Winter wheat phenological period monitoring method for heterogeneous regions

Abstract

The invention discloses a winter wheat phenological period monitoring method for heterogeneous regions, and relates to the technical field of remote sensing monitoring. According to the method, wave band reflection information is extracted by smoothing remote sensing image data of a phenological period monitoring area. EVI2 is calculated through wave band reflection information, and an EVI2 time sequence curve is obtained. And constructing an SMF model to perform curve fitting on the EVI2 time sequence curve, optimizing a model phenological period matching algorithm in combination with a spatial change rule of a phenological period monitoring area, and constructing a GSMF model. The phenological period difference is calculated according to the growth period difference caused by geographic factors of the phenological monitoring area, and the phenological period of the winter wheat in the heterogeneous area is monitored. According to the method, remote sensing information and geographic information of a phenological period monitoring area are combined to be used for monitoring the phenological period of the winter wheat, the method has high spatial expansibility and universality, and accurate prediction of the phenological period of the winter wheat in a heterogeneous area can be achieved.

Description

Technical Field

[0001] The invention relates to the technical field of remote sensing monitoring, and mainly to a winter wheat phenological period monitoring method for heterogeneous areas. Background Art

[0002] Crop phenology refers to the changes in external morphological characteristics during the growth and development of crops, which is crucial for understanding the dynamic growth of crops, the formation of yield and quality, and field management. Traditional phenological monitoring relies on field surveys, which have high data acquisition costs and difficulties, and cannot meet the spatial needs of phenological information for crop growth monitoring and yield estimation research in large areas. The methods suitable for large-scale phenological monitoring are mainly divided into two categories: phenological simulation of crop growth models based on meteorological data and phenological monitoring based on remote sensing data characteristics. The phenological simulation model of the crop growth model based on meteorological data depends on crop variety information, meteorological information, and management information. It is difficult to obtain input data in spatially heterogeneous areas, and the task of determining parameters in the phenological monitoring process is arduous. Phenological feature extraction based on remote sensing data can use the vegetation index VI time series for phenological remote sensing observation, which has shown great potential in timeliness and spatial heterogeneity.

[0003] The threshold method, change monitoring method and shape model fitting method are the three main methods for remote sensing monitoring of crop phenological periods. The threshold method defines a specific value of the VI time series as a specific phenological period, such as a fixed threshold or a dynamic threshold such as the mean, median, or rate in a certain period of time; the change monitoring method determines a specific phenological period by monitoring the maximum and minimum values ​​of the VI time series curve, i.e., the inflection point and other characteristic points. The phenological periods extracted by the threshold method and the change monitoring method are mainly mathematically defined, and the defined phenological periods are limited, mainly the seedling period, heading period, and harvest period. Moreover, the phenological periods predicted by this type of method are significantly different from the actual phenological periods on the ground. The shape model fitting method SMF can solve the problem of limited definition of phenological periods by linearly scaling and matching the phenological periods of the entire growing season. However, the SMF model assumes that the position of the date of a specific crop phenological period on the VI time series curve is relatively stable, and assumes that all growth stages change synchronously. This does not conform to the growth pattern of winter wheat before and after wintering, and does not consider the phenological period differences caused by crop growth differences in heterogeneous regions. In addition, using the mean of the VI time series of a certain area and its corresponding phenological period as a standard to determine the phenological period dates of different areas has advantages in a smaller range, but the large difference in crop growth in heterogeneous space will lead to large errors. Heterogeneous space, that is, heterogeneous areas, refers to local areas in remote sensing image data of winter wheat that are different in color from the surrounding areas. Summary of the invention

[0004] The present invention aims to solve the problems that the existing winter wheat phenological period monitoring methods have difficulty in monitoring crop growth in large areas, large monitoring errors in winter wheat in heterogeneous areas, and a limited number of monitored phenological periods, and thus provides a winter wheat phenological period monitoring method for heterogeneous areas.

