Crop change detection method, device and equipment

Abstract

The invention discloses a crop change detection method, a crop change detection device and crop change detection equipment, relates to the technical field of crop detection, and aims to eliminate intensity information of co-polarization channels by using normalized correlation of crops among the co-polarization channels so as to more accurately capture real changes in the growth process of the crops. The method comprises the following steps: after acquiring co-polarization synthetic aperture radar data, constructing a polarization covariance matrix according to image data of crops on a co-polarization channel; according to the polarization covariance matrix, calculating the correlation of the crops among the co-polarization channels after normalization; according to the normalized correlation of the crops among the co-polarization channels, calculating the polarization correlation distance of the crops among different time points; and according to the polarization correlation distance of the crop between different time points, determining a first variable quantity characteristic in the crop growth process, the first variable quantity characteristic at least comprising the change amplitude of the crop growth process.

Description

Technical Field

[0001] The present application relates to the technical field of crop detection, and in particular to a method, device and equipment for detecting crop changes. Background Art

[0002] In the era of rapid globalization, the status of crop production directly affects the national economic development. With the help of satellite remote sensing technology, large-scale, high-precision monitoring of crop production has become an important means to increase crop yields and optimize resource utilization. Among many remote sensing technologies, polarimetric synthetic aperture radar has significant advantages in the application of crop change detection due to its all-weather and all-day observation capabilities and high sensitivity to the physical properties of the target, providing reliable data support for precision agriculture and long-term dynamic monitoring of crops.

[0003] In related technologies, crop change detection technology based on polarization synthetic aperture radar can obtain more information about target characteristics by simultaneously transmitting and receiving radar waves in multiple polarization states. Polarization covariance matrix is ​​usually used for analysis to more comprehensively express target characteristics. Here, polarization covariance matrix is ​​composed of the intensity of different polarization channels and related information between channels, but the intensity is easily affected by environmental factors and cannot accurately reflect the real changes of crops. Summary of the invention

[0004] In view of this, the present application provides a crop change detection method, device and equipment, the main purpose of which is to solve the problem that the intensity in the polarization covariance matrix of the prior art is easily affected by environmental factors and cannot accurately reflect the real changes of crops.

[0005] According to a first aspect of the present application, a crop change detection method is provided, comprising: After acquiring the co-polarization synthetic aperture radar data, the polarization covariance matrix is ​​constructed according to the image data of crops on the co-polarization channel; Calculating the normalized correlation of crops between co-polarization channels according to the polarization covariance matrix; Calculating the polarization correlation distance of the crops at different time points according to the normalized correlation between the crops in the same polarization channels; A first variation characteristic in the growth process of the crop is determined according to the polarization correlation distance of the crop at different time points, wherein the first variation characteristic at least includes a variation amplitude in the growth process of the crop.

[0006] Furthermore, the calculating, according to the polarization covariance matrix, the normalized correlation of crops between co-polarization channels includes: According to the polarization covariance matrix, respectively extracting the intensity of the co-polarization channel after filtering and the correlation of the different polarization channels; According to the filtered intensity of the co-polarization channel and the correlation of different polarization channels, the normalized correlation of crops between the co-polarization channels is calculated.

[0007] Furthermore, the calculating the polarization correlation distance of the crops at different time points according to the normalized correlation of the crops between the co-polarization channels includes: Performing coordinate transformation on the normalized correlation of the crops between the co-polarized channels so that the correlation is transformed from a polar coordinate system to a Cartesian coordinate system; In the Cartesian coordinate system, the polarization correlation distance of the crops at different time points is calculated.

[0008] Further, after calculating the normalized correlation of crops between co-polarization channels according to the polarization covariance matrix, the method further includes: Calculating the polarization correlation distance of the crop on at least one standard scattering mechanism according to the normalized correlation of the crop between co-polarization channels; Establishing polarization state vectors of the crop at different time points according to the polarization correlation distance of the crop on at least one standard scattering mechanism, wherein the polarization state vector is used to represent the characteristics of the crop on at least one standard scattering mechanism; According to the difference between the polarization state vectors of the crops at different time points, a second change quantity characteristic of the crops during their growth process is determined, wherein the second change quantity characteristic at least includes a change type, a change direction, and a change amplitude in at least one standard scattering mechanism during the growth process of the crops.

[0009] Furthermore, before calculating the polarization correlation distance of the crop on at least one standard scattering mechanism according to the normalized correlation of the crop between co-polarization channels, the method further includes: The polarization correlation of at least one standard scattering mechanism is predefined, and the polarization correlation of the at least one standard scattering mechanism comprises at least one of the polarization correlation of volume scattering, the polarization correlation of surface scattering and the polarization correlation of secondary scattering.

[0010] Furthermore, the calculating, according to the normalized correlation of the crops between the co-polarization channels, the polarization correlation distance of the crops on at least one standard scattering mechanism comprises: Performing coordinate conversion on the normalized correlation of the crop between the co-polarized channels and the polarization correlation of at least one standard scattering mechanism, so that the correlation is converted from a polar coordinate system to a Cartesian coordinate system; In the Cartesian coordinate system, the distances between the normalized correlations of the crops between the same polarization channels and the polarization correlations of the at least one standard scattering mechanism are calculated respectively to obtain the polarization correlation distance of the crops on the at least one standard scattering mechanism.

[0011] Furthermore, establishing the polarization state vectors of the crops at different time points according to the polarization correlation distance of the crops on at least one standard scattering mechanism includes: Determining a weight ratio of the crop on at least one standard scattering mechanism according to the polarization correlation distance of the crop on at least one standard scattering mechanism; According to the weight proportion of the crop in at least one standard scattering mechanism, the polarization state vector of the crop at different time points is established.

[0012] According to a second aspect of the present application, a crop change detection device is provided, comprising: A construction unit is used to construct a polarization covariance matrix according to the image data of crops on the co-polarization channel after acquiring the co-polarization synthetic aperture radar data; A first calculation unit is used to calculate the normalized correlation of crops between co-polarization channels according to the polarization covariance matrix; A second calculation unit is used to calculate the polarization correlation distance of the crops at different time points according to the normalized correlation of the crops between the same polarization channels; The first determination unit is used to determine a first variation characteristic in a crop growth process according to the polarization correlation distance of the crop at different time points, wherein the first variation characteristic at least includes a variation amplitude in the crop growth process.

