A method for identifying crop low-temperature freezing disasters using meteorological and remote sensing data.

Abstract

This application relates to the field of data processing technology, and provides a method for identifying crop low-temperature freezing disasters using meteorological and remote sensing data. The method includes: determining the duration of freezing disasters suffered by crops in a target area, the coefficient of change of air temperature, the coefficient of change of chlorophyll index, the coefficient of change of vegetation index, and the coefficient of change of backscattering coefficient; inputting the duration of freezing disasters, the coefficient of change of air temperature, the coefficient of change of chlorophyll index, the coefficient of change of vegetation index, and the coefficient of change of backscattering coefficient into a target model to obtain crop yield data output by the target model; wherein, the target model is used for crop yield prediction; and based on the crop yield data, determining the crop loss in the target area. This application accurately determines the crop loss in a target area, thereby avoiding the shortcomings of insufficient accuracy in disaster information acquisition and improving the accuracy of crop loss assessment.

Description

Technical Field

[0001] This application relates to the field of data processing technology, specifically to a method for identifying crop low-temperature freezing disasters by combining meteorological and remote sensing data. Background Technology

[0002] Low-temperature freezing disasters are agricultural meteorological disasters caused by the intrusion of strong cold air and cold waves, resulting in a sharp drop in temperature over several consecutive days. This damages crops due to the excessively low ambient temperature, leading to reduced yields.

[0003] Currently, methods for obtaining the spatiotemporal distribution of low-temperature freezing disasters include manual statistical reporting, remote sensing visual interpretation, and remote sensing automatic identification. Manual statistical reporting relies heavily on manual reporting, and the consistency and objectivity of the data need to be confirmed, especially for large areas, making it difficult to effectively reflect the actual situation. Remote sensing visual interpretation requires significant human and material resources, and the results are often inconsistent due to the operator's skill level. Remote sensing automatic identification methods, due to the complexity of information acquisition, currently struggle to obtain reliable results and are subject to considerable uncertainty.

[0004] Therefore, when assessing crop losses based on the above methods, there is a deficiency in the accuracy of disaster information acquisition, which leads to low accuracy in crop loss assessment. Summary of the Invention

[0005] This application provides a method for identifying crop low-temperature freezing disasters by combining meteorological and remote sensing data, in order to solve the problem of low accuracy in crop loss assessment.

[0006] In a first aspect, embodiments of this application provide a method for identifying crop low-temperature freezing disasters using a combination of meteorological and remote sensing data, including:

[0007] Determine the duration of crop freezing damage, the coefficient of change of air temperature, the coefficient of change of chlorophyll index, the coefficient of change of vegetation index, and the coefficient of change of backscattering coefficient in the target area;

[0008] The duration, the coefficient of change of air temperature, the coefficient of change of chlorophyll index, the coefficient of change of vegetation index, and the coefficient of change of backscattering coefficient are input into the target model to obtain the crop yield data output by the target model; wherein, the target model is used for crop yield prediction.

[0009] Based on the crop yield data, the amount of crop loss in the target area is determined.

[0010] In one embodiment, determining the coefficient of variation of crop air temperature in the target area includes:

[0011] Determine the air temperature of the region where a pixel is located in a remote sensing image of the target area;

[0012] Based on the air temperature and the air temperature threshold of the crop corresponding to the pixel, the variation coefficient of crop air temperature in the target area is determined.

[0013] In one embodiment, determining the coefficient of variation of crop chlorophyll index in the target area includes:

[0014] Determine the chlorophyll index of the crop corresponding to the pixel in the remote sensing image of the target area during its current growth stage; wherein the chlorophyll index is determined based on the remote sensing reflectance of the green band and the remote sensing reflectance of the red band of the target area;

[0015] Based on the chlorophyll index and the chlorophyll index threshold of the crop corresponding to the pixel in the current growth stage, the change coefficient of crop chlorophyll index in the target area is determined.

[0016] In one embodiment, the chlorophyll index is determined based on the following formula:

[0017]

[0018] Among them, R g R represents the remote sensing reflectance of the green band corresponding to a pixel in a remote sensing image of the target area. r This represents the remote sensing reflectance of the red band in the region corresponding to the pixel.