[0005] A winter wheat phenological period monitoring method for heterogeneous regions, comprising:

[0006] S1. Obtain remote sensing image data of the winter wheat phenological period monitoring area through MODIS satellite, and smooth the remote sensing image data to obtain smooth remote sensing data;

[0007] S2. Extract the band reflectance information from the smoothed remote sensing data, and calculate the dual-band enhanced vegetation index EVI2 of the winter wheat phenological period monitoring area through the band reflectance information;

[0008] S3. extracting the time series curve of the dual-band enhanced vegetation index EVI2 to obtain the EVI2 time series curve;

[0009] S4. Construct a reference shape fitting model SMF, and obtain an EVI2 time series fitting curve based on the EVI2 time series curve;

[0010] S5. Statistically analyze the measured data of the winter wheat phenological period monitoring area, and combine it with the spatial data of the winter wheat phenological period monitoring area to obtain the spatial variation law of the phenological period monitoring area;

[0011] S6. Construct a geographically weighted shape fitting model GSMF and optimize the SMF phenological period matching algorithm based on the spatial variation law of the phenological period monitoring area;

[0012] S7. The S algorithm is optimized by golden section search and the bandwidth parameter is selected using the Akaike information criterion (AIC) to extract the phenological period of winter wheat in heterogeneous regions.

[0013] In S2, the calculation formula of the dual-band enhanced vegetation index EVI2 is as follows:

[0014]

[0015] In the formula, R NIR is the surface reflectivity in the near-infrared band, R RED is the surface reflectivity in the red light band.

[0016] In S3, the EVI2 time series curve h(x) is as follows:

[0017]

[0018] Where i and j are the amplitude factors of the EVI2 time series curve, which determine the peak height of the EVI2 time series curve; x is the number of days of phenological period monitoring; f and h are the center positions of the peaks of the EVI2 time series curve; a and b are the standard deviations, which determine the distance between the valleys of the EVI2 time series curve.

[0019] In S4, a reference shape fitting model SMF is constructed, and the EVI2 time series curve is averaged to obtain the reference EVI2 time series curve Q(x). Two curve segments are selected from Q(x), and the two curve segments are translated and scaled to obtain the EVI2 time series fitting curve g(x). The calculation process of g(x) is:

[0020] g(x)=yscale1×Q1(xscale1×x+tshift1)+yscale2×Q2(xscale2×x+tshift2);

[0021] Where Q1(x) and Q2(x) are two curves selected from Q(x), yscale1 and yscale2 are the Y-coordinate scaling factors corresponding to the two curves, xscale1 and xscale2 are the X-coordinate scaling factors corresponding to the two curves, and tshift1 and tshift2 are the translation steps corresponding to the two curves.

[0022] The Y-coordinate scaling factor, X-coordinate scaling factor and translation step length are all scale conversion parameters, which are selected by evaluating the root mean square error RMSE. The calculation formula is:

[0023]

[0024] Where f(x) is the predicted amount of winter wheat in the monitoring area, and m(x) is the actual amount of winter wheat in the monitoring area.

[0025] In S6, the GSMF model uses the model structure and target curve conversion method of SMF, and modifies the phenological period matching algorithm based on the SMF model. The phenological period of winter wheat in the monitoring area is estimated by combining the phenological period differences caused by crop growth differences. The modified model phenological period matching algorithm X est The calculation process is as follows:

[0026] X est =(a 0(ui,vi) +a 1(ui,vi) xscale (ui,vi) )×(X0+a 2(ui,vi) tshift (ui,vi) +ε (ui,vi) );

[0027] Where i = 1, 2, 3, ..., m; m represents the number of observations; a0(ui,vi) 、a 1(ui,vi) and a 2(ui,vi) Represents the coefficient vector of the phenological period estimation model at the position (ui,vi); xscale (ui,vi) Indicates the X-coordinate scaling factor at the (ui,vi) position; tshift (ui,vi) represents the translation step length at the position (ui,vi); ε (ui,vi) represents the estimation error of the phenological period at the position (ui,vi); X0 is the reference phenological period of SMF; X est The value represents the estimated phenological period of the target pixel.

[0028] Take the position of the pixel as the centroid, calculate the distance between pixels, and get the coefficient vector of the local pixel as follows:

[0029]

[0030] Where X is the reference phenological period of the GSMF model; X′ is the estimated phenological period of the GSMF model; W (ui,vi) represents an n×n-order diagonal weight matrix, which weights the corresponding observations based on the distance between each observation and the location (ui,vi) in the phenological observation area; [] -1 is the inverse of the matrix; y is the observed value of the phenological period.

[0031] In S7, the calculation formula of Akaike information criterion AIC is:

[0032]

[0033] Where AIC is the Akaike information entropy of the Akaike information criterion, m is the number of observations, RSS is the residual sum of squares, S is the phenological observation parameter matrix, and tr(s) is the trace number of matrix S.

