Oilseed rape flowering period extraction method based on Sentinel-1 dual-polarization pseudo entropy

Abstract

The invention discloses a rape flowering period extraction method based on Sentinel-1 dual-polarization pseudo entropy, and belongs to the technical field of crop remote sensing, and the method comprises the steps: obtaining time sequence Sentinel-1 images of a plurality of regions; acquiring a dual-polarization pseudo-entropy curve of each region; the area with the highest time resolution of the time sequence Sentinel-1 image in the multiple areas serves as a standard area, the mean value of the dual-polarization pseudo-entropy curve of the standard area is subjected to Fourier fitting interpolation processing and then serves as a standard dual-polarization pseudo-entropy curve, and rape flowering period key points on the standard dual-polarization pseudo-entropy curve are determined; aligning the standard dual-polarization pseudo-entropy curve with the dual-polarization pseudo-entropy curve of each region by adopting a DTW (Dynamic Time Warning) algorithm; and determining the rape flowering stage in the dual-polarization pseudo-entropy curve of each aligned region based on the rape flowering stage key points on the standard dual-polarization pseudo-entropy curve. According to the method, cross-regional rape flowering period extraction can be realized.

Description

Technical Field

[0001] This disclosure relates to the field of remote sensing technology for crops, and in particular to a method for extracting the flowering period of rapeseed based on Sentinel-1 dual-polarization pseudo-entropy. Background Technology

[0002] Food production is crucial to people's well-being and national security. Timely and accurate crop monitoring provides vital agricultural data support for efficient agricultural management, yield estimation, and the formulation of national food policies. Rapeseed, as the world's second-largest oilseed crop, ranks only after soybeans in terms of production and global trade volume, holding a very important position among food crops. Furthermore, rapeseed is one of the world's major sources of edible oil and biofuels. The flowering period is a critical phenological stage for rapeseed growth; the timing and quantity of flowering directly affect the final yield. On the other hand, excessive rainfall during the flowering period can lead to sclerotinia stem rot infection, thus reducing yield. Therefore, monitoring the flowering period of rapeseed is of great significance for preventing sclerotinia stem rot, estimating rapeseed yield, and ensuring national food security.

[0003] Traditional field survey methods, due to their high cost and inefficiency, can no longer meet the current requirements for efficient, rapid, and large-scale crop distribution monitoring. With the development of remote sensing technology, the high efficiency, wide coverage, and low cost of remote sensing imagery have made it an important tool for crop distribution monitoring. The method of using optical remote sensing imagery for monitoring the distribution of crops, including rapeseed, has also been widely applied, achieving good results in both large-scale qualitative analysis and small-scale quantitative analysis. However, passive optical remote sensing imaging is easily affected by weather changes; in areas with heavy cloud cover and rain, complete observation data cannot be obtained, making it difficult to acquire images with sufficient temporal resolution and low cloud coverage.

[0004] Synthetic Aperture Radar (SAR) data, due to its inherent physical characteristics, can effectively solve the problem of obtaining high-quality continuous images encountered by optical imagery. Firstly, SAR has a certain degree of penetration; the C-band (37.5mm-75mm) generally has good penetration, unaffected by clouds and rain, and to some extent overcomes the problem of vegetation index saturation. Secondly, SAR is an active imaging method, capable of stably acquiring temporal images. The use of SAR imagery for crop monitoring and parameter extraction has gradually gained attention from researchers in recent years, with significant results. Because SAR imagery is more stable than optical imagery, it has great application potential in rapeseed area extraction and growth monitoring. Existing research directly uses the changes in the backscattering coefficient of SAR images before and after flowering to extract the rapeseed flowering period. However, the backscattering coefficient is affected by many factors, and the temporal characteristics of rapeseed backscattering at different times and in different regions can fluctuate significantly. This means that related methods for extracting the rapeseed flowering period are only applicable to a single region. Summary of the Invention

[0005] This disclosure provides a method for extracting the flowering period of rapeseed based on Sentinel-1 dual-polarization pseudoentropy, which can achieve cross-regional extraction of the rapeseed flowering period. The technical solution includes at least the following components: Firstly, a method for extracting the flowering period of rapeseed based on Sentinel-1 dual-polarization pseudo-entropy is provided, comprising: acquiring temporal Sentinel-1 images of multiple regions, each region including multiple plots, each plot being planted with rapeseed; acquiring the dual-polarization pseudo-entropy curve of each plot in each region based on the temporal Sentinel-1 images of each region; taking the region with the highest temporal resolution of the temporal Sentinel-1 images in the multiple regions as a standard region, and interpolating the mean of the dual-polarization pseudo-entropy curves of each plot in the standard region using a Fourier fitting method to obtain a standard dual-polarization pseudo-entropy curve, and determining the key points of the rapeseed flowering period on the standard dual-polarization pseudo-entropy curve; aligning the standard dual-polarization pseudo-entropy curve with the dual-polarization pseudo-entropy curve of each plot using the DTW algorithm; and determining the rapeseed flowering period in the aligned dual-polarization pseudo-entropy curve of each plot based on the key points of the rapeseed flowering period on the standard dual-polarization pseudo-entropy curve.

[0006] Alternatively, the following formula can be used to obtain each bipolar pseudo-entropy value in the bipolar pseudo-entropy curve:

[0007] in, This is the pseudo-entropy value of bipolarization. This represents the backscattering coefficient of VV polarization. Represents the backscattering coefficient of VH polarization. These are all intermediate parameters.

[0008] Optionally, the key points of rapeseed flowering period on the standard dual-polarized pseudo-entropy curve are before the rapid rise phase and peak of the standard dual-polarized pseudo-entropy curve, wherein the rapid rise phase is used to indicate the phase in which the rate of change of the slope of the standard dual-polarized pseudo-entropy curve is greater than the rate of change threshold.