[0019] In one embodiment, determining the coefficient of change of crop vegetation index in the target area includes:

[0020] The vegetation index of the crop corresponding to a pixel in the remote sensing image of the target area during its current growth stage is determined. Based on the vegetation index and a threshold value for the vegetation index of the crop corresponding to the pixel during its current growth stage, a variation coefficient of the crop vegetation index in the target area is determined. The vegetation index is determined based on the remote sensing reflectance in the near-infrared band and the remote sensing reflectance in the red band of the target area. The improved soil-atmosphere correction index of the crop corresponding to a pixel in the remote sensing image of the target area during its current growth stage is determined. Based on the improved soil-atmosphere correction index and a threshold value for the improved soil-atmosphere correction index of the crop corresponding to the pixel during its current growth stage, a variation coefficient of the crop vegetation index in the target area is determined. The improved soil-atmosphere correction index is determined based on the remote sensing reflectance in the blue band, the remote sensing reflectance in the near-infrared band, and the remote sensing reflectance in the red band of the target area.

[0021] In one embodiment, the vegetation index is determined based on the following formula:

[0022]

[0023] Among them, R nir R represents the near-infrared reflectance of the region corresponding to a pixel in a remotely sensed image of the target area. r This represents the remote sensing reflectance of the red band in the region corresponding to the pixel;

[0024] The improved soil atmospheric correction index is determined based on the following formula:

[0025]

[0026] Among them, R b R represents the remote sensing reflectance of the blue band in the region corresponding to the pixel. nir R represents the remote sensing reflectance in the near-infrared band of the region corresponding to the pixel. r The values ​​represent the remote sensing reflectance of the red band in the region corresponding to the pixel; L = 1; C1 = 6; C2 = 7.5; G = 2.5.

[0027] In one embodiment, determining the variation coefficient of the crop backscattering coefficient in the target region includes:

[0028] Determine the backscattering coefficient value of the crop corresponding to the pixel in the remote sensing image of the target area during the current growth stage;

[0029] Based on the backscattering coefficient value and the backscattering coefficient threshold of the crop corresponding to the pixel in the current growth period, the variation coefficient of the crop backscattering coefficient in the target area is determined.

[0030] In one embodiment, determining the crop loss in the target area based on the crop yield data includes:

[0031] Determine the pixel area of ​​the remote sensing image of the target area and the number of pixels of damaged crops in the remote sensing image;

[0032] Based on the crop yield data, the pixel area, and the number of pixels, the crop loss in the target area is determined.

[0033] In one embodiment, determining the crop loss in the target area based on the crop yield data, the pixel area, and the number of pixels is achieved using the following expression:

[0034]

[0035] Where TA represents crop loss, p represents the pixel area of ​​the remote sensing image, and Q... r CY represents the number of damaged pixels for crop type r. kThe loss per unit area is represented by the damage level k, which is determined based on the crop yield data and the crop yield threshold; m represents the number of damage levels; and v represents the number of crop types.

[0036] In one embodiment, the expression for the target model is as follows:

[0037]

[0038] Where f(CY|C) represents the ensemble decision tree, n represents the number of sub-decision trees, and f i (CY|C) represents a sub-decision tree of the original output CY given the input variable C;

[0039] CY=f RF (C)+ε;

[0040] Where CY represents crop yield; C represents input variables, which include at least the duration of crop frost damage, the coefficient of change of air temperature, the coefficient of change of chlorophyll index, the coefficient of change of vegetation index, and the coefficient of change of backscattering coefficient; f RF (C) represents a nonlinear function that establishes the relationship between input variables and crop yield; ε represents the error value.

[0041] Secondly, embodiments of this application provide a crop low-temperature freezing disaster identification device that integrates meteorological and remote sensing data, comprising:

[0042] The first determining module is used to determine the duration of crop freezing damage, the coefficient of change of air temperature, the coefficient of change of chlorophyll index, the coefficient of change of vegetation index, and the coefficient of change of backscattering coefficient in the target area.

[0043] A transmission module is used to input the duration, the air temperature variation coefficient, the chlorophyll index variation coefficient, the vegetation index variation coefficient, and the backscattering coefficient variation coefficient into the target model to obtain crop yield data output by the target model; wherein, the target model is used for crop yield prediction.

[0044] The second determining module is used to determine the amount of crop loss in the target area based on the crop yield data.

[0045] Thirdly, embodiments of this application provide an electronic device, including a processor and a memory storing a computer program, wherein the processor executes the program to implement the crop low-temperature freezing disaster identification method based on meteorological and remote sensing data coordination as described in the first aspect.

[0046] Fourthly, embodiments of this application provide a storage medium, which is a computer-readable storage medium including a computer program. When the computer program is executed by a processor, it implements the crop low-temperature freezing disaster identification method based on the coordination of meteorological and remote sensing data as described in the first aspect.