[0034] Compared with the prior art, the present invention has the following beneficial effects:

[0035] On the basis of fully considering the feasibility and practicality of the technical solution, the present invention conducts in-depth research on the phenological period information of winter wheat in heterogeneous regions and the difficulty of obtaining the required data, and proposes a winter wheat phenological period monitoring method for heterogeneous regions. The remote sensing information of the phenological period monitoring area is combined with the geographic information for winter wheat phenological period monitoring, which has strong spatial scalability and universality, and can realize the accurate prediction of the key phenological period of winter wheat in a large regional scale. By modifying the phenological period parameter acquisition method in the classic shape model fitting model SMF, a GSMF model is constructed to weight the corresponding observations in terms of location, and the phenological period difference is calculated for the growth period difference caused by geographical factors to monitor the phenological period, so as to realize the extraction of the greening period, jointing period, booting period, heading period, flowering period and harvest period in different regions based on the band reflection information in the remote sensing data. Compared with other remote sensing data, the band reflection information is less difficult to obtain, and can be used to extract the phenological period information of winter wheat in different regions and different growing seasons, and has high accuracy, strong universality and stability. At the same time, this idea can be applied to phenological period monitoring of different remote sensing platforms and different crops (such as corn, rice, etc.). In practical applications, the spatiotemporal resolution and collection method of the required remote sensing data can be determined according to specific needs such as the area of ​​the crop monitoring area. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] The accompanying drawings in the specification, which constitute a part of the present invention, are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute improper limitations on the present invention.

[0037] Figure 1 A flow chart of phenological period monitoring based on a geographically weighted shape fitting model provided in an embodiment of the present invention;

[0038] Figure 2 A 1:1 scatter plot of the estimated phenological period and the measured phenological period based on the SMF model provided in an embodiment of the present invention;

[0039] Figure 3 A 1:1 scatter plot of the estimated phenological period and the measured phenological period based on the GSMF model provided in an embodiment of the present invention;

[0040] Figure 4 A 1:1 scatter plot of the estimated phenological period and the measured phenological period in 2014 based on the GSMF model provided in an embodiment of the present invention;

[0041] Figure 5 A 1:1 scatter plot of the estimated phenological period and the measured phenological period in 2015 based on the GSMF model provided in an embodiment of the present invention;

[0042] Figure 6A 1:1 scatter plot of the estimated phenological period and the measured phenological period in 2016 based on the GSMF model provided in an embodiment of the present invention;

[0043] Figure 7 A 1:1 scatter plot of the estimated phenological period and the measured phenological period in 2017 based on the GSMF model provided in an embodiment of the present invention;

[0044] Figure 8 A 1:1 scatter plot of the estimated phenological period and the measured phenological period in 2018 based on the GSMF model provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0045] The present invention discloses a method for monitoring winter wheat phenological period in heterogeneous regions. Those skilled in the art can refer to the content of this article and appropriately improve the process parameters to achieve it. It should be particularly pointed out that all similar substitutions and modifications are obvious to those skilled in the art, and they are all considered to be included in the present invention, and relevant personnel can obviously modify or appropriately change and combine the contents described herein without departing from the content, spirit and scope of the present invention to implement and apply the technology of the present invention.

[0046] In the present invention, unless otherwise specified, the scientific and technical terms used herein have the meanings commonly understood by those skilled in the art. In order to enable those skilled in the art to better understand the technical solution of the present invention, the present invention is further described in detail below in conjunction with specific embodiments.

[0047] The GSMF model uses the model structure and target curve conversion method of SMF, and modifies the phenological period matching algorithm based on the SMF model. It weights the corresponding observation values ​​according to the geographic spatial information of longitude, latitude, altitude and slope of the phenological monitoring area, calculates the crop growth differences caused by regional differences, and realizes the phenological period estimation of winter wheat in heterogeneous regions.

