CN113158570A - All-weather surface temperature near-real-time inversion method fusing multi-source satellite remote sensing - Google Patents
All-weather surface temperature near-real-time inversion method fusing multi-source satellite remote sensing Download PDFInfo
- Publication number
- CN113158570A CN113158570A CN202110451765.7A CN202110451765A CN113158570A CN 113158570 A CN113158570 A CN 113158570A CN 202110451765 A CN202110451765 A CN 202110451765A CN 113158570 A CN113158570 A CN 113158570A
- Authority
- CN
- China
- Prior art keywords
- surface temperature
- temperature
- time
- day
- period
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 45
- 238000007637 random forest analysis Methods 0.000 claims abstract description 27
- 238000013507 mapping Methods 0.000 claims abstract description 17
- 238000003384 imaging method Methods 0.000 claims abstract description 15
- 238000012937 correction Methods 0.000 claims abstract description 3
- 230000010287 polarization Effects 0.000 claims description 9
- 238000012795 verification Methods 0.000 claims description 7
- 238000012216 screening Methods 0.000 claims description 4
- 238000002310 reflectometry Methods 0.000 claims description 2
- 238000012544 monitoring process Methods 0.000 abstract description 7
- 238000012545 processing Methods 0.000 abstract description 6
- 238000004519 manufacturing process Methods 0.000 abstract description 4
- NEBFIUZIGRTIFY-BJDJZHNGSA-N Ala-Met-Ser-Arg Chemical compound CSCC[C@H](NC(=O)[C@H](C)N)C(=O)N[C@@H](CO)C(=O)N[C@H](C(O)=O)CCCNC(N)=N NEBFIUZIGRTIFY-BJDJZHNGSA-N 0.000 description 7
- 238000009826 distribution Methods 0.000 description 6
- 230000008569 process Effects 0.000 description 6
- 238000010586 diagram Methods 0.000 description 5
- 230000006870 function Effects 0.000 description 4
- 230000005855 radiation Effects 0.000 description 4
- 238000012549 training Methods 0.000 description 4
- 238000004422 calculation algorithm Methods 0.000 description 3
- 238000006243 chemical reaction Methods 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 238000007781 pre-processing Methods 0.000 description 3
- 238000007619 statistical method Methods 0.000 description 3
- 238000012360 testing method Methods 0.000 description 3
- 230000009466 transformation Effects 0.000 description 3
- 238000011161 development Methods 0.000 description 2
- 238000012417 linear regression Methods 0.000 description 2
- 238000010801 machine learning Methods 0.000 description 2
- 230000002093 peripheral effect Effects 0.000 description 2
- 238000001556 precipitation Methods 0.000 description 2
- 238000003908 quality control method Methods 0.000 description 2
- 230000002159 abnormal effect Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000005520 cutting process Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 230000002349 favourable effect Effects 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000001360 synchronised effect Effects 0.000 description 1
- 230000009897 systematic effect Effects 0.000 description 1
- 238000012876 topography Methods 0.000 description 1
- 230000007704 transition Effects 0.000 description 1
- 238000010200 validation analysis Methods 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
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/243—Classification techniques relating to the number of classes
- G06F18/24323—Tree-organised classifiers
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2111/00—Details relating to CAD techniques
- G06F2111/08—Probabilistic or stochastic CAD
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
- G06F2119/08—Thermal analysis or thermal optimisation
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Evolutionary Computation (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Software Systems (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Engineering & Computer Science (AREA)
- Medical Informatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Computer Hardware Design (AREA)
- Geometry (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Radiation Pyrometers (AREA)
Abstract
The invention discloses an all-weather earth surface temperature near-real-time inversion method fusing multi-source satellite remote sensing, and belongs to the technical field of satellite remote sensing earth surface temperature. The method comprises the steps of firstly constructing a regression mapping model between brightness temperature data of a microwave radiometer and passive microwave brightness temperature data, then constructing an estimation model of passive microwave brightness temperature in a track clearance area, obtaining seamless passive microwave brightness temperature in a specified period of a target area, constructing a random forest regression relation based on the microwave brightness temperature data of the passive microwave data of the target area and the earth surface temperature of a medium-resolution imaging spectrometer, obtaining an all-weather earth surface temperature estimation value in a reconstruction period, and carrying out system error correction processing based on the earth surface temperature of the corresponding medium-resolution imaging spectrometer. The method does not need to rely on the complete cycle data, can meet the data production requirements of different application requirements and monitoring time periods with different lengths, has good precision and higher image quality, and can show more space detail information of the earth surface temperature.
Description
Technical Field
The invention belongs to the technical field of satellite remote sensing earth surface temperature, and particularly relates to an all-weather earth surface temperature near-real-time inversion method fusing multi-source satellite remote sensing.
Background
Surface Temperature (LST) is a key physical quantity during the interaction of the surface with the atmosphere. The spatial-temporal distribution of the LST in regional and global scales is mastered, and especially comprehensive, complete and continuous information has important significance. With the continuous progress and development of remote sensing technology, the launching of sensors of various platforms makes it possible to acquire regional and global LSTs quickly and in real time. LST is of great importance for surface emissivity (LSE) studies between earth-gas radiation and equilibrium, which have become an important component of the field of quantitative remote sensing. The algorithm for inverting the LST by using the optical remote sensing is mature, the product precision is high, and the resolution basically can meet the requirements of practical application. However, due to the influence of cloud coverage, the integrity and continuity of LST spatio-temporal distribution are greatly damaged, and the application and development of LST products are seriously hindered and limited. Dynamic monitoring of resources and environments (e.g., climate change monitoring, agricultural drought prediction, etc.) across regions and the globe requires high-resolution all-weather LSTs. Over the last 5 years, integrating satellite thermal infrared remote sensing (TIR) and passive Microwave (MW) has proven to be a practical and feasible way to estimate all-weather LST. However, at present, all methods are focused on reconstructing historical archived data, and cannot meet the requirement of real-time or near real-time (NRT) monitoring on the earth surface. Meanwhile, the problem of data loss of the orbit gap region caused by the satellite polar orbit operation mode of passive microwave data (such as AMSR2) is not well solved, in other words, all-weather real sense cannot be realized by fusing passive microwave and thermal infrared remote sensing.
Disclosure of Invention
Aiming at the problems, the invention provides an all-weather surface temperature near-real-time inversion method fusing multi-source satellite remote sensing.