[0013] Furthermore, the first computing unit is specifically configured to: According to the polarization covariance matrix, respectively extracting the intensity of the co-polarization channel after filtering and the correlation of the different polarization channels; According to the filtered intensity of the co-polarization channel and the correlation of different polarization channels, the normalized correlation of crops between the co-polarization channels is calculated.

[0014] Furthermore, the second computing unit is specifically configured to: Performing coordinate transformation on the normalized correlation of the crops between the co-polarized channels so that the correlation is transformed from a polar coordinate system to a Cartesian coordinate system; In the Cartesian coordinate system, the polarization correlation distance of the crops at different time points is calculated.

[0015] Furthermore, the device also includes: a third calculation unit, configured to calculate, after calculating the normalized correlation of the crops between the co-polarization channels according to the polarization covariance matrix, a polarization correlation distance of the crops on at least one standard scattering mechanism according to the normalized correlation of the crops between the co-polarization channels; an establishing unit, configured to establish polarization state vectors of the crop at different time points according to the polarization correlation distance of the crop on at least one standard scattering mechanism, wherein the polarization state vector is used to represent a feature of the crop on at least one standard scattering mechanism; The second determination unit is used to determine a second change quantity characteristic of the crop during its growth process based on the difference between the polarization state vectors of the crop at different time points, wherein the second change quantity characteristic at least includes a change type, a change direction, and a change amplitude in at least one standard scattering mechanism during the crop growth process.

[0016] Furthermore, the device also includes: A definition unit is used to predefine the polarization correlation of at least one standard scattering mechanism before calculating the polarization correlation distance of the crop on at least one standard scattering mechanism based on the normalized correlation of the crop between the same polarization channels, wherein the polarization correlation of the at least one standard scattering mechanism includes at least one of the polarization correlation of body scattering, the polarization correlation of surface scattering and the polarization correlation of secondary scattering.

[0017] Furthermore, the third computing unit is specifically configured to: Performing coordinate conversion on the normalized correlation of the crop between the co-polarized channels and the polarization correlation of at least one standard scattering mechanism, so that the correlation is converted from a polar coordinate system to a Cartesian coordinate system; In the Cartesian coordinate system, the distances between the normalized correlations of the crops between the same polarization channels and the polarization correlations of the at least one standard scattering mechanism are calculated respectively to obtain the polarization correlation distance of the crops on the at least one standard scattering mechanism.

[0018] Furthermore, the establishing unit is specifically used for: Determining a weight ratio of the crop on at least one standard scattering mechanism according to the polarization correlation distance of the crop on at least one standard scattering mechanism; According to the weight proportion of the crop in at least one standard scattering mechanism, the polarization state vector of the crop at different time points is established.

[0019] According to a third aspect of the present application, a computer device is provided, comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the method described in the first aspect when executing the computer program.

[0020] According to a fourth aspect of the present application, a readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps of the method described in the first aspect are implemented.

[0021] By means of the above technical scheme, the present application provides a crop change detection method, device and equipment. Compared with the current prior art method of using polarization covariance matrix to detect crop changes, the present application constructs a polarization covariance matrix according to the image data of crops on the co-polarization channel after acquiring co-polarization synthetic aperture radar data; calculates the normalized correlation of crops between co-polarization channels according to the polarization covariance matrix; calculates the polarization correlation distance of crops at different time points according to the normalized correlation of crops between co-polarization channels; determines the first change quantity feature in the crop growth process according to the polarization correlation distance of crops at different time points, and the first change quantity feature at least includes the change amplitude of the crop growth process. The whole process calculates the normalized correlation of crops between co-polarization channels based on the polarization covariance matrix. The intensity of the co-polarization channels is eliminated through the normalized correlation of crops between co-polarization channels, which can effectively avoid the influence of environmental factors on the intensity. The change characteristics of crops are further detected through the polarization correlation distance between crops at different time points, which can more accurately capture the real changes in the growth process of crops.

[0022] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings: Figure 1 is a flow chart of a crop change detection method in one embodiment of the present application; Figure 2 yes Figure 1 A schematic flow chart of a specific implementation of step 102; Figure 3 yes Figure 1 A schematic flow chart of a specific implementation of step 103; Figure 4 is a flow chart of a crop change detection method in another embodiment of the present application; Figure 5 yes Figure 4 A schematic flow chart of a specific implementation of step 402; Figure 6 yes Figure 4 A schematic flow chart of a specific implementation of step 403; Figure 7 is the average polarization correlation of barley at different time points in this example Figure 8 is the polarization correlation distance of barley in different standard scattering mechanisms in this example; Fig. 9 is the change of the scattering state of barley at different time points in this embodiment; Fig.10 is the overall change in barley between any two time points in this embodiment; Fig.11 In this embodiment, the overall change amount of barley between any two time points is represented by the change matrix of geometric distance; Fig.12 In this embodiment, the overall change of barley between any two time points is represented by the change matrix of Wishart distance; Fig.13 is the change direction and amount of the secondary scattering corresponding to barley in this embodiment; Fig.14 is the change direction and amount of the scattering of the barley corresponding volume in this embodiment; Fig.15 is the change direction and amount of the surface scattering corresponding to barley in this embodiment; Fig.16 is a schematic diagram of the structure of a crop change detection device in one embodiment of the present application; Fig.17 It is a schematic diagram of the device structure of a computer device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0024] Now the content of the present invention will be discussed with reference to several exemplary embodiments. It should be understood that these embodiments are discussed only to enable those skilled in the art to better understand and thus implement the content of the present invention, rather than implying any limitation on the scope of the present invention.