[0009] Optionally, the method further includes: preprocessing, interpolating, and standardizing the dual-polarization pseudo-entropy curve of each plot. The preprocessing includes smoothing and denoising the dual-polarization pseudo-entropy curve using SG filtering. The interpolation process uses Fourier fitting to interpolate the preprocessed dual-polarization pseudo-entropy curve. The standardization process uses Z-score standardization to standardize the interpolated dual-polarization pseudo-entropy curve.

[0010] Optionally, aligning the standard bipolar pseudo-entropy curve with the bipolar pseudo-entropy curve of each plot using the DTW algorithm includes: aligning the standard bipolar pseudo-entropy curve with the first bipolar pseudo-entropy curve using the DTW algorithm in the following manner: obtaining a set of constraints, including boundary constraints, continuity constraints, and monotonicity constraints; calculating the cumulative distance matrix between the standard bipolar pseudo-entropy curve and the first bipolar pseudo-entropy curve; finding a matching path from the cumulative distance matrix that minimizes the cumulative distance at each step from the starting endpoint to the ending endpoint, thus obtaining the alignment result between the standard bipolar pseudo-entropy curve and the first bipolar pseudo-entropy curve; wherein the first bipolar pseudo-entropy curve is the bipolar pseudo-entropy curve of a plot in one of the multiple regions.

[0011] Secondly, a rapeseed flowering period extraction device based on Sentinel-1 dual-polarization pseudo-entropy is also provided, comprising: a first acquisition module for acquiring time-series Sentinel-1 images of multiple regions, wherein each region includes multiple plots, and each plot is planted with rapeseed; a second acquisition module for acquiring the dual-polarization pseudo-entropy curve of each plot in each region based on the time-series Sentinel-1 images of each region; and a standard curve determination module for dividing the time-series Sentinel-1 images of the multiple regions into time-series Sentinel-1 images. The region with the highest resolution is used as the standard region. The mean value of the dual-polarization pseudo-entropy curve of each plot in the standard region is interpolated using the Fourier fitting method to obtain the standard dual-polarization pseudo-entropy curve, and the key points of rapeseed flowering period on the standard dual-polarization pseudo-entropy curve are determined. The alignment module is used to align the standard dual-polarization pseudo-entropy curve with the dual-polarization pseudo-entropy curve of each plot using the DTW algorithm. The rapeseed flowering period determination module is used to determine the rapeseed flowering period in the aligned dual-polarization pseudo-entropy curve of each plot based on the key points of rapeseed flowering period on the standard dual-polarization pseudo-entropy curve.

[0012] Optionally, in the second acquisition module, each bipolar pseudo-entropy value in the bipolar pseudo-entropy curve is acquired using the following formula:

[0013] in, This is the pseudo-entropy value of bipolarization. This represents the backscattering coefficient of VV polarization. Represents the backscattering coefficient of VH polarization. These are all intermediate parameters.

[0014] Optionally, in the standard curve determination module, the key point of rapeseed flowering period on the standard dual-polarized pseudo-entropy curve is before the rapid rise phase and peak of the standard dual-polarized pseudo-entropy curve, wherein the rapid rise phase is used to indicate the phase in which the rate of change of the slope of the standard dual-polarized pseudo-entropy curve is greater than the rate of change threshold.

[0015] Optionally, the device further includes a preprocessing module, which is used to preprocess, interpolate, and standardize the dual-polarization pseudo-entropy curve of each plot. The preprocessing includes smoothing and denoising the dual-polarization pseudo-entropy curve using SG filtering. The interpolation process uses Fourier fitting to interpolate the preprocessed dual-polarization pseudo-entropy curve. The standardization process uses Z-score standardization to standardize the interpolated dual-polarization pseudo-entropy curve.

[0016] Optionally, the alignment module is further configured to align the standard bipolar pseudo-entropy curve with the first bipolar pseudo-entropy curve using the DTW algorithm in the following manner: obtaining a set of constraints, including boundary constraints, continuity constraints, and monotonicity constraints; calculating the cumulative distance matrix between the standard bipolar pseudo-entropy curve and the first bipolar pseudo-entropy curve; finding a matching path from the cumulative distance matrix that minimizes the cumulative distance at each step from the starting endpoint to the ending endpoint, thereby obtaining the alignment result between the standard bipolar pseudo-entropy curve and the first bipolar pseudo-entropy curve; wherein the first bipolar pseudo-entropy curve is the bipolar pseudo-entropy curve of a plot of land in one of the multiple regions.

[0017] Thirdly, a computer device is also provided, comprising: a memory and a processor, wherein the memory stores at least one computer program, the at least one computer program being loaded and executed by the processor to perform the rapeseed flowering period extraction method based on Sentinel-1 dual-polarization pseudo-entropy described in the above embodiments.

[0018] Fourthly, a computer-readable storage medium is also provided, wherein at least one computer program is stored in the computer-readable storage medium, and the at least one computer program is loaded and executed by a processor to perform the rapeseed flowering period extraction method based on Sentinel-1 dual-polarization pseudo-entropy described in the above embodiments.

[0019] Fifthly, a computer program product is provided, including a computer program / instructions that, when executed by a processor, implement the method described in the first aspect.

[0020] The beneficial effects of the technical solutions provided in this disclosure include at least the following: In this embodiment, time-series Sentinel-1 images of multiple regions are acquired. Each region includes multiple plots, and each plot is planted with rapeseed. Based on the time-series Sentinel-1 images of each region, a dual-polarization pseudo-entropy curve is obtained for each plot in each region. The region with the highest temporal resolution of the time-series Sentinel-1 images in the multiple regions is taken as the standard region. The mean of the dual-polarization pseudo-entropy curves of each plot in the standard region is interpolated using a Fourier fitting method to obtain the standard dual-polarization pseudo-entropy curve, and key points of rapeseed flowering period on the standard dual-polarization pseudo-entropy curve are determined. The DTW algorithm is used to align the standard dual-polarization pseudo-entropy curve with the dual-polarization pseudo-entropy curve of each plot. Based on the key points of rapeseed flowering period on the standard dual-polarization pseudo-entropy curve, the rapeseed flowering period in the aligned dual-polarization pseudo-entropy curve of each plot is determined. The dual-polarization pseudo-entropy improves the stability of rapeseed flowering period monitoring and has a good application effect in extracting rapeseed flowering period across regions.