The technical scheme of the invention comprises the following steps:
step 1: acquiring brightness temperature data of a target area, brightness temperature data of passive microwaves and surface temperature data of a resolution imaging spectrometer in the target area;
step 2: constructing a regression mapping model between the brightness temperature data of the microwave radiometer and the passive microwave brightness temperature data:
BTAMSR2-fp(n,dfine)＝RFn-fp[BTMWRI-fp(n,dfine)]
wherein n represents the arbitrary day of the target period, dfineCoordinates (row and column coordinates) representing passive microwave luminance temperature in non-track gap areas, RFn-fp[·]Representing a random forest mapping, BT, between the passive microwave luminance temperature of the non-track-gap region of the target area (outside the track-gap region) and the luminance temperature of the microwave radiometer for day nMWRI-fp(n,dfine) The brightness temperature of a microwave radiometer corresponding to a non-orbit gap area representing passive microwave brightness temperature data, and f and p respectively represent the frequency and polarization mode of an observation channel;
and step 3: constructing an estimation model of the passive microwave brightness temperature in the track gap region:
BTAMSR2-fp(n,dgap)＝RFn-fp[BTMWRI-fp(n,dgap)]
wherein, BTAMSR2-fp(n,dgap) Representing the estimated value of the passive microwave luminance temperature in the track gap region in the target region for the nth day, dgapCoordinates (row and column coordinates) representing passive microwave luminance temperature in the track gap region, BTMWRI-fp(n,dgap) Indicates for the nth day, the micro within the target areaBrightness temperature data of the radiometer;
mapping the random forest with RF according to the estimation precision of the brightness temperature of the passive microwaves by taking the brightness temperature data of the original passive microwaves outside the track clearance area as a verification setn-fp[·]Screening the regression tree;
based on the screening results, according to BTAMSR2-fp(n,dgap)＝RFn-fp[BTMWRI-fp(n,dgap)]Estimating the passive microwave brightness temperature in the track gap area within the specified period of the target area day by day to fill up the missing passive microwave brightness temperature in the track gap area within the specified period of the target area and obtain the seamless passive microwave brightness temperature within the specified period of the target area;
and 4, step 4: constructing a random forest regression relation based on microwave brightness temperature data of passive microwave data of the target area and the earth surface temperature of the medium-resolution imaging spectrometer:
Ts-cl(tcl)＝RFTIR-MW[BTcl(tcl),CFcl(tcl),Sacl(tcl)]
wherein, tclTime series, T, representing clear sky dayss-cl(tcl) Surface temperature, BT, representing a specified distance in clear skycl(tcl) Is a brightness temperature sequence of passive microwaves of a pixel containing the surface temperature of a certain TIR under clear sky conditions, BTi(tcl-j) The luminance temperature of the ith channel at day j, wherein I ═ h1, v1, h2, v2, …, hI, and vI, wherein h1 to hI represent channels with different horizontal polarizations, v1 to vI represent channels with different vertical polarizations, I represents the number of channels, CFcl(tcl) Denotes cloud amount, Sacl(tcl) Representing ground albedo, RFTIR-MW[·]A unique random forest map constructed of pixels representing surface reflectance for a certain TIR;
and 5: on the nth day of the target time period, the corresponding time of the specified period and the nth day form a reconstruction period, and all-weather surface temperature estimation is carried out on the basis of the seamless passive microwave brightness temperature in the reconstruction period to obtain an all-weather surface temperature estimation value in the reconstruction period;
the all-weather surface temperature estimate is:
for any time t of any day within the reconstruction perioddNear real-time all-weather surface temperature initial estimation value T's(td) Comprises the following steps:
wherein, T's-cl(td)、T's-ucl(td) Respectively represent tdThe earth surface temperature of the specified distance is estimated under the clear sky condition and the non-condition at the moment, and
T's-cl(td)＝RFTIR-MW[BTre-cl(td),CFre-cl(td),Sare-cl(td)]
T's-ucl(td)＝RFTIR-MW[BTre-ucl(td),CFre-ucl(td),Sare-ucl(td)]
wherein, BTre-cl(td)、BTre-ucl(td) Respectively represent tdA MW BT time sequence under a clear sky condition and a non-clear sky condition at a moment; CF (compact flash)re-cl(td)、CFre-ucl(td) Respectively represent tdCloud amount corresponding to the image element under clear sky condition and non-clear sky condition; sa (Sa)re-cl(td)、Sare-ucl(td) Respectively represent tdThe surface reflectivity under the clear sky condition and the non-clear sky condition at any moment;
step 6: acquiring the earth surface temperature composition time sequence of the published medium-resolution imaging spectrometer in the reconstruction period, and correcting the all-weather earth surface temperature of the nth day of the target time period (namely correcting the system error) based on the all-weather earth surface temperature estimated value obtained in the step 5:
wherein, T'1(td)、T'2(td) Respectively, t on the n-th day after correctiondFirst and second corrected surface temperatures at the time,
T's-clthe surface temperature sequence (ordered according to time) of the designated distance estimated under the clear sky condition on the nth day of the target time period is represented, namely the all-weather surface temperature sequence of the designated distance estimated under the clear sky condition on the nth day, Ts-clGround surface temperature sequence (ordered by time of day), T 'of a medium resolution imaging spectrometer representing a specified distance on the nth day of a target time period under clear sky conditions'1And T'2First and second corrected surface temperature sequences (ordered by time of day) representing day n, mean (-) representing a mean function; std (. circle.) represents a standard deviation function, Ts-NRT(td) T on the nth day representing a target perioddCorrected surface temperature at that time.
The technical scheme provided by the invention at least has the following beneficial effects:
firstly, missing value filling of a passive microwave data track gap area is realized, time consumption is short, and result precision is high. The method for producing all-weather earth surface temperature by integrating thermal infrared satellite remote sensing data and passive microwave data cannot well solve the problem of obtaining the measured value of the orbit clearance area for a long time, usually uses cubic spline interpolation to simply fill up, and the actual physical meaning of the result is unclear. The result of this method is therefore not "all-weather" in the true sense. Secondly, near real-time acquisition of all-weather surface temperature data at a specified distance (e.g., 1km) is achieved. All-weather surface temperature generation (estimation) methods have been applied only to historical data sets with complete time periods, which is related to the fact that most of their methods are based on the premise that the surface temperature is periodically changing and regular over a complete period of time (e.g., annual/daily scale). The mainstream satellite thermal infrared remote sensing earth surface temperature products (such as a medium-resolution imaging spectrometer) in the world can achieve near real-time acquisition (generally lags behind for 1.5-2 days), and the short board of the all-weather earth surface temperature estimation method on the timeliness greatly limits the further application of the all-weather earth surface temperature estimation method. According to the method, through flexible reconstruction of the annual scale period, a machine learning method is used for establishing a regression model between the brightness temperature value of the passive microwave and the brightness temperature value detected by the medium-resolution imaging spectrometer, and then the regression model is extended to the adjacent latest moment (target point), so that a near-real-time 1km all-weather earth surface temperature product which is synchronous with the brightness temperature value of the passive microwave and the brightness temperature value of the medium-resolution imaging spectrometer is obtained. The method gets rid of the dependence of the traditional integration method on the data of the complete cycle, can meet the data production requirements of different application requirements and monitoring time periods with different lengths, has stable quality of products day by day, has good consistency with the traditional thermal infrared satellite remote sensing earth surface temperature products, has high precision in an acceptable range and high image quality through the inspection of the actually measured earth surface temperature, and can show more space detail information of the earth surface temperature.
