daliuta Coal Mining Area, China | |
| Wang et al. | Simulation of precise scale land use change based on the Markov-cellular automata model | |
| Zhang et al. | Analysis of influence mechanism of spatial distribution of incoming solar radiation based on DEM | |
| Han et al. | Multi-sensor high spatial resolution leaf area index estimation by combining surface reflectance with vegetation indices for highly heterogeneous regions: A case study of the Chishui River Basin in southwest China |
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 |