[0021] Because the backscattering coefficient is affected by many factors, the backscattering time sequence characteristics of rapeseed at different times and in different regions can fluctuate significantly. This means that existing rapeseed flowering period extraction methods are only applicable to a single region. Therefore, when extracting the rapeseed flowering period across multiple regions, existing methods require calculating the flowering period separately for each region, resulting in a large computational load. In this embodiment, when calculating the rapeseed flowering period for multiple regions, only the key points of the rapeseed flowering period on the standard bipolar pseudo-entropy curve need to be calculated to determine the flowering period for multiple regions. In other words, the method in this embodiment only needs to calculate the rapeseed flowering period once, thereby improving the computational efficiency of the rapeseed flowering period. Attached Figure Description

[0022] To more clearly illustrate the technical solutions in the embodiments of this disclosure, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this disclosure. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0023] Figure 1 A flowchart illustrating a rapeseed flowering period extraction method based on Sentinel-1 bipolar pseudo-entropy provided in an exemplary embodiment of this disclosure is shown. Figure 2 A flowchart is shown below illustrating a method for extracting the flowering period of rapeseed based on Sentinel-1 dual-polarization pseudo-entropy, provided by another exemplary embodiment of this disclosure. Figure 3 This is a schematic diagram of the dual-polarization pseudo-entropy curves before and after SG filtering; Figure 4 This is a schematic diagram of key points during the flowering period of rapeseed on a standard dual-polarization pseudo-entropy curve; Figure 5 This is a schematic diagram of DTW alignment between the standard dual-polarization pseudo-entropy curve and the first dual-polarization pseudo-entropy curve; Figure 6 This illustration shows a schematic diagram of a rapeseed flowering period extraction device based on Sentinel-1 dual-polarization pseudo-entropy provided in an exemplary embodiment of this disclosure; Figure 7 This is a schematic diagram of the structure of a computer device provided in an embodiment of this disclosure. Detailed Implementation

[0024] Unless otherwise defined, the technical or scientific terms used herein shall have the ordinary meaning as understood by one of ordinary skill in the art to which this disclosure pertains. The terms “first,” “second,” “third,” and similar terms used in this patent application specification and claims do not indicate any order, quantity, or importance, but are merely used to distinguish different components. Similarly, the terms “an” or “a” and similar terms do not indicate a quantity limitation, but rather indicate the presence of at least one. The terms “comprising” or “including” and similar terms mean that the elements or objects preceding “comprising” or “including” encompass the elements or objects listed following “comprising” or “including” and their equivalents, and do not exclude other elements or objects. The terms “connected” or “linked” and similar terms are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect.

[0025] To make the objectives, technical solutions, and advantages of this disclosure clearer, the embodiments of this disclosure will be described in further detail below with reference to the accompanying drawings.

[0026] Figure 1 A flowchart illustrating a method for extracting rapeseed flowering period based on Sentinel-1 bipolar pseudo-entropy, provided in an exemplary embodiment of this disclosure, is shown. This method can be executed by a computer device. See also Figure 1 The method includes: In step 101, temporal Sentinel-1 images of multiple regions are acquired.

[0027] The multiple regions, each comprising multiple plots of land, are all planted with rapeseed. These multiple regions are geographically distinct. This disclosure does not limit the location of these multiple regions; that is, they can be geographically far apart (e.g., belonging to different cities or countries) or geographically close together (e.g., belonging to the same city or country).

[0028] Sentinel-1 imagery is a multi-temporal, multi-polarization SAR image. Existing research has shown that multi-temporal and multi-polarization SAR imagery can achieve higher accuracy in crop classification compared to single-temporal SAR imagery.

[0029] For any given region, a temporal Sentinel-1 image refers to Sentinel-1 images of that region at different times. That is, the temporal Sentinel-1 image of each region consists of multiple Sentinel-1 images, and each Sentinel-1 image corresponds to a specific time point.

[0030] After acquiring time-series Sentinel-1 images of multiple regions, it is necessary to perform filtering processing on the time-series Sentinel-1 images of multiple regions, for example, by using a speckle filtering algorithm to filter the time-series Sentinel-1 images of multiple regions.

[0031] For example, a 7×7 Refined Lee filter is used to filter the acquired Sentinel-1 image to reduce the influence of the inherent multiplicative noise of the SAR image, thus obtaining the temporal Sentinel-1 image of each region.

[0032] In step 102, based on the temporal Sentinel-1 image of each region, the dual-polarization pseudo-entropy curve of each plot in each region is obtained.

[0033] In step 103, the region with the highest temporal resolution of the temporal Sentinel-1 image in multiple regions is taken as the standard region. The mean value of the dual-polarization pseudo-entropy curve of each plot in the standard region is interpolated by the Fourier fitting method and taken as the standard dual-polarization pseudo-entropy curve. The key points of rapeseed flowering period on the standard dual-polarization pseudo-entropy curve are determined.

[0034] In step 104, the DTW algorithm is used to align the standard dual-polarization pseudo-entropy curve with the dual-polarization pseudo-entropy curve of each plot.

[0035] In step 105, based on the key points of rapeseed flowering period on the standard dual-polarization pseudo-entropy curve, the rapeseed flowering period in the dual-polarization pseudo-entropy curve of each aligned plot is determined.