[0047] The crop low-temperature freezing disaster identification method based on the synergy of meteorological and remote sensing data provided in this application can accurately determine crop yield data based on the target model by determining the duration of crop freezing disaster, the coefficient of change of air temperature, the coefficient of change of chlorophyll index, the coefficient of change of vegetation index, and the coefficient of change of backscattering coefficient in the target area. Based on the crop yield data, the crop loss in the target area can be accurately determined. This can avoid the defects of insufficient accuracy in disaster information acquisition and improve the accuracy of crop loss assessment. Attached Figure Description

[0048] To more clearly illustrate the technical solutions in this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0049] Figure 1 This is a flowchart illustrating the crop low-temperature freezing disaster identification method based on the collaboration of meteorological and remote sensing data provided in this application embodiment;

[0050] Figure 2 This is a schematic diagram of the process for determining crop loss in the crop low-temperature freezing disaster identification method based on the collaboration of meteorological and remote sensing data provided in the embodiments of this application;

[0051] Figure 3 This is a schematic diagram of the functional modules of an embodiment of the crop low-temperature freezing disaster identification device that integrates meteorological and remote sensing data according to this application;

[0052] Figure 4 This is a schematic diagram of the structure of the electronic device provided in the embodiments of this application. Detailed Implementation

[0053] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions of this application will be clearly and completely described below with reference to the accompanying drawings of the embodiments. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0054] In the process of creating this application, the applicant considered the following aspects:

[0055] Low-temperature freezing disasters are agricultural meteorological disasters mainly caused by the intrusion of strong cold air and cold waves, resulting in a sharp drop in temperature over several consecutive days. This damages crops due to the excessively low ambient temperatures, leading to reduced yields. Cold air is the primary cause of low-temperature freezing disasters. During spring and autumn, frequent convergence of cold air from the north and warm, humid air from the south often results in prolonged periods of low temperatures and rainy weather. Strong cold air, especially the outbreak and southward movement of cold waves, causes a sharp drop in temperature, leading to disasters such as "late spring cold snaps" and frost.

[0056] The accumulation, intensification, southward movement, and outbreak of cold air is a process that occurs under specific weather conditions. Under intense radiative cooling, large-scale cold air masses form in the Arctic Ocean and Siberia. The adiabatic expansion and cooling of the air within these cold air masses further intensifies the mass. When the cold air mass reaches a certain intensity, it will erupt southward under corresponding circulation patterns, resulting in a cold wave.

[0057] Frost typically occurs on cold, clear (cloudless or with very little cloud cover), windless or lightly windy nights with low humidity. It is primarily caused by the large amount of heat radiated outwards from the ground and plant surfaces, cooling the near-surface air to below 0°C. Frost is related to ground radiative cooling, and cloud cover is one of the most significant factors influencing the amount of radiation; clouds are unfavorable for frost formation. Wind promotes the mixing of near-surface and upper-level air, thus reducing the degree of ground cooling; therefore, strong winds are also unfavorable for frost formation.

[0058] Currently, methods for obtaining the spatiotemporal distribution of low-temperature freezing disasters include manual statistical reporting, remote sensing visual interpretation, and remote sensing automatic identification. Manual statistical reporting primarily involves reporting the extent and area of ​​damage through administrative hierarchical reporting. Remote sensing visual interpretation mainly uses medium- to high-resolution remote sensing imagery to visually identify the area and type of damage caused by low-temperature freezing. Remote sensing automatic identification mainly uses remote sensing classification and other methods combined with field investigations to obtain the overall situation of low-temperature freezing disasters. Manual statistical reporting has advantages such as speed and directness, but its main drawback is its reliance on manual reporting, raising concerns about data consistency and objectivity, especially for large areas where it may struggle to effectively reflect the actual situation; furthermore, the data update frequency needs improvement. Remote sensing visual interpretation offers the advantage of relatively objectively identifying disaster distribution and severity, but it is costly in terms of manpower and resources, and the results are often inconsistent due to the operator's skill level. Remote sensing automatic identification is currently the most widely used method, offering advantages such as objectivity, speed, and low cost. There are many remote sensing automatic identification methods, such as machine learning methods, time-series vegetation index extraction methods, and vegetation canopy moisture observation methods. These methods can automatically and quickly obtain crop distribution information. However, due to the complexity of information acquisition, current methods generally struggle to obtain effective results and are subject to significant uncertainty.