[0025] As used herein, the term "including" and variations thereof are to be interpreted as open-ended terms meaning "including but not limited to." The term "based on" is to be interpreted as "based, at least in part, on." The terms "one embodiment" and "an embodiment" are to be interpreted as "at least one embodiment." The term "another embodiment" is to be interpreted as "at least one other embodiment."

[0026] In the crop change detection scenario, the crop change detection technology based on polarization synthetic aperture radar can obtain more information about the target characteristics by simultaneously transmitting and receiving radar waves in multiple polarization states. The polarization covariance matrix is ​​usually used for analysis to more comprehensively express the target characteristics. Here, the polarization covariance matrix is ​​composed of the intensity of different polarization channels and the related information between channels, but the intensity is easily affected by environmental factors and cannot accurately reflect the real changes of crops.

[0027] In order to solve this problem, this embodiment provides a crop change detection method, such as Figure 1 As shown, the following steps are included: 101. After obtaining the co-polarization synthetic aperture radar data, the polarization covariance matrix is ​​constructed according to the image data of crops on the co-polarization channel.

[0028] Polarimetric synthetic aperture radar systems usually transmit and receive electromagnetic waves in different polarization channels, for example, horizontal polarization channel H and vertical polarization channel V. For a full polarization system, there may be four channels: HH, HV, VH, and VV.

[0029] In this embodiment, the co-polarization channel refers to the same polarization channel for transmission and reception, including two channels, HH and VV. Accordingly, the image data of crops on the co-polarization channel includes image data of the HH channel and image data of the VV channel. Accordingly, the polarization covariance matrix is ​​2×2, which is used to describe the statistical relationship between the HH channel and the VV channel.

[0030] The polarization covariance matrix elements can be constructed by the image data of each pixel on the HH channel and VV channel corresponding to the crop. The image data here is usually a complex number, including amplitude and phase information; then the covariance matrix elements of each pixel are calculated by conjugate multiplication and spatial averaging. For example, for the window around the pixel, the average value of the conjugate multiplication of the image data on the HH channel and VV channel of all pixels in the window is calculated, so as to construct the covariance matrix. It should be noted that since the noise of a single element data is large and cannot be accurately estimated, the covariance matrix needs to estimate the average value, which usually requires time series averaging or spatial averaging to reduce noise. Here, the area around each pixel can be spatially averaged to estimate the statistic.

[0031] Exemplarily, after obtaining the image data of the HH channel and VV channel image data Based on this, the elements of the polarization covariance matrix for each pixel are calculated, including , , , , the corresponding polarization covariance matrix is ​​constructed as follows: ; in, is the polarization covariance matrix, is the spatial averaging operation, is the Hermitian transpose, and accordingly, the elements in the polarization covariance matrix Represents the intensity of the HH channel after filtering, Represents the intensity of the VV channel after filtering, and Represents the correlation between the HH channel and the VV channel.

[0032] 102. Calculate the normalized correlation of crops between co-polarization channels according to the polarization covariance matrix.

[0033] In this embodiment, in the polarization covariance matrix, the diagonal elements represent the intensity of the HH channel and the VV channel after filtering, and the non-diagonal elements are the correlation between the HH channel and the VV channel. The normalized correlation between crops and co-polarized channels can be measured by the complex correlation coefficient, which is a complex number, but its amplitude and phase are usually concerned in practical applications. The specific calculation process of the normalized correlation between crops and co-polarized channels can be referred to the following formula: ; in, is the normalized complex coherence between HH and VV channels, The magnitude of the coherence ranges from 0 to 1. is the phase value of coherence, ranging from -180 degrees to 180 degrees.

[0034] It should be noted that the elements in the polarization covariance matrix are complex numbers, containing amplitude and phase information.

[0035] In actual application scenarios, the correlation between co-polarization channels of the same crop at different growth stages may be different. By analyzing the normalized correlation between co-polarization channels of crops, it can help crop phenology identification or change detection. For example, the co-polarization correlation amplitude of barley is between 0.7-0.9 in the early growth stage, with a high correlation, and between 0.3-0.5 in the middle growth stage, with a medium correlation. The co-polarization correlation phase of barley is around 0 degrees in the early growth stage and 90-180 degrees in the middle growth stage.

[0036] 103. Calculate polarization correlation distances of the crops at different time points according to the normalized correlations between the crops in the same polarization channels.

[0037] It can be understood that the normalized correlation between crops in the same polarization channel reflects the correlation between different polarization channels at the same time point, usually in complex form, including amplitude and phase. In order to obtain the difference between the polarization characteristics of crops at different time points, the correlation of crops at different time points can be compared, and the polarization correlation distance of crops at different time points can be calculated, so as to evaluate the change of crops over time.

[0038] Specifically, according to the normalized correlation of crops between co-polarized channels, the correlation at each time point can be used as a feature vector, and then the distance between the feature vectors at different time points can be calculated to obtain the polarization correlation distance of crops at different time points. For example, assuming that at time points t1 and t2, the correlation coefficients ρ1 and ρ2 of HH and VV are calculated respectively, the distance between the corresponding time points is |ρ1 - ρ2|. Here, the Euclidean distance can be used to calculate the distance between the feature vectors at different time points.

[0039] 104. Determine a first variation characteristic during the growth of the crop according to the polarization correlation distance of the crop at different time points.

[0040] Among them, the first variation feature at least includes the variation range of the crop growth process. A large variation range indicates that the overall scattering mechanism of the crop changes significantly, and a small variation range indicates that the overall scattering mechanism of the crop changes slightly.

[0041] For the change amplitude, it can directly reflect the distance of polarization correlation between different time points. The larger the distance, the larger the change amplitude. In order to more accurately evaluate the change amplitude, the relative difference can also be calculated, that is, the ratio of the difference to the initial polarization correlation distance. This can eliminate the influence of the absolute distance difference under different crops or different measurement conditions, and more intuitively reflect the relative degree of change.

[0042] As an implementation method, the data can be curve-fitted with time as the horizontal axis and the polarization correlation distance as the vertical axis. The slope of the fitting curve reflects the rate at which the polarization correlation distance changes over time. The larger the absolute value of the slope, the greater the change amplitude. By comparing the slopes of the fitting curves at different stages, the dynamic changes in the amplitude of the change during the growth of crops can be determined to determine whether it is a rapid growth stage or a slow growth stage.