Drawings
FIG. 1 is a diagram illustrating a training period and a prediction period in an implementation of the present invention.
FIG. 2 is a region map containing experimental regions in the practice of the present invention.
Fig. 3 is a schematic processing process diagram of the all-weather surface temperature near real-time inversion method for fusion multi-source satellite remote sensing provided by the embodiment of the invention.
Fig. 4 is a comparison graph of the filling effect of missing values in the track gap area in the implementation of the present invention.
Fig. 5 is a plot of the quantitative comparison of BT after padding with the original AMSR2BT implemented in accordance with the present invention.
FIG. 6 is a quantitative comparison of near real-time all-weather LST and MODIS LST obtained by the process of the present invention.
FIG. 7 is a spatial distribution diagram of the all-weather LST and MODIS LST obtained by the processing for the first target time interval in the embodiment of the present invention.
FIG. 8 is a spatial distribution diagram of the all-weather LST and MODIS LST obtained by the processing for the second target time interval in the embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, embodiments of the present invention will be described in detail with reference to the accompanying drawings.
The method for performing near real-time inversion on all-weather earth surface temperature fusing multi-source satellite remote sensing, provided by the embodiment of the invention, comprises the steps of filling a missing value in an AMSR2(Advanced Microwave Scanning Radiometer 2) Microwave brightness temperature data track gap based on Microwave brightness temperature data (BT) provided by a Microwave Radiometer (MWRI) carried by a Chinese wind cloud meteorological satellite, establishing a spatial seamless year scale model of AMSR2BT and TIR LST (such as MODIS), and applying the model to an adjacent year scale time slot to obtain LST at the latest target time.
The all-weather surface temperature near-real-time inversion method fusing multi-source satellite remote sensing provided by the embodiment of the invention can be divided into two parts. In the first part, a random forest regression mapping relation is established between microwave brightness and temperature data of a satellite-mounted microwave radiometer (MWRI) and an advanced microwave scanning radiometer (AMSR2) to obtain near real-time microwave brightness and temperature data (MW BT) of spatially seamless moving microwaves. And secondly, establishing microwave brightness and temperature data (AMSR2BT) of the space seamless advanced microwave scanning radiometer and an annual scale model of TIR LST and applying the model to target time to finally obtain the near-real-time all-weather LST. The microwave radiometer carried by the satellite can be a microwave radiometer carried by a Chinese wind and cloud meteorological satellite.
Although satellite-mounted microwave radiometers and mid-resolution Imaging spectrometers (MODIS) have different scan paths and approximate observation times. However, the inventors of the present invention found, based on statistical analysis, that there was still an observed time difference between MWRI and AMSR2 of 36-120 minutes. Therefore, the microwave light temperature data of MWRI cannot be directly used to fill the track gap region of AMSR 2. However, the inventors of the present invention also found, based on statistical analysis, that there is a high correlation between MWRI BT and AMSR2BT in the experimental region,
microwave radiometers MWRI and Aqua MODIS carried by China Fengyun meteorological satellites have different scanning paths and approximate observation time. However, according to our earlier statistical results, it was found that there is still an observed time difference of 36-120 minutes between MWRI and AMSR 2. Therefore, MWRI BT cannot be directly used to fill the track gap region of AMSR 2. According to our statistical analysis, there is a high correlation between MWRI BT and AMSR2BT for the experimental region, with a correlation R > 0.85.
In the formula (1), i is the sequence number of the effective value outside the AMSR2 track gap region, BTiThe BT value of the ith AMSR 2; MWRIiThe symbol "" represents the mean for the BT value of the ith MWRI. R has a value range of [ -1,1 [)]Wherein R is positive, which means that the two are in positive correlation.
Therefore, the brightness temperature value in the AMSR2 track gap region can be estimated by establishing a regression mapping relationship between MWRI BT and AMSR2 BT. Compared with other machine learning methods, the Random Forest (RF) can more fully characterize the relation between the regression parameters and the target characteristics and construct a regression mapping model with higher accuracy and generalization capability. Therefore, the embodiment of the present invention establishes a regression relationship between the brightness temperature value of each channel of MWRI and AMSR2BT using RF, and then applies within the track gap region of AMSR 2.
On day n of the target session for the experimental region (target region), RF regression mapping models for AMSR2BT and MWRI BT were established outside the track gap region of AMSR 2:
BTAMSR2-fp(n,dfine)＝RFn-fp[BTMWRI-fp(n,dfine)] (2)
in formula (2), f and p are respectively the frequency and polarization mode of the observation channel, BTAMSR2-fp(n,dfine) BT, d representing non-track gap region of AMSR2fineLine and column coordinates representing the effective value of the AMSR2BT non-track-gap area, BTMWRI-fp(n,dfine) Denotes AMSR2BT notMWRI bright temperature value, RF corresponding to track gap regionn-fp[·]Represents the unique random forest map established between BT and corresponding MWRI light temperature data for the AMSR2 non-track gap region within the experimental region for day n.
Then, applying the RF relationship constructed by equation (2) to the track gap region of AMSR2 to implement missing value padding of the track gap region:
BTAMSR2-fp(n,dgap)＝RFn-fp[BTMWRI-fp(n,dgap)] (3)
in the formula (3), BTAMSR2-fp(n,dgap) Estimate, d, representing the track gap region of AMSR2gapLine and row coordinates of the track gap area of BT representing AMSR2MWRI-fp(n,dgap) Represents the MWRI light temperature value corresponding to the BT track gap region of AMSR 2.
The above is the first part, and the spatial seamless AMSR2BT is finally obtained.
And then, establishing annual scale models of spatial seamless AMSR2BT and MODIS LST and applying the models to target time to finally obtain near-real-time all-weather LST.
First, a study period is divided into a training period and a prediction period. The training period is a previous year scale period (shown in fig. 1, a repetition cycle-1) of a date needing near real-time estimation, the prediction period is a target period needing near real-time estimation, and the specific time length is not fixed.