[0036] After aligning the standard bipolar pseudo-entropy curve with the bipolar pseudo-entropy curve of each plot, each plot's bipolar pseudo-entropy curve has a point corresponding to the key point of the rapeseed flowering period. The time indicated by this point is the rapeseed flowering period in the plot. In this way, the rapeseed flowering period of each plot in each region can be obtained.

[0037] In this embodiment, time-series Sentinel-1 images of multiple regions are acquired. Each region includes multiple plots, and each plot is planted with rapeseed. Based on the time-series Sentinel-1 images of each region, a dual-polarization pseudo-entropy curve is obtained for each plot in each region. The region with the highest temporal resolution of the time-series Sentinel-1 images in the multiple regions is taken as the standard region. The mean of the dual-polarization pseudo-entropy curves of each plot in the standard region is interpolated using a Fourier fitting method to obtain the standard dual-polarization pseudo-entropy curve, and key points of rapeseed flowering period on the standard dual-polarization pseudo-entropy curve are determined. The DTW algorithm is used to align the standard dual-polarization pseudo-entropy curve with the dual-polarization pseudo-entropy curve of each plot. Based on the key points of rapeseed flowering period on the standard dual-polarization pseudo-entropy curve, the rapeseed flowering period in the aligned dual-polarization pseudo-entropy curve of each plot is determined. The dual-polarization pseudo-entropy improves the stability of rapeseed flowering period monitoring and has a good application effect in extracting rapeseed flowering period across regions.

[0038] Because the backscattering coefficient is affected by many factors, the backscattering time sequence characteristics of rapeseed at different times and in different regions can fluctuate significantly. This means that existing rapeseed flowering period extraction methods are only applicable to a single region. Therefore, when extracting the rapeseed flowering period across multiple regions, existing methods require calculating the flowering period separately for each region, resulting in a large computational load. In this embodiment, when calculating the rapeseed flowering period for multiple regions, only the key points of the rapeseed flowering period on the standard bipolar pseudo-entropy curve need to be calculated to determine the flowering period for multiple regions. In other words, the method in this embodiment only needs to calculate the rapeseed flowering period once, thereby improving the computational efficiency of the rapeseed flowering period.

[0039] Figure 2 A flowchart illustrating a method for extracting rapeseed flowering period based on Sentinel-1 dual-polarization pseudo-entropy, provided in another exemplary embodiment of this disclosure, is shown. This method can be executed by a computer device. See also Figure 2 The method includes: In step 201, temporal Sentinel-1 images of multiple regions are acquired.

[0040] The area comprises multiple regions, each containing multiple plots of land, and each plot is planted with rapeseed.

[0041] The details of step 201 are the same as those in step 101 above, and will not be described in detail here.

[0042] In step 202, based on the temporal Sentinel-1 image of each region, the dual-polarization pseudo-entropy curve of each plot in each region is obtained.

[0043] Optionally, formula (1) can be used to obtain each bipolar pseudo-entropy value in the bipolar pseudo-entropy curve.

[0044] (1) In formula (1), This is the pseudo-entropy value of bipolarization. This represents the backscattering coefficient of VV polarization. Represents the backscattering coefficient of VH polarization. These are all intermediate parameters. The range of values ​​is .

[0045] For each Sentinel-1 image in the time series, the dual-polarization pseudo-entropy value of the corresponding block can be calculated. (Typically, Sentinel-1 images can capture an entire area, and different plots within that area belong to different pixels in the Sentinel-1 images.) Within a single area, the dual-polarization pseudo-entropy curve of a plot can be determined based on the dual-polarization pseudo-entropy value calculated from each Sentinel-1 image in the time-series Sentinel-1 images at the same plot.

[0046] On the one hand, unlike polarimetric decomposition scattering entropy, dual-polarization pseudo-entropy is based on Sentinel-1 ground distance data and only utilizes backscattering intensity information. Its calculation method is simpler and easier to implement in a cloud platform. On the other hand, dual-polarization data has a wider coverage than fully polarimetric data, and dual-polarization pseudo-entropy is mainly derived from cross-polarization ratio conversion, making it more suitable for large-scale monitoring.

[0047] Here, VV (Vertical-Vertical) polarization means that both the transmission and reception of electromagnetic waves are vertically polarized, while VH (Vertical-Horizontal) polarization means that the transmission of electromagnetic waves is vertically polarized, while the reception is horizontally polarized. Vertical polarization means that the electric field vector of the electromagnetic wave is parallel to the vertical plane containing the radar's line-of-sight (incident direction), while horizontal polarization means that the electric field vector of the electromagnetic wave is parallel to the horizontal plane containing the radar's line-of-sight (incident direction).

[0048] VV polarization is a co-polarization, while VH polarization is a cross-polarization. The echo intensity of the same ground feature varies under different polarizations, resulting in different image tones, thus increasing the information available for identifying ground features. By comprehensively using the backscattering coefficients of co-polarization (such as VV polarization) and cross-polarization (VH polarization), radar image information can be significantly increased. Moreover, the information difference between the polarization echoes of vegetation and other different ground features is more sensitive than the difference between different bands. Therefore, the dual-polarization pseudo-entropy value calculated by combining the backscattering coefficients of VV polarization and VH polarization is also quite sensitive to the changes in the phenological stage of rapeseed. The dual-polarization pseudo-entropy curve obtained based on the dual-polarization pseudo-entropy value can reflect the changes in the phenological stage of rapeseed, and thus the flowering period of rapeseed can be determined based on the dual-polarization pseudo-entropy curve.

[0049] In step 203, the dual-polarization pseudo-entropy curve of each plot is preprocessed, interpolated, and standardized.

[0050] Preprocessing includes smoothing and denoising the dual-polarization pseudo-entropy curve using SG filtering, interpolation using Fourier fitting to interpolate the preprocessed dual-polarization pseudo-entropy curve, and standardization using Z-score standardization to standardize the interpolated dual-polarization pseudo-entropy curve.