[0059] Based on the above considerations, the applicant has proposed various embodiments of this application.

[0060] The following describes in detail, with reference to embodiments, the crop low-temperature freezing disaster identification method, apparatus, electronic equipment and storage medium based on the collaboration of meteorological and remote sensing data provided in this application.

[0061] Figure 1 This is a flowchart illustrating the crop low-temperature freezing disaster identification method based on the collaboration of meteorological and remote sensing data, provided in an embodiment of this application. (Refer to...) Figure 1 This application provides a method for identifying crop low-temperature freezing disasters using meteorological and remote sensing data, which may include:

[0062] Step 100: Determine the duration of crop freezing damage, the coefficient of change of air temperature, the coefficient of change of chlorophyll index, the coefficient of change of vegetation index, and the coefficient of change of backscattering coefficient in the target area.

[0063] It should be noted that the execution subject of the crop low-temperature freezing disaster identification method based on meteorological and remote sensing data collaboration provided in this application embodiment can be a computer device, such as a server, mobile phone, tablet computer, laptop computer, handheld computer, vehicle electronic device, wearable device, ultra-mobile personal computer (UMPC), netbook, or personal digital assistant (PDA), etc.

[0064] In this application, the target area can be an area that needs to be assessed for crop loss, which can be determined based on location range information or by dividing it on an electronic map. This application does not impose any specific limitations.

[0065] It should be noted that one or more crops may be grown in the target area of ​​this application.

[0066] The criterion for judging whether a crop has suffered from freezing damage in this application can be achieved by whether the crop’s air temperature change coefficient is greater than the crop’s air temperature threshold. The air temperature threshold can be a temperature value that allows the crop to be in its optimal growth state, determined by human experience or other means.

[0067] In this application, the duration can be the time difference between the end time and the start time of the crop suffering from freezing damage.

[0068] Specifically, it can be determined using the following formula:

[0069] Δt=t i+1 -t i0 ;

[0070] Among them, t i+1 t represents the end time of frost damage suffered by crop of type i. i0 Indicates the start time when crops of type i suffer from freezing damage.

[0071] The crop air temperature variation coefficient is the coefficient between the monitored air temperature and the crop air temperature threshold.

[0072] The crop chlorophyll index variation coefficient is the coefficient between the crop's chlorophyll index at the current growth stage and the chlorophyll index threshold for the corresponding growth stage. The chlorophyll index threshold can be the chlorophyll index of the crop under optimal growth conditions, determined through human experience or other methods. In this application, the "current growth stage" refers to the growth stage of the crop at the time of crop loss assessment.

[0073] In this application, the chlorophyll index of crops in the current growth stage can be determined by the remote sensing reflectance of the green band and the remote sensing reflectance of the red band in the corresponding region.

[0074] The crop vegetation index variation coefficient can be the coefficient between the crop's vegetation index during the current growth stage and the vegetation index threshold for the corresponding growth stage. The vegetation index threshold can be a vegetation index value that characterizes a crop under high cover, determined based on human experience or other methods.

[0075] The vegetation index in this application can be determined based on the remote sensing reflectance in the near-infrared band and the remote sensing reflectance in the red band of the crop's corresponding area.

[0076] The coefficient of variation of crop vegetation index can also be the coefficient of variation between the improved soil-atmosphere correction index of the crop in the current growth stage and the threshold value of the improved soil-atmosphere correction index of the crop in the corresponding growth stage. The threshold value of the improved soil-atmosphere correction index can be a value of the improved soil-atmosphere correction index that characterizes the crop under medium to low cover, determined by human experience or other methods.

[0077] The improved soil atmospheric correction index in this application can be determined based on the remote sensing reflectance of the blue band, the near-infrared band, and the red band of the crop's corresponding region.

[0078] The crop backscattering coefficient variation coefficient is the coefficient by which the crop's backscattering coefficient value changes between the value at the current growth stage and the backscattering coefficient threshold at the corresponding growth stage. The backscattering coefficient threshold can be a coefficient value determined based on human experience or other methods.

[0079] It should be noted that this application can obtain data such as the duration of crop freezing damage, air temperature variation coefficient, chlorophyll index variation coefficient, vegetation index variation coefficient, and backscattering coefficient variation coefficient for each pixel in the remote sensing image of the target area. This is to facilitate the acquisition of crop yield data for each pixel and its corresponding crop.