All-weather LST estimation studies based on prior historical data reconstruction showed that the mapping between TIR LST and MW BT was similar over two adjacent year scale periods. Therefore, the mapping relationship obtained from the previous annual scale period (training period) can be applied to the prediction period to obtain near real-time all-weather LST. The mapping relation between the TIR LST and the MW BT can be well described by establishing a linear regression model, and the LST estimated by the method is high in accuracy. The method of the invention takes into account the stability of the mapping relationship and the ability to describe non-linear relationships, replacing linear or non-linear regression methods with RF regression methods. Establishing an RF model between an MODIS LST pixel and a multichannel MW BT at a clear sky specified distance (for example, 1 km):
in the formula (4), tclTime series consisting of clear sky days, Ts-cl(tcl) LST, BT indicating a specified distance in clear skycl(tcl) Is a MW BT time sequence containing a certain TIR LST pixel under clear sky conditioni(tcl-j) Is the luminance temperature of the ith channel on day j (i ═ 10H, 10V.., 89V; where H denotes horizontal polarization and V denotes vertical polarization), CFcl(tcl) Is cloud cover, Sacl(tcl) For ground albedo, RFTIR-MW[·]For a unique random forest map constructed for a certain TIR LST pel, the spatial resolution of MW BT is usually lower than TIR LST, as scaling factor, RFTIR-MW[·]The spatial scaling information from the MW pixel to the TIR LST pixel is the key for realizing the scale reduction, and actually plays a role in reducing the scale factor.
Then, the RF map (RF) constructed by the formula (4) is mappedTIR-MW[·]) Generalizing to adjacent annual time series (the repetition cycle-n interval). Substituting MW BT time sequence in a period as input data into a model RF when a time group reconstruction period (interval-n) between a certain day and the corresponding moment of a previous year scale period in a prediction period is usedTIR-MW[·]To obtain t within the reconstruction perioddThe all-weather surface temperature initial estimation value at the moment is shown as an equation (5):
in formula (II) T's(td) Is tdNear real-time all-weather surface temperature initial estimation value of time, T's-cl(td)、T's-ucl(td) Are each tdL of the specified distance estimated under the clear sky condition and the non-condition of the timeST；BTre-cl(td)、BTre-ucl(td) Are each tdA MW BT time sequence under a clear sky condition and a non-clear sky condition at a moment; CF (compact flash)re-cl(td)、CFre-ucl(td) Are each tdCloud amount corresponding to the image element under the condition of clear sky and the condition of non-clear sky at the moment; sa (Sa)re-cl(td)、Sare-ucl(td) Are each tdAnd (3) the earth surface albedo under the conditions of clear sky and non-clear sky at the moment. It should be noted that the embodiment of the present invention re-estimates the clear sky LST (T's-cl(td) Instead of using MODIS LST in the generated all-weather LST, in order to better embody the performance of the real-time inversion method provided by the embodiments of the present invention.
The MODIS LST is used as a thermal infrared surface temperature product which is widely applied, the real-time inversion method provided by the embodiment of the invention needs to correct the estimated near-real-time all-weather LST to the MODIS LST level in a 'systematic' manner (namely, the system deviation with the MODIS LST is reduced as far as possible), and the corrected near-real-time all-weather LST is shown in a formula (5):
in the formula (6), T's-cl(td) Is tdLST, T of specified distance obtained by estimation under clear sky condition of times-cl(td) For the surface temperature of a medium-resolution imaging spectrometer at a specified distance (for example, 1km) under clear sky conditions, mean () is a mean function; std (-) is a function of standard deviation; t iss-NRT(td) Near real-time all-weather surface temperature required for a target (i.e. FIG. 1, T)d-1The value represented by the point).
For example, the actual implementation area of the embodiment of the present invention selects the black river basin and its peripheral area, and the area map thereof is shown in fig. 2. The black river is the second continental river of China, and the total flow area is 14.3 km2. The area is typical of moderate temperature zone drought area/subarctic zone drought area, has dry climate and little precipitation, and is suitable for the average yearThe precipitation is less than 500 mm. The main types of the ground surfaces in the area are oasis farmland, gobi, desert, wetland and alpine grassland. In addition, the area has a high-quality ground actual measurement station data set provided by a black river basin ecological-hydrological process comprehensive remote sensing observation combined test, and the accuracy verification of the ground surface temperature data can be conveniently carried out.
The data set specifically adopted in this embodiment includes:
1) MODIS products: MODIS daily land surface temperature product (MYD11A1,1km), daily MODIS cloud product (MOD06_ L2,5km), daily land albedo product (MCD43A3, 1km), daily normalized snow index (NDSI) product (MOD10A1, 500m), 16-day synthetic normalized vegetation index (NDVI) (MOD12A2,1 km);
2) passive microwave data set: the resolution ratio of a day-by-day near real-time AMSR2BT three-level product is 0.1 degrees, and a near real-time three-level bright temperature product (FY-3B/3D MWRI BT,0.1 degrees) is provided by a China wind and cloud meteorological satellite microwave radiometer (MWRI);
3) ground measured data set: long-wave radiation data sets of 6 automatic meteorological stations in 2014, which are provided by a black river basin ecological-hydrology process comprehensive remote sensing observation combined test.
Referring to fig. 3, the process of performing near real-time inversion processing on all-weather surface temperature specifically includes:
1. and data preprocessing and space-time matching.
The targeted data objects include near real-time data and long-time data.
Firstly, carrying out batch format conversion, projection transformation, splicing and cutting on MYD11A1 by using an IDL (interface Description language) language and an MRT (multimedia Description Topography) tool to obtain a MODIS LST data set containing a black river basin and peripheral areas thereof; and for MYD11A1 data, removing low-quality MODIS LST pixels by using a Quality Control (QC) layer.
Secondly, format conversion, projection and clipping are carried out on AMSR2BT data in batches by using IDL (inverse discrete cosine transformation) language (pixel elements with observation time difference larger than a specified value (for example, 5min) with MODIS LST), and then longitude and latitude information of AMSR2BT and MODIS LST are respectively read and stored as numerical value matrixes. And performing space matching in the MATLAB according to the read longitude and latitude information to generate a lookup table corresponding to the position relationship. For MWRI BT data, due to the existence of track drift, the MWRI BT and AMSR2BT need to be subjected to space position matching day by day on an MATLAB platform, and pixels with observation time difference exceeding a specified value (for example, 5min) are removed;
still further, 5km of MODIS cloud size product (MOD06_ L2) and 500m of Normalized Daily Snow Index (NDSI) product MOD10A1 were resampled to 1km resolution, respectively, to match the spatial resolution of MODIS LST data. NDVI data at 16 days and 1km resolution is then time interpolated (cubic spline interpolation) to match the day-by-day time resolution of MODIS LST.
And finally, filling data missing parts of NDSI caused by cloud coverage by using a time filter method on MATLAB.
Namely, the purpose of data preprocessing is to realize spatial resolution matching and time resolution between different data sources, wherein the data preprocessing comprises: format conversion, projective transformation, splicing and clipping.