[0051] Optionally, the SG filtering method is used to smooth and denoise the dual-polarization pseudo-entropy curve. This includes: when using the SG (Savitzky–Golay) filter to smooth and denoise different regions, the fitting order is always selected as 2. For regions with lower time resolution, the window size is set to 5, while for regions with higher time resolution, the window size is set to 11.

[0052] By smoothing and denoising the dual-polarization pseudo-entropy curve using SG filtering, the curve fluctuations caused by differences in satellite flight direction and incident angle can be reduced.

[0053] Figure 3 This is a schematic diagram of the dual-polarization pseudo-entropy curves before and after SG filtering. It can be seen that the dual-polarization pseudo-entropy curve becomes smoother after SG filtering.

[0054] Optionally, the preprocessed dual-polarization pseudo-entropy curve can be interpolated using the Fourier fitting method, including: interpolating the preprocessed dual-polarization pseudo-entropy curve day by day using formula (2).

[0055] (2) In formula (2), The independent variable of the bipolar pseudoentropy curve is time (unit: days). The dependent variable of the bipolar pseudo-entropy curve is the bipolar pseudo-entropy value. These are the fitting coefficients. The frequency of the trigonometric function is related to the growth cycle of rapeseed.

[0056] Although the bipolar pseudo-entropy curves of rapeseed in different regions and plots have similar trends, the absolute values ​​of the bipolar pseudo-entropy values ​​of rapeseed in different regions and plots still differ due to factors such as planting environment and rapeseed species. In order to uniformly use the ideal curve to match rapeseed in different regions, it is necessary to standardize the rapeseed data to reduce the matching error caused by the difference in the magnitude of the bipolar pseudo-entropy values. In this embodiment, the Z-score standardization method is used to standardize the bipolar pseudo-entropy curves of each plot after daily interpolation. Z-score standardization can be expressed by formula (3).

[0057] (3) In formula (3), This is the raw data, that is, the data before standardization. For the standardized results, for The expected value of the bipolar pseudo-entropy value of the corresponding region. The standard deviation is usually represented by the sample mean and standard deviation.

[0058] In step 204, the region with the highest temporal resolution of the temporal Sentinel-1 image in multiple regions is taken as the standard region. The mean value of the dual-polarization pseudo-entropy curve of each plot in the standard region is interpolated by the Fourier fitting method and taken as the standard dual-polarization pseudo-entropy curve. The key points of rapeseed flowering period on the standard dual-polarization pseudo-entropy curve are then determined.

[0059] If a region has the highest temporal resolution in its temporal Sentinel-1 imagery, it indicates that the imagery accuracy for that region is high, and it can be used as a standard region. Furthermore, the rapeseed in the standard region must be in normal flowering condition.

[0060] After averaging the bipolar pseudo-entropy curves of each plot in the standard area, an average curve can be obtained. This average curve also needs to be processed in step 203, that is, Fourier fitting interpolation. Then, based on the relationship between rapeseed phenological period and bipolar pseudo-entropy value, the part of the average curve with large fluctuations in the early stage and the part with obvious decline in the later stage are discarded, thus obtaining the standard bipolar pseudo-entropy curve.

[0061] There is a strong correlation between the phenological stages of rapeseed, especially the flowering and pod-setting stages, and the following patterns are observed: The temperature rises rapidly during the flowering period of rapeseed, then reaches saturation due to the limited penetration ability of C-band electromagnetic waves. During the pod-setting period, the temperature rises slightly but remains relatively stable, and finally declines during the maturity period.

[0062] Based on the above pattern, optionally, the key points of rapeseed flowering period on the standard bipolar pseudo-entropy curve are in the rapid rising phase and before the peak of the standard bipolar pseudo-entropy curve. The rapid rising phase is used to indicate the stage where the rate of change of the slope of the standard bipolar pseudo-entropy curve is greater than the rate of change threshold.

[0063] Figure 4 This is a schematic diagram of the key point of rapeseed flowering period on the standard dual-polarization pseudo-entropy curve. The time corresponding to this key point is the rapeseed flowering period on the standard dual-polarization pseudo-entropy curve.

[0064] Key points on the rapeseed flowering period on the standard bipolar pseudo-entropy curve require manual calibration. In related technologies, extracting the rapeseed flowering period involves calculating the backscattering coefficients of different regions separately and then calibrating the flowering period for each region individually. When multiple regions exist, this requires multiple calibrations to extract the flowering period. In this embodiment, however, determining the flowering period for multiple regions only requires calibrating the key points on the standard bipolar pseudo-entropy curve, meaning the method in this embodiment only requires one calibration, thus improving the efficiency of rapeseed flowering period extraction.

[0065] In step 205, the DTW algorithm is used to align the standard bipolar pseudo-entropy curve with the bipolar pseudo-entropy curve of each plot.

[0066] To reduce the impact of curve changes during the early stages of rapeseed planting and after rapeseed maturity on the matching results, it is also necessary to truncate the standard dual-polarization pseudo-entropy curve and the dual-polarization pseudo-entropy curve of each region. When trunculating, remove the data before the minimum value of the curve and the data after the maximum value of the curve before proceeding to step 205.

[0067] The DTW (Dynamic Time Warping) algorithm includes global matching and local matching. In this embodiment, global matching is used to implement step 205.

[0068] In this scenario, the following three steps are used to align the standard bipolar pseudo-entropy curve with the first bipolar pseudo-entropy curve using the DTW algorithm. The first bipolar pseudo-entropy curve is the bipolar pseudo-entropy curve of a single plot within one of multiple regions.

[0069] The first step is to obtain the set of constraints.

[0070] The set of constraints includes boundary constraints, continuity constraints, and monotonicity constraints.

[0071] Let the standard dual-polarization pseudo-entropy curve be The first bipolar pseudoentropy curve is Among them, subscript It indicates the chronological order in which points on the curve appear, with larger subscripts indicating later times.