[0048] A winter wheat phenological period monitoring method for heterogeneous regions, comprising:

[0049] S1. Obtain remote sensing image data of the winter wheat phenological period monitoring area through MODIS satellite, and smooth the remote sensing image data to obtain smooth remote sensing data;

[0050] S2. Extract the band reflectance information from the smoothed remote sensing data, and calculate the dual-band enhanced vegetation index EVI2 of the winter wheat phenological period monitoring area through the band reflectance information;

[0051] S3. extracting the time series curve of the dual-band enhanced vegetation index EVI2 to obtain the EVI2 time series curve;

[0052] S4. Construct a reference shape fitting model SMF, and obtain an EVI2 time series fitting curve based on the EVI2 time series curve;

[0053] S5. Statistically analyze the measured data of the winter wheat phenological period monitoring area, and combine it with the spatial data of the winter wheat phenological period monitoring area to obtain the spatial variation law of the phenological period monitoring area;

[0054] S6. Construct a geographically weighted shape fitting model GSMF and optimize the SMF phenological period matching algorithm based on the spatial variation law of the phenological period monitoring area;

[0055] S7. The S algorithm is optimized by golden section search and the bandwidth parameter is selected using the Akaike information criterion (AIC) to extract the phenological period of winter wheat in heterogeneous regions.

[0056] In S2, the calculation formula of the dual-band enhanced vegetation index EVI2 is as follows:

[0057]

[0058] In the formula, R NIR is the surface reflectivity in the near-infrared band, R RED is the surface reflectivity in the red light band.

[0059] In S3, the EVI2 time series curve h(x) is as follows:

[0060]

[0061] Where i and j are the amplitude factors of the EVI2 time series curve, which determine the peak height of the EVI2 time series curve; x is the number of days of phenological period monitoring; f and h are the center positions of the peaks of the EVI2 time series curve; a and b are the standard deviations, which determine the distance between the valleys of the EVI2 time series curve.

[0062] In S4, a reference shape fitting model SMF is constructed, and the EVI2 time series curve is averaged to obtain the reference EVI2 time series curve Q(x). Two curve segments are selected from Q(x), and the two curve segments are translated and scaled to obtain the EVI2 time series fitting curve g(x). The calculation process of g(x) is:

[0063] g(x)=yscale1×Q1(xscale1×x+tshift1)+yscale2×Q2(xscale2×x+tshift2);

[0064] Where Q1(x) and Q2(x) are two curves selected from Q(x), yscale1 and yscale2 are the Y-coordinate scaling factors corresponding to the two curves, xscale1 and xscale2 are the X-coordinate scaling factors corresponding to the two curves, and tshift1 and tshift2 are the translation steps corresponding to the two curves.

[0065] The Y-coordinate scaling factor, X-coordinate scaling factor and translation step length are all scale conversion parameters, which are selected by evaluating the root mean square error RMSE. The calculation formula is:

[0066]

[0067] Where f(x) is the predicted amount of winter wheat in the monitoring area, and m(x) is the actual amount of winter wheat in the monitoring area.

[0068] In S6, the GSMF model uses the model structure and target curve conversion method of SMF, and modifies the phenological period matching algorithm based on the SMF model. The phenological period of winter wheat in the monitoring area is estimated by combining the phenological period differences caused by crop growth differences. The modified model phenological period matching algorithm X est The calculation process is as follows:

[0069] X est =(a 0(ui,vi) +a 1(ui,vi) xscale (ui,vi) )×(X0+a 2(ui,vi) tshift (ui,vi) +ε (ui,vi) );

[0070] Where i = 1, 2, 3, ..., m; m represents the number of observations; a 0(ui,vi) 、a 1(ui,vi) and a 2(ui,vi) Represents the coefficient vector of the phenological period estimation model at the position (ui,vi); xscale (ui,vi) Indicates the X-coordinate scaling factor at the (ui,vi) position; tshift (ui,vi) represents the translation step length at the position (ui,vi); ε (ui,vi) represents the estimation error of the phenological period at the position (ui,vi); X0 is the reference phenological period of SMF; X est The value represents the estimated phenological period of the target pixel.

[0071] Take the position of the pixel as the centroid, calculate the distance between pixels, and get the coefficient vector of the local pixel as follows:

[0072]

[0073] Where X is the reference phenological period of the GSMF model; X′ is the estimated phenological period of the GSMF model; W (ui,vi) represents an n×n-order diagonal weight matrix, which weights the corresponding observations based on the distance between each observation and the location (ui,vi) in the phenological observation area; [] -1 is the inverse of the matrix; y is the observed value of the phenological period.