2. The day-by-day AMSR2BT track gap area lacks data padding.
2-1) an RF-based regression relationship was established on MATLAB for MWRI and AMSR2BT on a certain day of the target time period outside the track gap region according to equation (2).
2-2) estimating the AMSR2BT value in the track gap region using formula (3) according to the regression result obtained by formula (2).
2-3) according to the preset quantity of regression trees, taking the original AMSR2BT outside the track gap area as a verification set, estimating the AMSR2BT value by using a formula (3), comparing with the verification set, and selecting the result of the regression tree with the best precision as the final track gap area filling result.
3. And (4) near real-time all-weather surface temperature estimation.
3-1) for the nth day of the target period, establishing a random forest regression relationship (formula (4)) between the AMSR2BT and the MODIS LST with complete history cycles in the previous year or years, in order to obtain enough time law information of local LST cycle changes.
3-2) for the nth day of the target time interval, taking the corresponding time of the previous year scale cycle and forming a reconstruction cycle with the time, and substituting the spatial seamless AMSR2BT in the cycle into a formula (5) to obtain the all-weather surface temperature estimated value in the cycle.
3-3) in the reconstruction period in the step 3-2), acquiring the published MODIS LST to form a time sequence (which is not necessarily complete), combining the initial estimation value obtained by the formula (5) and substituting the initial estimation value into the formula (6), and obtaining the all-weather LST which is consistent with the MODIS LST and is in the target time period for the nth day.
According to the currently adopted data source, the embodiment of the invention can realize two-day-delayed release and meet the requirement of near-real-time monitoring.
Further, the specific results were analyzed as follows:
due to the fact that the ground observation data are difficult to obtain and store, the real-time inversion method provided by the embodiment of the invention uses the long-wave radiation data sets of 6 automatic meteorological stations in 2014, which are provided by the black river basin ecological-hydrological process comprehensive remote sensing observation combined test, so as to verify the near-real-time all-weather LST of 1 km. The evaluation indexes selected by the real-time inversion method provided by the embodiment of the invention comprise mean deviation error (MBE), standard deviation (STD), Root Mean Square Error (RMSE) and determination coefficient (R)2). In addition, the example also generates all-weather LST from 1 month 2020 to 9 months 2020, so as to embody the timeliness of the real-time inversion method provided by the embodiment of the invention.
Fig. 4 shows spatial distribution of AMSR2BT and padded BT in the experimental region, taking 10GHz and 89GHz horizontal channels as examples. The padded BT not only complements the missing values of AMSR2 in the track gap region, but also maintains a high degree of consistency with the original AMSR2BT in terms of spatial distribution and numerical range. In addition, the edge transition of the original track gap area on the filled AMSR2BT image is smooth, and no spatial discontinuity phenomenon occurs. This shows that the gap filling algorithm in the method of the present invention has a high reliability.
Fig. 5 shows a scatter plot between padded BT and AMSR2 BT. MBE is 0.03-0.25K. RMSE of 1.02-1.58K, R20.97-0.99, indicating that the padded BT has high consistency with AMSR2BT, further demonstrating thatGood performance of the gap filling algorithm.
Fig. 6 shows the quantitative comparison results of near real-time all-weather LST and MODIS LST products of 1km a day under clear sky conditions. The near real-time LST has a certain system error compared with MODIS LST, and the MBE of MODIS LST in 2014 and 2020 is 0.12-0.19K in the daytime and-0.07-0.06K at night. STD less than 1.8K, R2Greater than 0.95, indicating close agreement between near real time LST and MODIS LST. This shows that the good performance of the RF regression mapping under clear sky conditions describes the relationship between MW BT and LST.
In addition, the image quality of MODIS LST and near real-time all-weather LST were also compared. Fig. 7 shows the spatial pattern of daytime MODIS LST and near real-time all-weather LST in 2014 in four different seasons (DOY 15, 105, 196 and 288). Near real-time all-weather LSTs are spatially seamless regardless of season and highly consistent with the original MODIS LST in spatial configuration and range of valid values. Meanwhile, the near real-time all-weather LST image has no mosaic or spatial discontinuity, which shows that the image quality of the result of the method is good. Further, FIG. 8 shows the spatial patterns of MODIS LST and near real-time all-weather LST in 9, 30 and 2020 by day. Near real-time all-weather LST is spatially seamless and highly consistent with the original MODIS LST in spatial configuration and range of valid values. The LST missing in the cloud coverage range is well recovered, and no abnormal value occurs. At the same time, there is no mosaic or spatial discontinuity in the near real-time all-weather LST image, which indicates that the RF mapping in equation (4) is indeed effective in overcoming the scale mismatch between the coarse resolution BT and its 1km sub-pel (i.e., MODIS pel) LST.
Tables 1 and 2 show the near real-time all-weather LST and Site (Site) measured LST verification results.
TABLE 1
TABLE 2
In tables 1 and 2, "ARO", "DAM", "HZS", "GB", "SSW", and "HZZ" respectively denote different station names, and a schematic location diagram of each station may be referred to in fig. 2.
By combining the verification results in the daytime and at night, the near-real-time all-weather LST obtained by the real-time inversion method provided by the embodiment of the invention is usually lower than the actually measured LST of the station under the clear sky and non-clear conditions. The main reason for this phenomenon is the cold bias of the MYD11a1 product in arid and semi-arid regions. In the validation results, the MBE was between-1.77K to-0.68K during the night and-2.58-0.17K during the day. The fluctuation of the RMSE and the MBE of other sites except the HZZ site does not exceed 1K, which shows that the real-time inversion method provided by the embodiment of the invention has satisfactory performance and limited uncertainty under all-weather conditions.
In all six sites, GB, SSW, HZS and HZZ all gave worse results than the others, with RMSE between 2.5K and 4.94K. Possible causes are three aspects 1) subsurface heat radiation caused by passive microwave thermal sampling depth; 2) spatial heterogeneity of the area in which the site is located; 3) the inherent error of MODIS LST is mainly the underestimation of temperature values in arid and semi-arid regions.
In summary, it can be shown that the real-time inversion method provided by the embodiment of the invention has good applicability and robustness on different time scales. Near real-time all-weather LST has stable and satisfactory precision under both clear and non-clear air conditions. Although the embodiments of the present invention use AMSR2 and MODIS as data sources, it has the potential to be ported to other sensor platforms. For example, the real-time inversion method provided by the embodiment of the invention can be applied to integrating TIR LST (such as earth temperature products of ESA) and MW BT of geostationary meteorological satellites, and then can generate near real-time all-weather LST with high resolution, which has high value for regional or global meteorological observation (weather forecast, drought and fire monitoring, runoff prediction and the like). Namely, the processing result obtained by the real-time inversion method provided by the embodiment of the invention has good precision, and the RMSE range is 2.46-4.62K. The method is favorable for the production of near-real-time all-weather LST products, and can be used for marking international mainstream thermal infrared remote sensing surface temperature products in timeliness.