[0080] Step 200: Input the duration, air temperature variation coefficient, chlorophyll index variation coefficient, vegetation index variation coefficient and backscattering coefficient variation coefficient into the target model to obtain the crop yield data output by the target model.

[0081] The target model in this application is used for crop yield prediction.

[0082] The target model can be a random forest model consisting of multiple sub-decision trees. Its expression can be shown in the following formula:

[0083]

[0084] Where f(CY|C) represents the ensemble decision tree, n represents the number of sub-decision trees, and f i (CY|C) represents a sub-decision tree of the original output CY given the input variable C;

[0085] in,

[0086] CY=f RF (C)+ε;

[0087] CY represents crop yield; C represents input variables, which include at least the duration of crop frost damage, the coefficient of change in air temperature, the coefficient of change in chlorophyll index, the coefficient of change in vegetation index, and the coefficient of change in backscattering coefficient; f RF (C) represents a nonlinear function that establishes the relationship between input variables and crop yield; ε represents the error value.

[0088] In this application, a nonlinear functional relationship between various input variables and crop yield can be established first. The specific functional relationship is not limited in this application. The ultimate goal is to establish a nonlinear functional relationship between various input variables and crop yield.

[0089] Furthermore, a sub-decision tree can be constructed based on the aforementioned nonlinear functional relationship, given the crop yield from the training input variables. A random forest model can then be formed based on multiple sub-decision trees.

[0090] This application acquires sample remote sensing images and obtains the following parameters for each pixel in the sample images: duration of crop frost damage, coefficient of change in air temperature, coefficient of change in chlorophyll index, coefficient of change in vegetation index, and coefficient of change in backscattering coefficient. It also obtains labels for each pixel indicating whether the crop has suffered frost damage, whether there have been changes in chlorophyll content, whether there have been changes in crop cover, whether there have been changes in crop morphology, and yield. The resulting random forest model is trained based on these samples and labels. After training and validation, the target model is obtained.

[0091] In this application, in order to determine the changes in chlorophyll, cover and plant morphology of crops, remote sensing models of crop biophysical and structural parameters are selected and established. These remote sensing models of crop biophysical and structural parameters can also be referred to as time series estimation models of crop biophysical and structural parameters.

[0092] It should be noted that, in the process of training the random forest model, this application can also add the crop damage level corresponding to the pixel as a label, so that the trained target model can not only predict the crop yield based on the input data, but also predict the crop damage level corresponding to the pixel.

[0093] It should be noted that the degree of crop damage in this application can be divided into multiple levels according to the actual situation.

[0094] It's important to note that random forests build multiple sub-decision trees during the training phase, and then calculate the average prediction value of these sub-decision trees as the output of the method. The random forest model can divide the input feature space into a large number of regression trees, called a forest, where each tree is generated from a guide sample. A guide sample contains approximately two-thirds of the training samples, and the remaining one-third of the data is used to validate each tree. The result of a random forest is the average prediction result of each sub-decision tree.

[0095] Therefore, after obtaining information such as the duration of crop freezing damage in the target area, the coefficient of change in air temperature, the coefficient of change in chlorophyll index, the coefficient of change in vegetation index, and the coefficient of change in backscattering coefficient, this application can input the above data into the target model. After the target model completes the crop yield prediction, the crop yield data output by the target model is obtained. It should be noted that the crop yield data in this application may include both crop yield and crop damage degree.

[0096] Step 300: Based on crop yield data, determine the amount of crop loss in the target area.

[0097] After obtaining crop yield data, this application can obtain the highest yield of each type of crop in the area corresponding to the pixel as a reference yield.

[0098] Furthermore, the unit area loss for each degree of damage can be determined based on the crop yield at each level of damage and the corresponding reference yield.

[0099] Furthermore, the crop loss caused by freezing disaster in the target area can be determined based on the unit area loss of each degree of damage, the pixel area of ​​the remote sensing image, and the number of damaged pixels of each crop.

[0100] The crop low-temperature freezing disaster identification method based on the synergy of meteorological and remote sensing data provided in this application can accurately determine crop yield data based on the target model by determining the duration of crop freezing disaster, the coefficient of change of air temperature, the coefficient of change of chlorophyll index, the coefficient of change of vegetation index, and the coefficient of change of backscattering coefficient in the target area. Based on the crop yield data, the crop loss in the target area can be accurately determined. This can avoid the defects of insufficient accuracy in disaster information acquisition and improve the accuracy of crop loss assessment.