[0043] As another implementation, the standard deviation of the change in polarization correlation distance of multiple measurement points within the same time interval can be calculated; a larger standard deviation indicates that the magnitude of crop growth changes within this time period varies greatly between different samples, which may be due to the uneven growth of crops in the farmland, or the influence of local environmental factors; a smaller standard deviation indicates that the magnitude of growth changes is relatively consistent.

[0044] The crop change detection method provided in the embodiment of the present application is compared with the current prior art method of using polarization covariance matrix to detect crop changes. After acquiring co-polarization synthetic aperture radar data, the present application constructs a polarization covariance matrix according to the image data of crops on the co-polarization channel; according to the polarization covariance matrix, calculates the normalized correlation of crops between co-polarization channels; according to the normalized correlation of crops between co-polarization channels, calculates the polarization correlation distance of crops at different time points; according to the polarization correlation distance of crops at different time points, determines the first change quantity feature in the crop growth process, and the first change quantity feature at least includes the change amplitude of the crop growth process. The whole process calculates the normalized correlation of crops between co-polarization channels on the basis of the polarization covariance matrix, eliminates the intensity of the co-polarization channel through the normalized correlation of crops between co-polarization channels, can effectively avoid the influence of environmental factors on the intensity, and further detects the change quantity feature of crops through the polarization correlation distance of crops at different time points, can more accurately capture the real changes in the crop growth process.

[0045] Considering that the polarization covariance matrix contains the scattering information of the target on the two channels HH and VV, it is necessary to extract the elements related to the co-polarization channel from the polarization covariance matrix. In the above embodiment, specifically, Figure 2 As shown, step 102 includes the following steps: 201. According to the polarization covariance matrix, extract the filtered intensity of the same-polarization channel and the correlation of different polarization channels respectively.

[0046] 202. Calculate the normalized correlation of crops between the co-polarization channels according to the filtered intensity of the co-polarization channel and the correlation of different polarization channels.

[0047] In this embodiment, for the co-polarized channels, i.e., the horizontally polarized and vertically polarized channels, the intensity of the co-polarized channels after filtering extracted from the polarization covariance matrix includes the intensity of the HH channel after filtering and the intensity of the VV channel after filtering, and the filtered intensity can reflect the average power of the scattered signals of the HH channel and the VV channel. The correlation of different polarized channels extracted from the polarization covariance matrix includes the correlation between the HH channel and the VV channel, and the correlation can reflect the degree of association between the scattered signals of the HH channel and the VV channel.

[0048] Specifically, the normalized correlation of crops between co-polarized channels can be calculated based on the correlation coefficient calculation principle in statistics. The calculation formula for the normalized correlation of crops between co-polarized channels can be used to take the correlation between the HH channel and the VV channel as the numerator, multiply the intensity of the HH channel after filtering and the intensity of the VV channel after filtering and take the square root as the denominator, and finally divide the numerator by the denominator to obtain the normalized correlation of crops between co-polarized channels.

[0049] It can be understood that the correlation includes correlation amplitude and correlation phase, where the correlation amplitude ranges from 0 to 1. The correlation amplitude is close to 1, indicating that the stronger the correlation between the HH channel and the VV channel, the more similar the scattering characteristics of the crops under these two polarization channels; the correlation amplitude is close to 0, indicating that there is almost no correlation between the two. The correlation phase ranges from -180 degrees to 180 degrees. The correlation phase is close to 0 degrees, indicating that the crops are close to the surface scattering mechanism or the volume scattering mechanism. The correlation phase is close to ±180 degrees, indicating that the crops are close to the secondary scattering mechanism.

[0050] In the above embodiment, specifically, Figure 3 As shown, step 103 includes the following steps: 301. Perform coordinate transformation on the normalized correlation of the crops between co-polarization channels, so that the correlation is transformed from a polar coordinate system to a Cartesian coordinate system.

[0051] 302. In the Cartesian coordinate system, calculate polarization correlation distances of crops at different time points.

[0052] In this embodiment, the polar coordinate system uses radius and angle to represent the position of a point, and the Cartesian coordinate system uses x and y coordinates. Coordinate conversion requires converting the amplitude and phase (polar coordinates) into real and imaginary parts (Cartesian), or converting a correlation parameter represented by a certain polar coordinate into a parameter under rectangular coordinates.

[0053] Specifically, the normalized correlation of crops between co-polarized channels is expressed in polar coordinates as follows: ; in, is the amplitude, indicating the coherence strength of the HH channel and the VV channel, ranging from 0 to 1, and φ is the phase, reflecting the interference phase difference between the two channels, which contains information about the difference in topography, deformation or scattering mechanism.

[0054] Accordingly, in the process of converting the correlation from polar coordinates to Cartesian coordinates, the complex coherence needs to be Decomposed into real and imaginary parts, it can be expressed as follows: ; It can be understood that through coordinate transformation, the normalized correlation between crops in the same polarization channels can be more intuitively analyzed and processed in the Cartesian coordinate system. For example, distance calculation and other operations can be performed on the correlation, providing a more convenient mathematical basis for further research on the scattering characteristics and growth status of crops.

[0055] Furthermore, in the Cartesian coordinate system, it is assumed that the polarization correlation of crops at the first time point is ( ), the polarization correlation of crops at the second time point is ( ). Accordingly, the polarization correlation distance between the crops at the first time point and the second time point can be expressed as follows: ; In the above embodiment, further, Figure 4 As shown, after step 103, the method further includes the following steps: 401. Calculate the polarization correlation distance of the crop on at least one standard scattering mechanism according to the normalized correlation of the crop between co-polarization channels.

[0056] 402. Establish polarization state vectors of the crop at different time points according to the polarization correlation distance of the crop on at least one standard scattering mechanism.

[0057] 403. Determine a second variation characteristic of the crop during its growth process according to the difference between the polarization state vectors of the crop at different time points.