Finally, it should be noted that: the above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; 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; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.
What has been described above are merely some embodiments of the present invention. It will be apparent to those skilled in the art that various changes and modifications can be made without departing from the inventive concept thereof, and these changes and modifications can be made without departing from the spirit and scope of the invention.
Claims (3)
1. An all-weather surface temperature near real-time inversion method fusing multi-source satellite remote sensing is characterized by comprising the following steps:
step 1: acquiring brightness temperature data of a target area, brightness temperature data of passive microwaves and earth surface temperature data of a medium-resolution imaging spectrometer;
step 2: constructing a regression mapping model between the brightness temperature data of the microwave radiometer and the passive microwave brightness temperature data:
BTAMSR2-fp(n,dfine)＝RFn-fp[BTMWRI-fp(n,dfine)]
wherein n represents the arbitrary day of the target period, dfineCoordinates representing passive microwave luminance temperature, RF, of non-track gap regionsn-fp[·]Passive representation of non-track gap regions of a target area for day nRandom forest mapping between microwave brightness temperature and brightness temperature of microwave radiometer, BTMWRI-fp(n,dfine) The brightness temperature of a microwave radiometer corresponding to a non-orbit gap area representing passive microwave brightness temperature data, and f and p respectively represent the frequency and polarization mode of an observation channel;
and step 3: constructing an estimation model of the passive microwave brightness temperature in the track gap region:
BTAMSR2-fp(n,dgap)＝RFn-fp[BTMWRI-fp(n,dgap)]
wherein, BTAMSR2-fp(n,dgap) Representing the estimated value of the passive microwave luminance temperature in the track gap region in the target region for the nth day, dgapCoordinate, BT, representing the passive microwave luminance temperature in the track gap regionMWRI-fp(n,dgap) Representing light temperature data for the microwave radiometer within the target area for day n;
mapping the random forest with RF according to the estimation precision of the brightness temperature of the passive microwaves by taking the brightness temperature data of the original passive microwaves outside the track clearance area as a verification setn-fp[·]Screening the regression tree;
based on the screening results, according to BTAMSR2-fp(n,dgap)＝RFn-fp[BTMWRI-fp(n,dgap)]Estimating the passive microwave brightness temperature in the track gap area within the specified period of the target area day by day to fill up the missing passive microwave brightness temperature in the track gap area within the specified period of the target area and obtain the seamless passive microwave brightness temperature within the specified period of the target area;
and 4, step 4: constructing a random forest regression relation based on microwave brightness temperature data of passive microwave data of the target area and the earth surface temperature of the medium-resolution imaging spectrometer:
Ts-cl(tcl)＝RFTIR-MW[BTcl(tcl),CFcl(tcl),Sacl(tcl)]
wherein, tclTime series, T, representing clear sky dayss-cl(tcl) Surface temperature, BT, representing a specified distance in clear skycl(tcl) Is a brightness temperature sequence of passive microwaves of a pixel containing the surface temperature of a certain TIR under clear sky conditions, BTi(tcl-j) The luminance temperature of the ith channel at day j, wherein I ═ h1, v1, h2, v2, …, hI, and vI, wherein h1 to hI represent channels with different horizontal polarizations, v1 to vI represent channels with different vertical polarizations, I represents the number of channels, CFcl(tcl) Denotes cloud amount, Sacl(tcl) Representing ground albedo, RFTIR-MW[·]A unique random forest map constructed of pixels representing surface reflectance for a certain TIR;
and 5: on the nth day of the target time period, the corresponding time of the specified period and the nth day form a reconstruction period, and all-weather surface temperature estimation is carried out on the basis of the seamless passive microwave brightness temperature in the reconstruction period to obtain an all-weather surface temperature estimation value in the reconstruction period;
the all-weather surface temperature estimate is:
for any time t of any day within the reconstruction perioddNear real-time all-weather surface temperature initial estimation value T's(td) Comprises the following steps:
wherein, T's-cl(td)、T's-ucl(td) Respectively represent tdThe earth surface temperature of the specified distance is estimated under the clear sky condition and the non-condition at the moment, and
T's-cl(td)＝RFTIR-MW[BTre-cl(td),CFre-cl(td),Sare-cl(td)]
T's-ucl(td)＝RFTIR-MW[BTre-ucl(td),CFre-ucl(td),Sare-ucl(td)]
wherein, BTre-cl(td)、BTre-ucl(td) Respectively represent tdA MW BT time sequence under a clear sky condition and a non-clear sky condition at a moment; CF (compact flash)re-cl(td)、CFre-ucl(td) Respectively represent tdCloud amount corresponding to the image element under clear sky condition and non-clear sky condition; sa (Sa)re-cl(td)、Sare-ucl(td) Respectively represent tdThe surface reflectivity under the clear sky condition and the non-clear sky condition at any moment;
step 6: acquiring the earth surface temperature composition time sequence of the published medium-resolution imaging spectrometer in the reconstruction period, and correcting the all-weather earth surface temperature of the nth day of the target time period based on the all-weather earth surface temperature estimated value acquired in the step 5:
wherein, T'1(td)、T'2(td) Respectively, t on the n-th day after correctiondFirst and second corrected surface temperatures at the time,
T's-cla surface temperature sequence T of a specified distance estimated under clear sky conditions on the nth day of the target time periods-clGround surface temperature sequence, T ', of a medium resolution imaging spectrometer at a specified distance under clear sky conditions representing the nth day of a target period'1And T'2Representing the first and second corrected surface temperature sequences at day n, mean (-) representing the mean function; std (. circle.) represents a standard deviation function, Ts-NRT(td) T on the nth day representing a target perioddCorrected surface temperature at that time.