[0074] In S7, the calculation formula of Akaike information criterion AIC is:

[0075]

[0076] In the formula, the AIC value is the Akaike information entropy of the Akaike information criterion, m represents the number of observations, RSS represents the residual sum of squares, S represents the phenological period observation parameter matrix, and tr(s) is the trace number of the matrix S. The Akaike information criterion AIC can be used to select the minimum model bandwidth and reduce the complexity of model construction.

[0077] The embodiment of the present invention comprehensively considers factors such as the spatial differences in phenological periods and the difficulty of obtaining the required data. Based on the field phenological period survey data, the phenological period parameter acquisition method in the classic shape model fitting model SMF is modified to realize the extraction of the greening period, jointing period, booting period, heading period, flowering period and harvesting period in different regions based on the band reflectance information in the remote sensing data. The model is applied and verified in the Huanghuaihai region. The RMSE comparison results of the SMF model and the GSMF model in different phenological periods are shown in Table 1.

[0078] Table 1 RMSE table of SMF and GSMF models in different phenological periods;

[0079]

[0080] The smaller the root mean square error RMSE, the smaller the gap between the estimated phenological period and the measured phenological period, and the better the model's estimation effect on the phenological period of winter wheat. Table 1 shows that in each phenological period of winter wheat, the root mean square error RMSE of the GSMF model is smaller than that of the SMF model for each phenological period of winter wheat. Therefore, the GSMF model has a better estimation effect on the phenological period of winter wheat than the SMF model.

[0081] Figure 1This is a flowchart of phenological period monitoring based on the geographically weighted shape fitting model. The remote sensing image data of the phenological period monitoring area is obtained through the MODIS satellite, and the remote sensing image data is smoothed. The crop EVI2 time series curve is calculated based on the band reflectance information in the remote sensing image data. The spatial variation law of the phenological period is obtained based on the spatial data of longitude, altitude, and slope of the phenological period monitoring area combined with the measured data of the phenological period. The geographically weighted shape fitting model is constructed through shape fitting model conversion, estimation algorithm modification, bandwidth parameter selection and other phenological period estimation optimization to obtain the phenological period estimation results. The estimation results are verified through independent samples to realize phenological period monitoring based on the geographically weighted shape fitting model.

[0082] Figure 2 The 1:1 scatter plot of the estimated results and measured results of winter wheat phenological period in Huanghuaihai region based on SMF model is shown in Figure 1. Figure 2 It can be seen that the estimated phenological period based on the SMF model is generally later than the measured phenological period, and the root mean square error of the estimation result is large, so it is impossible to make a more accurate estimate of the filling period of winter wheat.

[0083] Figure 3 The 1:1 scatter plot of the estimated results and measured results of winter wheat phenological period in Huanghuaihai region based on GSMF model is shown in Figure 1. Figure 3 It can be seen that the estimated winter wheat phenological period based on the GSMF model has a high degree of overlap with the measured phenological period. Figure 2 and Figure 3 Combined with Table 1, it can be seen that the GSMF model has better estimation effect than the SMF model, and has higher precision and accuracy in monitoring the phenological period of winter wheat.

[0084] pass Figures 4 to 8 It can be seen that when the winter wheat phenological period is monitored based on the GSMF model, the monitoring effect of the phenological period in different years is relatively stable, and the model's estimation results for each phenological period of winter wheat are basically consistent with the measured results of the phenological period. The error between the phenological period estimation results and the actual survey results is small, that is, the GSMF model can accurately estimate multiple phenological periods of wheat in different regions.

[0085] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A winter wheat phenological period monitoring method for heterogeneous regions. It is characterized in that include: S1. Obtain remote sensing image data of the winter wheat phenological period monitoring area through MODIS satellite, and smooth the remote sensing image data to obtain smooth remote sensing data; S2. Extract the band reflectance information from the smoothed remote sensing data, and calculate the dual-band enhanced vegetation index EVI2 of the winter wheat phenological period monitoring area through the band reflectance information; S3. extracting the time series curve of the dual-band enhanced vegetation index EVI2 to obtain the EVI2 time series curve; S4. Construct a reference shape fitting model SMF, and obtain an EVI2 time series fitting curve based on the EVI2 time series curve; S5. Statistically analyze the measured data of the winter wheat phenological period monitoring area, and combine it with the spatial data of the winter wheat phenological period monitoring area to obtain the spatial variation law of the phenological period monitoring area; S6. Construct a geographically weighted shape fitting model GSMF and optimize the SMF phenological period matching algorithm based on the spatial variation law of the phenological period monitoring area; S7. The S algorithm is optimized by golden section search and the bandwidth parameter is selected using the Akaike information criterion (AIC) to extract the phenological period of winter wheat in heterogeneous regions.