[0058] In this embodiment, at least one standard scattering mechanism includes volume scattering, surface scattering and secondary scattering. Accordingly, the polarization correlation distance of crops on at least one standard scattering mechanism can be expressed as the distance between the correlations of different standard scattering mechanisms and the co-polarization correlation, for example, the distance between the correlation of volume scattering and the co-polarization correlation, and the distance between the correlation of secondary scattering and the co-polarization correlation.

[0059] Specifically, before calculating the polarization correlation distance of crops on at least one standard scattering mechanism, the polarization correlation of at least one standard scattering mechanism may be predefined, where the polarization correlation of at least one standard scattering mechanism includes at least one of the polarization correlation of volume scattering, the polarization correlation of surface scattering, and the polarization correlation of secondary scattering. Accordingly, the polarization correlation of volume scattering may be expressed as , the polarization dependence of surface scattering can be expressed as , the polarization dependence of secondary scattering can be expressed as .

[0060] Specifically, in the above embodiment, if Figure 5As shown, step 402 includes the following steps: 501. Perform coordinate transformation on the normalized correlation of the crop between co-polarization channels and the polarization correlation of at least one standard scattering mechanism, so that the correlation is transformed from a polar coordinate system to a Cartesian coordinate system.

[0061] 502. In the Cartesian coordinate system, respectively calculate the distance between the normalized correlation of the crop between the co-polarization channels and the polarization correlation of the at least one standard scattering mechanism to obtain the polarization correlation distance of the crop on the at least one standard scattering mechanism.

[0062] In this embodiment, the normalized correlation of crops between co-polarized channels is converted from a polar coordinate system to a Cartesian coordinate system. See the content of step 302 above. After the correlation is converted from a polar coordinate system to a Cartesian coordinate system, the correlation can be expressed as: Correspondingly, after the polarization dependence of at least one standard scattering mechanism is converted from polar coordinates to Cartesian coordinates, the polarization dependence of volume scattering is expressed as , the polarization dependence of surface scattering is expressed as , the polarization dependence of secondary scattering is expressed as .

[0063] Correspondingly, the distance between the polarization correlation of the calculated volume scattering and the co-polarization correlation can be expressed as follows: ; Correspondingly, the distance between the polarization correlation of surface scattering and the co-polarization correlation can be expressed as follows: ; Correspondingly, the distance between the polarization correlation of secondary scattering and the co-polarization correlation can be expressed as follows: ; It can be understood that the polarization correlation distance of crops on at least one standard scattering mechanism is used to measure the difference between the scattering characteristics of crops and at least one standard scattering mechanism, and can reflect the type characteristics and changes of the crop scattering mechanism.

[0064] Specifically, the polarization correlation distance of the above crops on at least one standard scattering mechanism can be used to determine the dominant component of the crop scattering mechanism. If the polarization correlation distance with surface scattering is small, it indicates that the crop is close to the standard surface scattering mechanism. If the polarization correlation distance with volume scattering is small, it indicates that the crop scattering mechanism conforms to the standard scattering mechanism with strong depolarization, such as volume scattering.

[0065] Furthermore, in the standard scattering mechanism, the polarization correlation distance of crops in at least one standard scattering mechanism represents the difference of crops in the standard scattering mechanism. The smaller the difference, the greater the co-occurrence of the mechanism. Here, the inverse of the polarization correlation distance can be used as a weight to emphasize the similarity between crops and different standard scattering mechanisms. In this way, the smaller the distance, the higher the similarity, and the weight is naturally greater.

[0066] Specifically, the polarization correlation distance of crops on at least one standard scattering mechanism can be converted into weights, and the weights are further normalized. The weights of the normalized scattering mechanism are used to establish the polarization state vectors of crops at different time points. The normalization here is to make the sum of all weights equal to 1, and the corresponding polarization state vector can represent a probability distribution or proportion. In this way, the characteristics of crops on at least one standard scattering mechanism can be represented by the polarization vector state, and each component corresponds to a different standard scattering mechanism. For example, if at least one standard scattering mechanism includes: volume scattering, surface scattering, and secondary scattering, the corresponding polarization vector state corresponds to three components, and each component corresponds to the normalized weight of a standard scattering mechanism. In this way, the polarization state vector can comprehensively reflect the performance of crops on different standard scattering mechanisms, especially the characteristics on at least one standard scattering mechanism.

[0067] Specifically, in the above embodiment, if Figure 6 As shown, step 403 includes the following steps: 601. Determine a weight ratio of the crop in at least one standard scattering mechanism according to a polarization correlation distance of the crop in at least one standard scattering mechanism.

[0068] 602. Establish polarization state vectors of the crop at different time points according to the weight ratio of the crop in at least one standard scattering mechanism.

[0069] Correspondingly, according to the distance between the polarization correlation of volume scattering and the co-polarization correlation, the weight of volume scattering can be expressed as follows: ; Correspondingly, according to the distance between the polarization correlation of surface scattering and the co-polarization correlation, the weight of surface scattering can be expressed as follows: ; Correspondingly, according to the distance between the polarization correlation of secondary scattering and the co-polarization correlation, the weight of secondary scattering can be expressed as follows: ; The process of further normalizing the weights requires summing the weights, which can be expressed as follows: ; Accordingly, the weight of the post-scattering mechanism is normalized to establish the polarization state vector of the crop at different time points, which can be expressed as follows: ; in, corresponds to the normalized weight of volume scattering, corresponds to the normalized weight of surface scattering, Corresponds to the normalized weight of secondary scattering.

[0070] In actual application scenarios, assuming that the normalized weights of the three standard scattering mechanisms of crops are 0.21, 0.78, and 0.01, respectively, the corresponding polarization state vectors are: = The polarization state vector shows that the scattering characteristics of crops are 78% close to volume scattering (probably vegetation in the growing period) and 21% close to surface scattering (probably soil exposure).