2. The method of claim 1, wherein the specified distance is 1 km.
3. The method of claim 1, wherein the specified period is a previous year scale period.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110451765.7A CN113158570A (en) | 2021-04-26 | 2021-04-26 | All-weather surface temperature near-real-time inversion method fusing multi-source satellite remote sensing |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110451765.7A CN113158570A (en) | 2021-04-26 | 2021-04-26 | All-weather surface temperature near-real-time inversion method fusing multi-source satellite remote sensing |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113158570A true CN113158570A (en) | 2021-07-23 |
Family
ID=76870784
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110451765.7A Pending CN113158570A (en) | 2021-04-26 | 2021-04-26 | All-weather surface temperature near-real-time inversion method fusing multi-source satellite remote sensing |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113158570A (en) |
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113743000A (en) * | 2021-08-13 | 2021-12-03 | 电子科技大学 | Method for generating all-weather surface temperature with high time resolution |
| CN113834572A (en) * | 2021-08-26 | 2021-12-24 | 电子科技大学 | Method for removing drift of temperature measurement result of non-refrigeration thermal imager of unmanned aerial vehicle |
| CN113866764A (en) * | 2021-08-13 | 2021-12-31 | 成都理工大学 | Landslide susceptibility improvement evaluation method based on InSAR and LR-IOE model |
| CN113901966A (en) * | 2021-12-07 | 2022-01-07 | 武汉光谷信息技术股份有限公司 | Crop classification method fusing multi-source geographic information data |
| CN113987782A (en) * | 2021-10-26 | 2022-01-28 | 武汉大学 | Near-ground ultraviolet irradiance satellite missing data completion method based on cloud information |
| CN114019579A (en) * | 2021-08-24 | 2022-02-08 | 中国农业科学院农业资源与农业区划研究所 | High-space-time resolution near-surface air temperature reconstruction method, system and equipment |
| CN114184280A (en) * | 2021-12-07 | 2022-03-15 | 自然资源部国土卫星遥感应用中心 | Earth surface temperature time normalization method based on heat balance |
| CN114202705A (en) * | 2022-02-16 | 2022-03-18 | 清华大学 | Spectral feature time sequence construction method and system |
| CN114436086A (en) * | 2022-01-24 | 2022-05-06 | 浙江新再灵科技股份有限公司 | Dim identification and alarm method and system for elevator car |
| CN116358709A (en) * | 2023-06-02 | 2023-06-30 | 中国科学院空天信息创新研究院 | Surface temperature inversion method based on passive microwave multiband temperature model |
| CN116450700A (en) * | 2023-03-09 | 2023-07-18 | 中国农业科学院农业资源与农业区划研究所 | Polar orbit satellite earth surface temperature time normalization method and device and electronic equipment |
| CN117725375A (en) * | 2024-02-08 | 2024-03-19 | 中国科学院空天信息创新研究院 | Satellite remote sensing data reconstruction method and device, electronic equipment and storage medium |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090022359A1 (en) * | 2007-07-19 | 2009-01-22 | University Of Delaware | Vegetation index image generation methods and systems |
| CN101799561A (en) * | 2010-02-05 | 2010-08-11 | 民政部国家减灾中心 | Snow disaster remote sensing monitoring simulation evaluation method based on disaster reduction small satellite |
| CN105321157A (en) * | 2015-12-09 | 2016-02-10 | 中国科学院遥感与数字地球研究所 | Passive microwave brightness temperature data downscaling method and passive microwave brightness temperature data downscaling system |
| CN107576417A (en) * | 2017-09-04 | 2018-01-12 | 电子科技大学 | A kind of round-the-clock surface temperature generation method |
| CN108898049A (en) * | 2018-05-11 | 2018-11-27 | 中南林业科技大学 | Forest fires recognition methods based on MODIS data |
| US20190057171A1 (en) * | 2017-02-10 | 2019-02-21 | Guangxi Institute of Meteorlogical Disaster- Reducing Research | Novel nonlinear method for area-wide near surface air temperature precision retrieval |
| CN110516816A (en) * | 2019-08-30 | 2019-11-29 | 中国科学院、水利部成都山地灾害与环境研究所 | Round-the-clock surface temperature generation method and device based on machine learning |
| CN111982294A (en) * | 2020-07-21 | 2020-11-24 | 电子科技大学 | All-weather earth surface temperature generation method integrating thermal infrared and reanalysis data |
| CN112016052A (en) * | 2020-08-20 | 2020-12-01 | 广东省气象探测数据中心 | Near-surface daily maximum air temperature estimation method, system and terminal based on multi-source data |
| CN112464980A (en) * | 2020-10-26 | 2021-03-09 | 中国农业科学院农业资源与农业区划研究所 | Method for inverting earth surface temperature by fusing thermal infrared and passive microwave remote sensing data |
-
2021
- 2021-04-26 CN CN202110451765.7A patent/CN113158570A/en active Pending
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090022359A1 (en) * | 2007-07-19 | 2009-01-22 | University Of Delaware | Vegetation index image generation methods and systems |
| CN101799561A (en) * | 2010-02-05 | 2010-08-11 | 民政部国家减灾中心 | Snow disaster remote sensing monitoring simulation evaluation method based on disaster reduction small satellite |
| CN105321157A (en) * | 2015-12-09 | 2016-02-10 | 中国科学院遥感与数字地球研究所 | Passive microwave brightness temperature data downscaling method and passive microwave brightness temperature data downscaling system |
| US20190057171A1 (en) * | 2017-02-10 | 2019-02-21 | Guangxi Institute of Meteorlogical Disaster- Reducing Research | Novel nonlinear method for area-wide near surface air temperature precision retrieval |
| CN107576417A (en) * | 2017-09-04 | 2018-01-12 | 电子科技大学 | A kind of round-the-clock surface temperature generation method |
| CN108898049A (en) * | 2018-05-11 | 2018-11-27 | 中南林业科技大学 | Forest fires recognition methods based on MODIS data |
| CN110516816A (en) * | 2019-08-30 | 2019-11-29 | 中国科学院、水利部成都山地灾害与环境研究所 | Round-the-clock surface temperature generation method and device based on machine learning |
| CN111982294A (en) * | 2020-07-21 | 2020-11-24 | 电子科技大学 | All-weather earth surface temperature generation method integrating thermal infrared and reanalysis data |
| CN112016052A (en) * | 2020-08-20 | 2020-12-01 | 广东省气象探测数据中心 | Near-surface daily maximum air temperature estimation method, system and terminal based on multi-source data |
| CN112464980A (en) * | 2020-10-26 | 2021-03-09 | 中国农业科学院农业资源与农业区划研究所 | Method for inverting earth surface temperature by fusing thermal infrared and passive microwave remote sensing data |
Non-Patent Citations (4)
| Title |
|---|
| TANG WENBIN 等: "Near-Real-Time Estimation of 1-km All-Weather Land Surface Temperature by Integrating Satellite Passive Microwave and Thermal Infrared Observations", 《IEEE GEOSCIENCE AND REMOTE SENSING LETTERS》 * |
| 吕京国: "MODIS地表产品数据的相关算法及处理过程", 《遥感应用》 * |
| 张晓东: "多源遥感协同下的全天候地表温度估算研究", 《中国博士学位论文全文数据库 (基础科学辑)》 * |
| 王浩 等: "基于多源遥感数据的锡尔河中下游农田土壤水分反演", 《自然资源学报》 * |