2. A winter wheat phenological period monitoring method for heterogeneous regions according to claim 1, It is characterized in that In S2, the calculation formula of the dual-band enhanced vegetation index EVI2 is as follows: In the formula, R NIR is the surface reflectivity in the near-infrared band, R RED is the surface reflectivity in the red light band.

3. A winter wheat phenological phase monitoring method for heterogeneous regions according to claim 2, It is characterized in that In S3, the EVI2 time series curve h(x) is as follows: Where i and j are the amplitude factors of the EVI2 time series curve, which determine the peak height of the EVI2 time series curve; x is the number of days of phenological period monitoring; f and h are the center positions of the peaks of the EVI2 time series curve; a and b are the standard deviations, which determine the distance between the valleys of the EVI2 time series curve.

4. The method for monitoring winter wheat phenology in heterogeneous regions according to claim 3, It is characterized in that In S4, a reference shape fitting model SMF is constructed, and the EVI2 time series curve is averaged to obtain the reference EVI2 time series curve Q(x). Two curve segments are selected from Q(x), and the two curve segments are translated and scaled to obtain the EVI2 time series fitting curve g(x). The calculation process of g(x) is: g(x)=yscale1×Q1(xscale1×x+tshift1)+yscale2×Q2(xscale2×x+tshift2); Where Q1(x) and Q2(x) are two curves selected from Q(x), yscale1 and yscale2 are the Y-coordinate scaling factors corresponding to the two curves, xscale1 and xscale2 are the X-coordinate scaling factors corresponding to the two curves, and tshift1 and tshift2 are the translation steps corresponding to the two curves.

5. The method for monitoring winter wheat phenology in heterogeneous regions according to claim 4, It is characterized in that The Y-coordinate scaling factor, X-coordinate scaling factor and translation step length are all scale conversion parameters, which are selected by evaluating the root mean square error RMSE. The calculation formula is: Where f(x) is the predicted amount of winter wheat in the monitoring area, and m(x) is the actual amount of winter wheat in the monitoring area.

6. A method for monitoring winter wheat phenology in heterogeneous regions according to claim 5, It is characterized in that In S6, the GSMF model uses the model structure and target curve conversion method of SMF, and modifies the phenological period matching algorithm based on the SMF model. The phenological period of winter wheat in the monitoring area is estimated by combining the phenological period differences caused by crop growth differences. The modified model phenological period matching algorithm X est The calculation process is as follows: X est =(a 0(ui,vi) +a 1(ui,vi) xscale (ui,vi) )×(X 0 +a 2(ui,vi) shift (ui,vi) +e (ui,vi) ); Wherein, i = 1, 2, 3, ..., m; m represents the number of observations; a 0(ui,vi) 、a 1(ui,vi) and a 2(ui,vi) Represents the coefficient vector of the phenological period estimation model at the position (ui,vi); xscale (ui,vi) Indicates the X-coordinate scaling factor at the (ui,vi) position; tshift (ui,vi) represents the translation step length at the position (ui,vi); ε (ui,vi) represents the estimation error of the phenological period at the position (ui,vi); X 0 is the reference phenological period of SMF; X est The value represents the estimated phenological period of the target pixel.

7. The method for monitoring winter wheat phenology in heterogeneous regions according to claim 6, It is characterized in that The position of the pixel is taken as the centroid, and the distance between pixels is calculated to obtain the coefficient vector of the local pixel as follows: Where X is the reference phenological period of the GSMF model; X′ is the estimated phenological period of the GSMF model; W (ui,vi) represents an n×n-order diagonal weight matrix, which weights the corresponding observations based on the distance between each observation and the location (ui,vi) in the phenological observation area; [] -1 is the inverse of the matrix; y is the observed value of the phenological period.

8. The method for monitoring winter wheat phenology in heterogeneous regions according to claim 7, It is characterized in that In S7, the calculation formula of Akaike information criterion AIC is: Where AIC is the Akaike information entropy of the Akaike information criterion, m is the number of observations, RSS is the residual sum of squares, S is the phenological observation parameter matrix, and tr(s) is the trace number of matrix S.

Images