[0101] In one embodiment, determining the coefficient of variation of crop air temperature in the target area includes:

[0102] Step 1011: Determine the air temperature of the area where the pixel is located in the remote sensing image of the target area;

[0103] This application can obtain spatial distribution maps of air temperature and humidity over time series using data such as air temperature and humidity released by meteorological stations in various locations, through spatial interpolation and other methods. Based on the time series optical and microwave data, remote sensing indices reflecting changes in different growth stages can be calculated.

[0104] Furthermore, the air temperature of the region where each pixel in the remote sensing image of the target area is located can be determined based on the spatial distribution map, and the air temperature of each region can be used as the air temperature of the crop corresponding to that pixel.

[0105] There may be multiple pixels that correspond to the same air temperature.

[0106] Step 1012: Based on the air temperature and the air temperature threshold of the corresponding crop in the pixel, determine the variation coefficient of crop air temperature in the target area.

[0107] Furthermore, this application can also obtain the air temperature threshold of the crop corresponding to each pixel, and determine the air temperature variation coefficient of the crop corresponding to each pixel in the target area based on the following formula, combining each set of air temperatures with the air temperature threshold:

[0108]

[0109] Among them, T i T represents the air temperature threshold for crop type i. i,j This represents the air temperature during the growth period j for crop of type i.

[0110] This application also proposes a meteorological early warning method for crop low-temperature freezing disasters.

[0111] In this application, the coefficient of change in crop air temperature can be called the temperature drop index. When the temperature drop index is greater than or equal to a preset temperature drop index threshold, the crop is determined to have suffered frost damage. It should be noted that different temperature drop index thresholds can be set for different crops.

[0112] This embodiment can accurately determine the crop air temperature variation coefficient in the target area, enabling the accurate determination of crop yield data based on the target model combined with the crop air temperature variation coefficient and other data. Based on the crop yield data, the crop loss in the target area can be accurately determined, thereby avoiding the deficiency of insufficient accuracy in disaster information acquisition and improving the accuracy of crop loss assessment.

[0113] Furthermore, the coefficient of variation of crop chlorophyll index in the target area is determined, including:

[0114] Step 1021: Determine the chlorophyll index of the crop corresponding to the pixel in the remote sensing image of the target area during the current growth stage;

[0115] The chlorophyll index is determined based on the remote sensing reflectance of the green band and the remote sensing reflectance of the red band of the target area.

[0116] This application can determine the chlorophyll index of the corresponding pixel based on the remote sensing reflectance of the region corresponding to each pixel in the remote sensing image of the target area, and determine the chlorophyll index of each pixel as the chlorophyll index of the crop corresponding to that pixel in the current growth stage.

[0117] The chlorophyll index can be determined using the following formula:

[0118]

[0119] Among them, R g R r These represent the remote sensing reflectance of the green band and the remote sensing reflectance of the red band, respectively.

[0120] In this application, R g R r These represent the remote sensing reflectance of the green band and the remote sensing reflectance of the red band in the remote sensing image corresponding to the pixel in the target area, respectively.

[0121] Step 1022: Based on the chlorophyll index and the chlorophyll index threshold of the crop corresponding to the pixel in the current growth stage, determine the change coefficient of crop chlorophyll index in the target area.

[0122] After obtaining the chlorophyll index of the crop corresponding to each pixel in the current growth stage, this application can obtain the chlorophyll index threshold of each crop in the corresponding growth stage. For example, after obtaining the chlorophyll index of crop of type i in growth stage j, the chlorophyll index threshold of crop of type i in growth stage j can be obtained.

[0123] Specifically, the coefficient of variation of the chlorophyll index of the crop corresponding to each pixel can be determined using the following formula:

[0124]

[0125] Among them, CHI i CHI represents the chlorophyll index threshold for crop type i at growth stage j. i,j This represents the chlorophyll index of crop type i during growth stage j.

[0126] It should be noted that if the chlorophyll index change coefficient is greater than or equal to the pre-set chlorophyll index change coefficient threshold, then it is determined that the chlorophyll of the crop corresponding to that pixel has changed.

[0127] This embodiment can accurately determine the variation coefficient of crop chlorophyll index in the target area, so that crop yield data can be accurately determined based on the target model combined with the variation coefficient of crop chlorophyll index and other data. Based on the crop yield data, the crop loss in the target area can be accurately determined, thereby avoiding the deficiency of insufficient accuracy in disaster information acquisition and improving the accuracy of crop loss assessment.