Cited By (23)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113743000A (en) * | 2021-08-13 | 2021-12-03 | 电子科技大学 | Method for generating all-weather surface temperature with high time resolution |
| CN113866764A (en) * | 2021-08-13 | 2021-12-31 | 成都理工大学 | Landslide susceptibility improvement evaluation method based on InSAR and LR-IOE 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 |
| CN114019579B (en) * | 2021-08-24 | 2023-11-24 | 中国农业科学院农业资源与农业区划研究所 | High space-time resolution near-surface air temperature reconstruction method, system and equipment |
| CN114019579A (en) * | 2021-08-24 | 2022-02-08 | 中国农业科学院农业资源与农业区划研究所 | High-space-time resolution near-surface air temperature reconstruction method, system and equipment |
| CN113834572A (en) * | 2021-08-26 | 2021-12-24 | 电子科技大学 | Method for removing drift of temperature measurement result of non-refrigeration thermal imager of unmanned aerial vehicle |
| CN113834572B (en) * | 2021-08-26 | 2023-05-12 | 电子科技大学 | Unmanned aerial vehicle non-refrigeration type thermal imager temperature measurement result drift removal method |
| CN113987782A (en) * | 2021-10-26 | 2022-01-28 | 武汉大学 | Near-ground ultraviolet irradiance satellite missing data completion method based on cloud information |
| CN113987782B (en) * | 2021-10-26 | 2024-07-12 | 武汉大学 | Near-ground ultraviolet irradiance satellite missing data complement method based on cloud information |
| CN113901966B (en) * | 2021-12-07 | 2022-03-11 | 武汉光谷信息技术股份有限公司 | Crop classification method fusing multi-source geographic information data |
| CN114184280A (en) * | 2021-12-07 | 2022-03-15 | 自然资源部国土卫星遥感应用中心 | Earth surface temperature time normalization method based on heat balance |
| CN114184280B (en) * | 2021-12-07 | 2024-03-19 | 自然资源部国土卫星遥感应用中心 | Surface temperature time normalization method based on heat balance |
| CN113901966A (en) * | 2021-12-07 | 2022-01-07 | 武汉光谷信息技术股份有限公司 | Crop classification method fusing multi-source geographic information data |
| CN114436086A (en) * | 2022-01-24 | 2022-05-06 | 浙江新再灵科技股份有限公司 | Dim identification and alarm method and system for elevator car |
| CN114436086B (en) * | 2022-01-24 | 2023-09-19 | 浙江新再灵科技股份有限公司 | Method and system for identifying and alarming darkness of elevator car |
| CN114202705B (en) * | 2022-02-16 | 2022-05-20 | 清华大学 | Spectral feature time sequence construction method and system |
| CN114202705A (en) * | 2022-02-16 | 2022-03-18 | 清华大学 | Spectral feature time sequence construction method and system |
| CN116450700A (en) * | 2023-03-09 | 2023-07-18 | 中国农业科学院农业资源与农业区划研究所 | Polar orbit satellite earth surface temperature time normalization method and device and electronic equipment |
| CN116450700B (en) * | 2023-03-09 | 2024-02-13 | 中国农业科学院农业资源与农业区划研究所 | Polar orbit satellite earth surface temperature time normalization method and device and electronic equipment |
| CN116358709B (en) * | 2023-06-02 | 2023-08-29 | 中国科学院空天信息创新研究院 | Surface temperature inversion method based on passive microwave multiband temperature model |
| CN116358709A (en) * | 2023-06-02 | 2023-06-30 | 中国科学院空天信息创新研究院 | Surface temperature inversion method based on passive microwave multiband temperature model |
| CN117725375A (en) * | 2024-02-08 | 2024-03-19 | 中国科学院空天信息创新研究院 | Satellite remote sensing data reconstruction method and device, electronic equipment and storage medium |
| CN117725375B (en) * | 2024-02-08 | 2024-05-14 | 中国科学院空天信息创新研究院 | Satellite remote sensing data reconstruction method and device, electronic equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113158570A (en) | All-weather surface temperature near-real-time inversion method fusing multi-source satellite remote sensing | |
| CN106776481B (en) | Downscaling correction method for satellite precipitation data | |
| Hughes | Global cloud climatologies: A historical review | |
| Muster et al. | Spatio-temporal sensitivity of MODIS land surface temperature anomalies indicates high potential for large-scale land cover change detection in Arctic permafrost landscapes | |
| CN103413272A (en) | Low-spatial-resolution multisource remote sensing image space consistency correction method | |
| CN114019579B (en) | High space-time resolution near-surface air temperature reconstruction method, system and equipment | |
| Fang et al. | A dataset of daily near-surface air temperature in China from 1979 to 2018 | |
| CN111982294B (en) | All-weather earth surface temperature generation method integrating thermal infrared and reanalysis data | |
| Parkinson | Spatial patterns in the length of the sea ice season in the Southern Ocean, 1979–1986 | |
| Song et al. | An improved surface soil moisture downscaling approach over cloudy areas based on geographically weighted regression | |
| Ma et al. | Generating gapless land surface temperature with a high spatio-temporal resolution by fusing multi-source satellite-observed and model-simulated data | |
| CN116011342B (en) | All-weather reconstruction method for high-resolution thermal infrared surface temperature | |
| Hong et al. | A simple yet robust framework to estimate accurate daily mean land surface temperature from thermal observations of tandem polar orbiters | |
| CN114564767A (en) | Under-cloud surface temperature estimation method based on sun-cloud-satellite observation geometry | |
| Shamim et al. | An improved technique for global daily sunshine duration estimation using satellite imagery | |
| Ioannidis et al. | Assessment of the global digital elevation models ASTER and SRTM in Greece | |
| Cao et al. | A new global gridded sea surface temperature data product based on multisource data | |
| Frey et al. | Land-surface temperature dynamics in the Upper Mekong Basin derived from MODIS time series | |
| Hu et al. | Spatiotemporal downscaling method of land surface temperature based on daily change model of temperature | |
| Duan et al. | A Novel Approach to All-Weather LST Estimation using XGBoost Model and Multi-Source Data | |
| CN104360040B (en) | Remote sensing soil moisture content monitoring method based on STARFM fusion technology | |
| AU2021105536A4 (en) | A High Spatial-Temporal Resolution Method for Near-Surface Air Temperature Reconstruction | |
| Cui et al. | Extraction of vegetation fraction based on the dimidiate pixel model and vegetation index transform plan | |
| Meng et al. | Estimating Hourly Land Surface Temperature from FY-4A AGRI Using an Explicitly Emissivity-Dependent Split-Window Algorithm | |
| CN111914396A (en) | High-resolution DEM data-based quick parameterization method for sub-grid terrain three-dimensional surface solar radiation forced effect |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210723 |
|
| RJ01 | Rejection of invention patent application after publication |