[0072] In this case, the boundary constraints include: the endpoints of the standard bipolar pseudo-entropy curve and the first bipolar pseudo-entropy curve are aligned with each other, that is, the starting position of the first bipolar pseudo-entropy curve corresponds to the starting position of the standard bipolar pseudo-entropy curve, and similarly, the end of the first bipolar pseudo-entropy curve corresponds to the end of the standard bipolar pseudo-entropy curve.

[0073] Continuity constraints include: if the currently matched point is The next matching point of this matching point is Then it must satisfy That is, every point in the first bipolar pseudo-entropy curve can find a matching point in the standard bipolar pseudo-entropy curve, and vice versa.

[0074] Monotonicity constraints include: if the currently matched point is The next matching point is Then it satisfies This means that cross-matching is not allowed, and the time of the next pair of matching points must not be earlier than the time of the previous pair of matching points. This monotonicity constraint ensures that the matching results conform to the time order and prevents matching with time discrepancies.

[0075] The second step is to calculate the cumulative distance matrix between the standard dual-polarization pseudo-entropy curve and the first dual-polarization pseudo-entropy curve.

[0076] The cumulative distance matrix includes multiple matching paths, each of which represents a possible matching result, and the cumulative distance of each matching path can be calculated.

[0077] The third step is to find a matching path from the cumulative distance matrix that minimizes the cumulative distance at every step from the starting endpoint to the ending endpoint, thus obtaining the alignment result between the standard bipolar pseudo-entropy curve and the first bipolar pseudo-entropy curve.

[0078] The matching path with the minimum cumulative distance at each step from the starting endpoint to the ending endpoint is the alignment result between the standard bipolar pseudo-entropy curve and the first bipolar pseudo-entropy curve.

[0079] Figure 5 This is a schematic diagram of DTW alignment between the standard dual-polarization pseudo-entropy curve and the first dual-polarization pseudo-entropy curve. Figure 5In the diagram, the blue curve is the first bipolar pseudo-entropy curve, and the black curve is the standard bipolar pseudo-entropy curve. It can be seen that each point on the standard bipolar pseudo-entropy curve corresponds to a point on the first bipolar pseudo-entropy curve.

[0080] For the dual-polar pseudo-entropy curves of other plots besides the first dual-polar pseudo-entropy curve, the same method as the first dual-polar pseudo-entropy curve can be used to align them with the standard dual-polar pseudo-entropy curve, thereby aligning the standard dual-polar pseudo-entropy curve with the dual-polar pseudo-entropy curve of each plot.

[0081] In step 206, based on the key points of rapeseed flowering period on the standard dual-polarization pseudo-entropy curve, the rapeseed flowering period in the dual-polarization pseudo-entropy curve of each aligned plot is determined.

[0082] After aligning the standard bipolar pseudo-entropy curve with the bipolar pseudo-entropy curve of each plot, each plot's bipolar pseudo-entropy curve has a point corresponding to the key point of the rapeseed flowering period. The time indicated by this point is the rapeseed flowering period. In this way, the rapeseed flowering period of each region can be obtained.

[0083] After obtaining the rapeseed flowering period for each region, the extracted rapeseed flowering period can be verified.

[0084] Different verification methods can be used to determine the flowering period of rapeseed in different climates and of different varieties. The verification methods adopted in this invention are as follows: (1) Field data verification. In some areas, the dates corresponding to different phenological stages of rapeseed, including the flowering period, were recorded. For these areas, field data can be directly used for verification. If the extracted flowering period falls within the recorded flowering period, the extraction result is considered to be without deviation. In some areas, the rapeseed flowering period was not recorded, but the rapeseed planting date was. Generally, rapeseed planted earlier flowers relatively earlier. For study areas where the flowering period was not recorded but the planting date was, the Pearson Correlation Coefficient (PCC) between the rapeseed planting date and flowering period can be calculated to indirectly verify the results.

[0085] (2) Validation based on time-series optical data. Existing studies have shown that the Normalized Difference Yellowness Index (NDYI) has a good monitoring ability for the flowering period of rapeseed. Previous studies have used the maximum value of the NDYI time-series curve as a characteristic point of the rapeseed flowering period. For areas where optical images can be obtained before and after the flowering period, in the absence of field data, this invention utilizes the acquired time-series Sentinel-2 optical images to construct the NDYI time-series curve, and uses the rapeseed flowering period extracted by this index to cross-validate the rapeseed flowering period extracted by Sentinel-1. The principle of NDYI extraction of the rapeseed flowering period is as follows: After rapeseed reaches the flowering period, its reflection of blue light decreases slightly, while its reflection of green light increases significantly, similar to the Normalized Difference Vegetation Index (NDVI). By calculating and combining these two fluctuations, the NDYI is obtained.

[0086] For example, NDYI is calculated using formula (4).

[0087] (4) In formula (4), Green represents the green band and Blue represents the blue band.

[0088] (3) Verification using optical images during the flowering period. Generally, it is difficult to obtain high-quality time-series optical data, while it is relatively easier to obtain single-scene optical images. For areas where high-quality time-series optical data cannot be obtained, this invention uses single-scene optical images obtained during the rapeseed flowering period for visual verification.

[0089] The flowering periods of rapeseed under different climates and different varieties extracted in the embodiments of this disclosure were verified using these three methods. The verification results show that the accuracy of the rapeseed flowering periods extracted in the embodiments of this disclosure is relatively high.

[0090] The following are device embodiments of this application. For details not described in detail in the device embodiments, please refer to the above method embodiments.

[0091] Figure 6 A schematic diagram of a rapeseed flowering period extraction device based on Sentinel-1 dual-polarization pseudo-entropy, provided in an exemplary embodiment of this disclosure, is shown. See also Figure 6 The rapeseed flowering period extraction device 600 based on Sentinel-1 dual-polarization pseudo-entropy includes: a first acquisition module 601, a second acquisition module 602, a standard curve determination module 603, an alignment module 604, a rapeseed flowering period determination module 605, and a preprocessing module 606.