[0071] Furthermore, setting and Representing two different time points, calculating the difference between the polarization state vectors of crops at different time points can be expressed as follows: ; In this embodiment, calculating the difference between the polarization state vectors of crops at different time points may require comparing the changes in the scattering mechanism composition of the same crop area at different time points. For example, surface scattering may be dominant in the early stage, and as the crop grows, secondary scattering increases, and there may be more volume scattering at the mature stage.

[0072] Specifically, the second variation feature includes at least the type, direction and amplitude of changes in at least one standard scattering mechanism during the crop growth process. The type of change is which scattering mechanism changes, such as surface scattering, secondary scattering or volume scattering. The direction of change may refer to which scattering mechanism increases or decreases. The amplitude of change is the size of the change value. Assume that at time point t1, surface scattering accounts for 40%, volume scattering 50%, and secondary scattering 10%; at time point t2, surface scattering 20%, volume scattering 60%, and secondary scattering 20%. The second variation feature includes: surface scattering decreases by 20% (negative direction), volume scattering increases by 10% (positive), and secondary scattering increases by 10% (positive). The type and direction of change are mainly surface scattering decreases, secondary scattering and volume scattering increase, and the amplitudes of change are 20%, 10%, and 10%, respectively.

[0073] For the change type, the distance of the same polarization correlation relative to the polarization correlation point of different standard scattering mechanisms represents the contribution of the corresponding scattering mechanism of the crop. For example, in the early growth stage, the crop may be dominated by body surface scattering, and as it grows, it may gradually change to secondary scattering or other body scattering mechanisms. By comparing the polarization correlation distances of crops at different time points, the change trend of the scattering mechanism can be determined, and then the transition of the crop growth stage can be determined, for example, the transition from the vegetative growth stage to the reproductive growth stage.

[0074] Specifically, the dominant scattering mechanism can be used to roughly determine the growth stage of the crop. Volume scattering mainly describes the scattering effect caused by the increase in leaf cover of the crop canopy. Secondary scattering shows significant changes during the jointing and rapid growth period of the crop. In contrast, the signal of surface scattering mainly comes from the ground surface and is more obvious in the early and mature stages of the crop.

[0075] For the direction of change, if the component of volume scattering in the difference of polarization state vectors of crops at different time points is positive, it indicates that the structure of the crops has become more complex, which may reflect the increase in the density of the vegetation canopy. A positive component of secondary scattering usually means that the stems of the crops are beginning to grow taller. A positive component of surface scattering indicates an increase in surface signals, which usually corresponds to the bare land state before sowing or the farmland after harvest. When the crop coverage is low, this scattering mechanism is dominant. In addition, if the direction of change of a scattering feature is negative, it means that the contribution of this scattering mechanism is relatively reduced, which may be due to crop structure adjustments or growth stage transitions.

[0076] It can be understood that the polarization correlation distance of different scattering mechanisms is a quantification of the difference in polarization state vectors, which can reflect the main direction and intensity of crop scattering. By comparing the polarization state vectors at different time points, if the value of a certain scattering state is negative, it means that the scattering feature of the crop has decreased, which may mean that the growth posture or structure of the crop has changed.

[0077] In the actual application scenario, 11 images of barley corresponding to different time points on the same polarization channel were collected in advance. During the acquisition of the image data, the barley went through different growth stages, first entering the tillering stage on April 19, and then starting stem growth on May 3. On May 24, the plants began to head, and then entered the flowering stage on June 7. On June 13, the fruits began to develop until they entered the maturity stage on July 5, and were finally harvested on July 26.

[0078] In the process of change detection for barley, the average polarization correlation of barley on different dates can be obtained by preprocessing the image. Figure 7 As shown in Figure 1, the amplitude is represented by the radius of the unit circle and the phase is represented by the angle. Figure 7 It can be seen that the position of the polarization correlation on the unit circle is different at different time points. This shows that the polarization complex coherence is sensitive to the growth of barley. Then mark the position of different standard scattering mechanisms on the unit circle, see Figure 8 As shown, secondary scattering is represented by square points, volume scattering is represented by star points, and surface scattering is represented by triangle points. Combining the distance between the average polarization correlation (circle) and the standard scattering mechanism, the scattering state of barley at different time points is determined, see Fig. 9 As shown, from Fig. 9 It can be seen that secondary scattering is dominant in the early to middle stages of barley growth, and then gradually turns to volume scattering. After harvesting, surface scattering becomes the dominant scattering mechanism.

[0079] Furthermore, by calculating the average polarization coherence distance between any two time points, the overall change in barley can be extracted. Fig.10 As shown, the calculated change between any two time points can be expressed as a positive value by the geometric distance, which is represented by the upper triangular part of the matrix. The process is shown in Fig.11 Compared with the Wishart distance including intensity, this process can be seen in Fig.12 The coherence distance proposed in this embodiment can more accurately capture the changes of the crops themselves without being affected by environmental factors. For example, during the growth of barley, the Wishart distance between 0511 and 0515 is large, but in fact the change of the crops between these two time points is very small, and they are still in the same growth stage.

[0080] Furthermore, according to the scattering state of barley at different time points, the change detection matrix of barley is determined. Fig.13 is the changing direction and amplitude of secondary scattering corresponding to barley, Fig.14 is the change direction and amplitude of the scattering of the barley body, Fig.15 is the change direction and amplitude of the surface scattering of barley. The upper triangle indicates an increase, and the lower triangle indicates a decrease. Figure 13-Figure 15 It can be seen that in the middle of barley growth, secondary scattering gradually decreases and volume scattering gradually increases; while in the late growth and harvest period of barley, surface scattering becomes the main growth component. For the convenience of analysis, the change detection matrices of these three different standard scattering mechanisms can be further synthesized through RGB to form a color change detection matrix.

[0081] Further, as Figure 1-Figure 6 The specific implementation of the method, the embodiment of the present application provides a crop change detection device, such as Fig.16 As shown, the device includes: a construction unit 71, a first calculation unit 72, a second calculation unit 73, and a first determination unit 74.