[0128] Furthermore, the coefficients of variation of crop vegetation indices in the target area are determined, including:

[0129] Step 1031: Determine the vegetation index of the crop corresponding to the pixel in the remote sensing image of the target area during the current growth period. Based on the vegetation index and the vegetation index threshold of the crop corresponding to the pixel during the current growth period, determine the change coefficient of the crop vegetation index in the target area. The vegetation index is determined based on the remote sensing reflectance of the near-infrared band and the remote sensing reflectance of the red band in the target area.

[0130] This application can determine the vegetation index of the crop corresponding to each pixel in the remote sensing image of the target area during the current growth period, and obtain the vegetation index threshold of the crop or all crops corresponding to each pixel during the current or all growth periods.

[0131] Furthermore, the variation coefficient of the vegetation index for each crop corresponding to each pixel is determined using the following formula, thereby obtaining the variation coefficient of the crop vegetation index in the target area:

[0132]

[0133] Among them, EVI i EVI represents the vegetation index threshold for crop type i during growth stage j, used for monitoring crops with high cover. i,j This represents the vegetation index of crop type i during its growth stage j, used for monitoring crops with high cover.

[0134] The vegetation index for each pixel during its current growth stage is determined using the following formula:

[0135]

[0136] Wherein, NDVI represents the normalized vegetation index, R nir R represents the remote sensing reflectance in the near-infrared band. r This represents the remote sensing reflectance in the red band.

[0137] In this application, R nir R represents the near-infrared reflectance of the region corresponding to a pixel in a remotely sensed image of the target area. rThis represents the remote sensing reflectance of the red band corresponding to the pixel in the remote sensing image of the target area.

[0138] as well as,

[0139] Step 1041: Determine the improved soil-atmosphere correction index of the crop corresponding to the pixel in the remote sensing image of the target area during the current growth period. Based on the improved soil-atmosphere correction index and the threshold of the improved soil-atmosphere correction index of the crop corresponding to the pixel during the current growth period, determine the change coefficient of the crop vegetation index in the target area. The improved soil-atmosphere correction index is determined based on the remote sensing reflectance of the blue band, the remote sensing reflectance of the near-infrared band, and the remote sensing reflectance of the red band in the target area.

[0140] This application can also determine the improved soil-atmosphere correction index of the crop corresponding to each pixel in the remote sensing image of the target area during the current growth period, and obtain the improved soil-atmosphere correction index threshold of the crop or all crops corresponding to each pixel during the current or all growth periods.

[0141] Furthermore, the variation coefficient of the improved soil-atmosphere correction index for each crop corresponding to each pixel is determined using the following formula, thereby obtaining the variation coefficient of the crop vegetation index in the target area:

[0142]

[0143] Among them, NDVI i The modified soil-atmosphere correction index threshold for crop type i during growth stage j is used for monitoring crops with medium to low cover; NDVI i,j The improved soil-atmosphere correction index represents the growth stage j of crop of type i, used for monitoring crops with medium to low cover.

[0144] The improved soil-atmosphere correction index for each pixel during the current growth stage is determined using the following formula:

[0145]

[0146] Wherein, EVI represents the improved soil-atmosphere correction index; R b R represents the remote sensing reflectance in the blue band. nir R represents the remote sensing reflectance in the near-infrared band. r This represents the remote sensing reflectance in the red band; parameters L = 1; C1 = 6; C2 = 7.5; G = 2.5.

[0147] In this application, R b R represents the remote sensing reflectance of the blue band corresponding to a pixel in a remote sensing image of the target area. nirR represents the near-infrared reflectance of the region corresponding to a pixel in a remotely sensed image of the target area. r This represents the remote sensing reflectance of the red band corresponding to the pixel in the remote sensing image of the target area.

[0148] This embodiment can accurately determine the variation coefficient of crop vegetation index in the target area, so that crop yield data can be accurately determined based on the target model combined with the variation coefficient of crop vegetation index and other data. Based on the crop yield data, the crop loss in the target area can be accurately determined, thereby avoiding the deficiency of insufficient accuracy in disaster information acquisition and improving the accuracy of crop loss assessment.

[0149] If the change coefficient of the crop vegetation index is greater than or equal to the preset change coefficient threshold of the vegetation index, or greater than or equal to the preset change coefficient threshold of the improved soil-atmosphere correction index, then it is determined that the crop cover corresponding to that pixel has changed.

[0150] It should be noted that, in determining whether there has been a change in crop cover, this application can use EVI for comparison when the NDVI value is greater than 0.45 to 0.5, and use NDVI for comparison when it is less than this range.