[0092] The first acquisition module 601 is used to acquire time-series Sentinel-1 images of multiple regions, each region including multiple plots, and each plot is planted with rapeseed.

[0093] The second acquisition module 602 is used to acquire the dual-polarization pseudo-entropy curve of each plot in each region based on the temporal Sentinel-1 image of each region.

[0094] The standard curve determination module 603 is used to take the region with the highest temporal resolution of the temporal Sentinel-1 image in multiple regions as the standard region, and to take the mean of the dual-polarization pseudo-entropy curve of each plot in the standard region as the standard dual-polarization pseudo-entropy curve after interpolation by the Fourier fitting method, and to determine the key points of rapeseed flowering period on the standard dual-polarization pseudo-entropy curve.

[0095] Alignment module 604 is used to align the standard bipolar pseudo-entropy curve with the bipolar pseudo-entropy curve of each plot using the DTW algorithm.

[0096] The rapeseed flowering period determination module 605 is used to determine the rapeseed flowering period in the aligned bipolar pseudo-entropy curve of each plot based on the key points of the rapeseed flowering period on the standard bipolar pseudo-entropy curve.

[0097] Optionally, in the second acquisition module 602, each bipolar pseudo-entropy value in the bipolar pseudo-entropy curve is acquired using the following formula:

[0098] in, This is the pseudo-entropy value of bipolarization. This represents the backscattering coefficient of VV polarization. Represents the backscattering coefficient of VH polarization. These are all intermediate parameters.

[0099] Optionally, in the standard curve determination module 603, the key point of rapeseed flowering period on the standard bipolar pseudo-entropy curve is before the rapid rise phase and peak of the standard bipolar pseudo-entropy curve. The rapid rise phase is used to indicate the phase where the rate of change of the slope of the standard bipolar pseudo-entropy curve is greater than the rate of change threshold.

[0100] Optionally, the device further includes a preprocessing module 606, which is used to preprocess, interpolate, and standardize the dual-polarization pseudo-entropy curve of each plot. The preprocessing includes smoothing and denoising the dual-polarization pseudo-entropy curve using SG filtering, interpolating the preprocessed dual-polarization pseudo-entropy curve using Fourier fitting, and standardizing the interpolated dual-polarization pseudo-entropy curve using Z-score standardization.

[0101] Optionally, the alignment module 604 is further configured to align the standard bipolar pseudo-entropy curve with the first bipolar pseudo-entropy curve using the DTW algorithm in the following manner: obtaining a set of constraints, including boundary constraints, continuity constraints, and monotonicity constraints; calculating the cumulative distance matrix between the standard bipolar pseudo-entropy curve and the first bipolar pseudo-entropy curve; finding a matching path from the cumulative distance matrix that minimizes the cumulative distance at each step from the starting endpoint to the ending endpoint, thereby obtaining the alignment result between the standard bipolar pseudo-entropy curve and the first bipolar pseudo-entropy curve; wherein the first bipolar pseudo-entropy curve is the bipolar pseudo-entropy curve of a plot of land in one of multiple regions.

[0102] It should be noted that the rapeseed flowering period extraction device based on Sentinel-1 dual-polarization pseudo-entropy provided in the above embodiments is only an example of the division of the above functional modules. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. In addition, the rapeseed flowering period extraction device based on Sentinel-1 dual-polarization pseudo-entropy and the rapeseed flowering period extraction method based on Sentinel-1 dual-polarization pseudo-entropy provided in the above embodiments belong to the same concept. The specific implementation process is detailed in the method embodiment and will not be repeated here.

[0103] The module division in this embodiment is illustrative and represents only one logical functional division. In actual implementation, other division methods are possible. Furthermore, the functional modules in the various embodiments of this disclosure can be integrated into a single processor, exist as separate physical entities, or be integrated into a single module. The integrated modules described above can be implemented in hardware or as software functional modules.

[0104] If the integrated module is implemented as a software functional module and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this disclosure, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a terminal device (which may be a personal computer, mobile phone, or communication device, etc.) or processor to execute all or part of the steps of the methods of the various embodiments of this disclosure. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0105] Figure 7 This is a schematic diagram of the structure of a computer device provided in an embodiment of this disclosure. For example... Figure 7 As shown, the computer device 700 includes a processor 701 and a memory 702.

[0106] Processor 701 may include one or more processing cores, such as a quad-core processor, an octa-core processor, etc. Processor 701 may be implemented using at least one hardware form selected from DSP (Digital Signal Processing), FPGA (Field-Programmable Gate Array), and PLA (Programmable Logic Array). Processor 701 may also include a main processor and a coprocessor. The main processor, also known as a CPU (Central Processing Unit), is used to process data in the wake-up state; the coprocessor is a low-power processor used to process data in the standby state. In some embodiments, processor 701 may integrate a GPU (Graphics Processing Unit), which is responsible for rendering and drawing the content to be displayed on the screen. In some embodiments, processor 701 may also include an AI (Artificial Intelligence) processor, which is used to handle computational operations related to machine learning.

[0107] The memory 702 may include one or more computer-readable storage media, which may be non-transitory. The memory 702 may also include high-speed random access memory and non-volatile memory, such as one or more disk storage devices or flash memory devices. In some embodiments, the non-transitory computer-readable storage media in the memory 702 is used to store at least one instruction, which is executed by the processor 701 to implement the rapeseed flowering period extraction method based on Sentinel-1 dual-polarization pseudo-entropy provided in this disclosure embodiment.

[0108] Those skilled in the art will understand that Figure 7 The structure shown does not constitute a limitation on the computer device 700, and may include more or fewer components than shown, or combine certain components, or use different component arrangements.