[0082] A construction unit 71 is used to construct a polarization covariance matrix according to the image data of crops on the co-polarization channel after acquiring the co-polarization synthetic aperture radar data; A first calculation unit 72 is used to calculate the normalized correlation of crops between co-polarization channels according to the polarization covariance matrix; A second calculation unit 73 is used to calculate the polarization correlation distance of the crops at different time points according to the normalized correlation of the crops between the same polarization channels; The first determination unit 74 is used to determine a first variation characteristic in the growth process of the crop according to the polarization correlation distance of the crop at different time points, wherein the first variation characteristic at least includes a variation amplitude in the growth process of the crop.

[0083] Compared with the current prior art method of using polarization covariance matrix to detect crop changes, the crop change detection device provided by the embodiment of the present invention constructs a polarization covariance matrix according to the image data of crops on the co-polarization channel after acquiring co-polarization synthetic aperture radar data; calculates the normalized correlation of crops between co-polarization channels according to the polarization covariance matrix; calculates the polarization correlation distance of crops at different time points according to the normalized correlation of crops between co-polarization channels; determines the first change quantity feature in the crop growth process according to the polarization correlation distance of crops at different time points, and the first change quantity feature at least includes the change amplitude of the crop growth process. The whole process calculates the normalized correlation of crops between co-polarization channels on the basis of the polarization covariance matrix, eliminates the intensity of the co-polarization channel through the normalized correlation of crops between co-polarization channels, can effectively avoid the influence of environmental factors on the intensity, and further detects the change quantity feature of crops through the polarization correlation distance of crops at different time points, can more accurately capture the real changes in the crop growth process.

[0084] In a specific application scenario, the first computing unit is specifically used to: According to the polarization covariance matrix, respectively extracting the intensity of the co-polarization channel after filtering and the correlation of the different polarization channels; According to the filtered intensity of the co-polarization channel and the correlation of different polarization channels, the normalized correlation of crops between the co-polarization channels is calculated.

[0085] In a specific application scenario, the second computing unit is specifically used to: Performing coordinate transformation on the normalized correlation of the crops between the co-polarized channels so that the correlation is transformed from a polar coordinate system to a Cartesian coordinate system; In the Cartesian coordinate system, the polarization correlation distance of the crops at different time points is calculated.

[0086] In a specific application scenario, the device further includes: a third calculation unit, configured to calculate, after calculating the normalized correlation of the crops between the co-polarization channels according to the polarization covariance matrix, a polarization correlation distance of the crops on at least one standard scattering mechanism according to the normalized correlation of the crops between the co-polarization channels; an establishing unit, configured to establish polarization state vectors of the crop at different time points according to the polarization correlation distance of the crop on at least one standard scattering mechanism, wherein the polarization state vector is used to represent a feature of the crop on at least one standard scattering mechanism; The second determination unit is used to determine a second change quantity characteristic of the crop during its growth process based on the difference between the polarization state vectors of the crop at different time points, wherein the second change quantity characteristic at least includes a change type, a change direction, and a change amplitude in at least one standard scattering mechanism during the crop growth process.

[0087] In a specific application scenario, the device further includes: A definition unit is used to predefine the polarization correlation of at least one standard scattering mechanism before calculating the polarization correlation distance of the crop on at least one standard scattering mechanism based on the normalized correlation of the crop between the same polarization channels, wherein the polarization correlation of the at least one standard scattering mechanism includes at least one of the polarization correlation of body scattering, the polarization correlation of surface scattering and the polarization correlation of secondary scattering.

[0088] In a specific application scenario, the third computing unit is specifically used to: Performing coordinate conversion on the normalized correlation of the crop between the co-polarized channels and the polarization correlation of at least one standard scattering mechanism, so that the correlation is converted from a polar coordinate system to a Cartesian coordinate system; In the Cartesian coordinate system, the distances between the normalized correlations of the crops between the same polarization channels and the polarization correlations of the at least one standard scattering mechanism are calculated respectively to obtain the polarization correlation distance of the crops on the at least one standard scattering mechanism.

[0089] In a specific application scenario, the establishing unit is specifically used to: Determining a weight ratio of the crop on at least one standard scattering mechanism according to the polarization correlation distance of the crop on at least one standard scattering mechanism; According to the weight proportion of the crop in at least one standard scattering mechanism, the polarization state vector of the crop at different time points is established.

[0090] It should be noted that for other corresponding descriptions of the functional units involved in the crop change detection device provided in this embodiment, reference can be made to Figure 1-Figure 6 The corresponding description in will not be repeated here.

[0091] Based on the above Figure 1-Figure 6 The method shown in the embodiment of the present application accordingly provides a storage medium on which a computer program is stored, and when the program is executed by a processor, the above-mentioned Figure 1-Figure 6 The crop change detection method shown.

[0092] Based on this understanding, the technical solution of the present application can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.), and includes a number of instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each implementation scenario of the present application.

[0093] Based on the above Figure 1-Figure 6 The method shown, and Fig.16 In order to achieve the above-mentioned purpose, the embodiment of the present application also provides a physical device for crop change detection, which can be a computer, a smart phone, a tablet computer, a smart watch, a server, or a network device, etc. The physical device includes a storage medium and a processor; the storage medium is used to store a computer program; the processor is used to execute the computer program to achieve the above-mentioned Figure 1-Figure 6 The crop change detection method shown.

[0094] Optionally, the physical device may also include a user interface, a network interface, a camera, a radio frequency (RF) circuit, a sensor, an audio circuit, a WI-FI module, etc. The user interface may include a display, an input unit such as a keyboard, etc., and the optional user interface may also include a USB interface, a card reader interface, etc. The network interface may optionally include a standard wired interface, a wireless interface (such as a WI-FI interface), etc.

[0095] In an exemplary embodiment, see Fig.17 The physical device includes a communication bus, a processor, a memory and a communication interface, and may also include an input / output interface and a display device, wherein each functional unit can communicate with each other through the bus. The memory stores a computer program, and the processor is used to execute the program stored in the memory and execute the crop change detection method in the above embodiment.

[0096] Those skilled in the art will appreciate that the physical device structure for detecting crop changes provided in this embodiment does not constitute a limitation on the physical device, and may include more or fewer components, or a combination of certain components, or different arrangements of components.

[0097] The storage medium may also include an operating system and a network communication module. The operating system is a program that manages the hardware and software resources of the physical device for crop change detection, and supports the operation of information processing programs and other software and / or programs. The network communication module is used to achieve communication between components within the storage medium, and communication with other hardware and software in the information processing physical device.

[0098] Through the description of the above implementation methods, those skilled in the art can clearly understand that the present application can be implemented by means of software plus necessary general hardware platforms, or by hardware. By applying the technical solution of the present application, compared with the current existing methods, the present application calculates the normalized correlation of crops between co-polarized channels based on the polarization covariance matrix, and eliminates the intensity of the co-polarized channels through the normalized correlation of crops between co-polarized channels, which can effectively avoid the influence of environmental factors on the intensity, and further detects the change characteristics of crops through the polarization correlation distance between crops at different time points, which can more accurately capture the real changes in the growth process of crops.

[0099] Those skilled in the art will appreciate that the accompanying drawings are only schematic diagrams of a preferred implementation scenario, and the modules or processes in the accompanying drawings are not necessarily necessary for implementing the present application. Those skilled in the art will appreciate that the modules in the devices in the implementation scenario can be distributed in the devices of the implementation scenario according to the description of the implementation scenario, or can be changed accordingly and located in one or more devices different from the present implementation scenario. The modules of the above-mentioned implementation scenario can be combined into one module, or can be further split into multiple submodules.

[0100] The above serial numbers of this application are only for description and do not represent the advantages and disadvantages of the implementation scenarios. The above disclosure is only a few specific implementation scenarios of this application, but this application is not limited to them, and any changes that can be thought of by technicians in this field should fall within the scope of protection of this application.

Claims

1. A method for detecting crop changes, characterized in that: include: After acquiring the co-polarization synthetic aperture radar data, the polarization covariance matrix is ​​constructed according to the image data of crops on the co-polarization channel; Calculating the normalized correlation of crops between co-polarization channels according to the polarization covariance matrix; Calculating the polarization correlation distance of the crops at different time points according to the normalized correlation between the crops in the same polarization channels; A first variation characteristic in the growth process of the crop is determined according to the polarization correlation distance of the crop at different time points, wherein the first variation characteristic at least includes a variation amplitude in the growth process of the crop.

2. The method according to claim 1, characterized in that The step of calculating the normalized correlation of crops between co-polarization channels according to the polarization covariance matrix includes: According to the polarization covariance matrix, respectively extracting the intensity of the co-polarization channel after filtering and the correlation of the different polarization channels; According to the filtered intensity of the co-polarization channel and the correlation of different polarization channels, the normalized correlation of crops between the co-polarization channels is calculated.

3. The method according to claim 1, characterized in that Calculating the polarization correlation distance of the crops at different time points according to the normalized correlation of the crops between the co-polarization channels includes: Performing coordinate transformation on the normalized correlation of the crops between the co-polarized channels so that the correlation is transformed from a polar coordinate system to a Cartesian coordinate system; In the Cartesian coordinate system, the polarization correlation distance of the crops at different time points is calculated.

4. The method according to claim 1, characterized in that: After calculating the normalized correlation of crops between co-polarization channels according to the polarization covariance matrix, the method further includes: Calculating the polarization correlation distance of the crop on at least one standard scattering mechanism according to the normalized correlation of the crop between co-polarization channels; Establishing polarization state vectors of the crop at different time points according to the polarization correlation distance of the crop on at least one standard scattering mechanism, wherein the polarization state vector is used to represent the characteristics of the crop on at least one standard scattering mechanism; According to the difference between the polarization state vectors of the crops at different time points, a second change quantity characteristic of the crops during their growth process is determined, wherein the second change quantity characteristic at least includes a change type, a change direction, and a change amplitude in at least one standard scattering mechanism during the growth process of the crops.

5. The method according to claim 4, characterized in that Before calculating the polarization correlation distance of the crop on at least one standard scattering mechanism according to the normalized correlation of the crop between co-polarization channels, the method further includes: The polarization correlation of at least one standard scattering mechanism is predefined, and the polarization correlation of the at least one standard scattering mechanism comprises at least one of the polarization correlation of volume scattering, the polarization correlation of surface scattering, and the polarization correlation of secondary scattering.

6. The method according to claim 4, characterized in that The step of calculating the polarization correlation distance of the crop on at least one standard scattering mechanism according to the normalized correlation of the crop between co-polarization channels comprises: Performing coordinate conversion on the normalized correlation of the crop between co-polarized channels and the polarization correlation of at least one standard scattering mechanism, so that the correlation is converted from a polar coordinate system to a Cartesian coordinate system; In the Cartesian coordinate system, the distances between the normalized correlations of the crops between the same polarization channels and the polarization correlations of the at least one standard scattering mechanism are calculated respectively to obtain the polarization correlation distance of the crops on the at least one standard scattering mechanism.

7. The method according to claim 4, characterized in that The step of establishing polarization state vectors of the crop at different time points according to the polarization correlation distance of the crop on at least one standard scattering mechanism comprises: Determining a weight ratio of the crop on at least one standard scattering mechanism according to the polarization correlation distance of the crop on at least one standard scattering mechanism; According to the weight proportion of the crop in at least one standard scattering mechanism, the polarization state vector of the crop at different time points is established.

8. A crop change detection device, characterized in that: include: A construction unit is used to construct a polarization covariance matrix according to the image data of crops on the co-polarization channel after acquiring the co-polarization synthetic aperture radar data; A first calculation unit is used to calculate the normalized correlation of crops between co-polarization channels according to the polarization covariance matrix; A second calculation unit is used to calculate the polarization correlation distance of the crops at different time points according to the normalized correlation of the crops between the same polarization channels; The first determination unit is used to determine a first variation characteristic in a crop growth process according to the polarization correlation distance of the crop at different time points, wherein the first variation characteristic at least includes a variation amplitude in the crop growth process.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the crop change detection method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the crop change detection method according to any one of claims 1 to 7 are implemented.

Images