[0109] This disclosure also provides a non-transitory computer-readable storage medium, which, when the instructions in the storage medium are executed by the processor of a computer device, enables the computer device to execute the rapeseed flowering period extraction method based on Sentinel-1 dual-polarization pseudo-entropy provided in this disclosure.

[0110] This disclosure also provides a computer program product, including a computer program / instruction, which, when executed by a processor, implements the rapeseed flowering period extraction method based on Sentinel-1 dual-polarization pseudo-entropy provided in this disclosure.

[0111] The above description is merely an optional embodiment of this disclosure and is not intended to limit this disclosure. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this disclosure should be included within the protection scope of this disclosure.

Claims

1. A method for extracting the flowering period of rapeseed based on Sentinel-1 dual-polarization pseudo-entropy, characterized in that, The method includes: Acquire temporal Sentinel-1 images of multiple regions, each region comprising multiple plots, each plot being planted with rapeseed; Based on the temporal Sentinel-1 imagery of each region, the dual-polarization pseudo-entropy curve of each plot in each region is obtained; The region with the highest temporal resolution of the temporal Sentinel-1 image in the multiple regions is taken as the standard region. The mean value of the dual-polarization pseudo-entropy curve of each plot in the standard region is interpolated by the Fourier fitting method and taken as the standard dual-polarization pseudo-entropy curve. The key points of rapeseed flowering period on the standard dual-polarization pseudo-entropy curve are determined. The DTW algorithm is used to align the standard bipolar pseudo-entropy curve with the bipolar pseudo-entropy curve of each plot. Based on the key points of rapeseed flowering period on the standard dual-polarized pseudo-entropy curve, the rapeseed flowering period in the dual-polarized pseudo-entropy curve of each plot after alignment is determined.

2. The method according to claim 1, characterized in that, The following formula is used to obtain each bipolar pseudo-entropy value in the bipolar pseudo-entropy curve: in, This is the pseudo-entropy value of bipolarization. This represents the backscattering coefficient of VV polarization. Represents the backscattering coefficient of VH polarization. These are all intermediate parameters.

3. The method according to claim 1, characterized in that, The key points of the rapeseed flowering period on the standard dual-polarized pseudo-entropy curve are before the rapid rise phase and the peak of the standard dual-polarized pseudo-entropy curve. The rapid rise phase is used to indicate the stage where the rate of change of the slope of the standard dual-polarized pseudo-entropy curve is greater than the rate of change threshold.

4. The method according to any one of claims 1 to 3, characterized in that, The method further includes: For each plot, the dual-polarization pseudo-entropy curve is preprocessed, interpolated, and standardized. The preprocessing includes smoothing and denoising the dual-polarization pseudo-entropy curve using SG filtering. The interpolation uses Fourier fitting to interpolate the preprocessed dual-polarization pseudo-entropy curve. The standardization uses Z-score standardization to standardize the interpolated dual-polarization pseudo-entropy curve.

5. The method according to any one of claims 1 to 3, characterized in that, The step of aligning the standard dual-polarization pseudo-entropy curve with the dual-polarization pseudo-entropy curve of each plot using the DTW algorithm includes: The standard dual-polarization pseudo-entropy curve is aligned with the first dual-polarization pseudo-entropy curve using the DTW algorithm in the following manner: Obtain a set of constraints, which includes boundary condition constraints, continuity constraints, and monotonic constraints; Calculate the cumulative distance matrix between the standard dual-polarization pseudo-entropy curve and the first dual-polarization pseudo-entropy curve; Find a matching path from the cumulative distance matrix that minimizes the cumulative distance at each step from the starting endpoint to the ending endpoint, and obtain the alignment result between the standard dual-polarization pseudo-entropy curve and the first dual-polarization pseudo-entropy curve. The first bipolar pseudo-entropy curve is the bipolar pseudo-entropy curve of a plot of land in one of the multiple regions.

6. A rapeseed flowering stage extraction device based on Sentinel-1 dual-polarization pseudo-entropy, characterized in that, The device includes: The first acquisition module is used to acquire temporal Sentinel-1 images of multiple regions, each region including multiple plots, and each plot is planted with rapeseed; The second acquisition module is used to acquire the dual-polarization pseudo-entropy curve of each plot in each region based on the temporal Sentinel-1 image of each region. The standard curve determination module is used to take the region with the highest temporal resolution of the temporal Sentinel-1 image in the multiple regions as the standard region, and to take the mean of the dual-polarization pseudo-entropy curve of each plot in the standard region as the standard dual-polarization pseudo-entropy curve after interpolation by the Fourier fitting method, and to determine the key points of rapeseed flowering period on the standard dual-polarization pseudo-entropy curve. The alignment module is used to align the standard bipolar pseudo-entropy curve with the bipolar pseudo-entropy curve of each plot using the DTW algorithm; The rapeseed flowering period determination module is used to determine the rapeseed flowering period in the aligned bipolar pseudo-entropy curve of each plot based on the key points of the rapeseed flowering period on the standard bipolar pseudo-entropy curve.

7. The rapeseed flowering stage extraction device based on Sentinel-1 dual-polarization pseudo-entropy according to claim 6, characterized in that, In the second acquisition module, the following formula is used to obtain each bipolar pseudo-entropy value in the bipolar pseudo-entropy curve: in, This is the pseudo-entropy value of bipolarization. This represents the backscattering coefficient of VV polarization. Represents the backscattering coefficient of VH polarization. These are all intermediate parameters.

8. A computer device, characterized in that, The computer device includes a memory and a processor, wherein the memory stores at least one computer program, which is loaded and executed by the processor to implement the method according to any one of claims 1 to 5.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores at least one computer program, which is loaded and executed by a processor to implement the method of any one of claims 1 to 5.

10. A computer program product comprising a computer program / instructions, characterized in that, When the computer program / instructions are executed by the processor, they implement the method described in any one of claims 1 to 5.

Images