[0151] Furthermore, the variation coefficient of the crop backscattering coefficient in the target area is determined, including:

[0152] Step 1051: Determine the backscattering coefficient value of the crop corresponding to the pixel in the remote sensing image of the target area during the current growth period;

[0153] This application can determine the backscattering coefficient value of the crop corresponding to each pixel in the remote sensing image of the target area during the current growth period, and obtain the backscattering coefficient threshold of the crop or all crops corresponding to each pixel during the current or all growth periods.

[0154] It should be noted that the backscattering coefficient value of the crop corresponding to each pixel in the current growth stage can be directly obtained in this application. The specific method of obtaining the value is not limited and can be implemented based on existing technologies.

[0155] Step 1052: Based on the backscattering coefficient value and the backscattering coefficient threshold of the crop corresponding to the pixel in the current growth period, determine the variation coefficient of the crop backscattering coefficient in the target area.

[0156] Furthermore, this application can determine the variation coefficient of the backscattering coefficient of the crop corresponding to each pixel in the target area using the following formula:

[0157]

[0158] Among them, BCI iBCI represents the backscattering coefficient value of crop type i during growth stage j. i，j This represents the backscattering coefficient of crop type i during growth stage j. The backscattering coefficient value can be obtained using microwave data such as P-band, L-band, and C-band.

[0159] It should be noted that if the change coefficient of the backscattering coefficient is greater than or equal to a preset threshold for the change coefficient of the backscattering coefficient, then it is determined that the morphology of the crop corresponding to that pixel has changed. The threshold for the change coefficient of the backscattering coefficient can be set to different values ​​for different crops.

[0160] This embodiment can accurately determine the variation coefficient of crop backscattering coefficient in the target area, so that crop yield data can be accurately determined based on the target model combined with the variation coefficient of crop backscattering coefficient and other data. Based on the crop yield data, the crop loss in the target area can be accurately determined, thereby avoiding the deficiency of insufficient accuracy in disaster information acquisition and improving the accuracy of crop loss assessment.

[0161] Figure 2 This is a schematic diagram illustrating the process of determining crop loss in the crop low-temperature freezing disaster identification method based on the collaboration of meteorological and remote sensing data provided in this application embodiment. (Refer to...) Figure 2 Based on crop yield data, determine the amount of crop loss in the target area, including:

[0162] Step 301: Determine the pixel area of ​​the remote sensing image of the target area and the number of pixels of damaged crops in the remote sensing image;

[0163] Step 302: Based on crop yield data, pixel area and number of pixels, determine the amount of crop loss in the target area.

[0164] This application can obtain the pixel area comprised of all pixels in a remote sensing image of a target area, as well as the number of pixels in the remote sensing image showing crop damage. Specifically, it can obtain the number of pixels showing damage for each type of crop.

[0165] In addition, obtain the number of damage levels for each crop and the number of crop types.

[0166] Furthermore, this application can determine the crop loss in the target area based on the following expression and the data obtained above:

[0167]

[0168] Where TA represents crop loss, p represents the pixel area of ​​the remote sensing image, and Q... r CY represents the number of damaged pixels for crop type r. kThis represents the loss per unit area with a damage level of k, which is determined based on crop yield data and crop yield thresholds; m represents the number of damage levels; and v represents the number of crop types.

[0169] This embodiment can accurately determine the crop loss in a target area based on crop yield data, the pixel area of ​​the remote sensing image of the target area, and the number of damaged crop pixels in the remote sensing image. This can avoid the deficiency of insufficient accuracy in disaster information acquisition and thus improve the accuracy of crop loss assessment.

[0170] This application aims to provide technical support for the rapid identification of regional crops affected by low temperature and freezing disasters, serving agricultural meteorological disaster monitoring and early warning, and agricultural production management.

[0171] By combining regional crop types, growth stages, and meteorological conditions, early warning information and spatial extent of crop damage caused by freezing can be obtained through meteorological data. Remote sensing image representations of crop damage can be established, time-series growth curves of normally growing crops can be created, freezing disaster identification algorithms can be proposed, and random forest machine learning methods can be combined to estimate the degree of damage and loss of different types of crops in different regions, in order to achieve the goal of accurate identification and rapid loss assessment of freezing disasters.

[0172] Furthermore, this application also provides a crop low-temperature freezing disaster identification device that integrates meteorological and remote sensing data.

[0173] Reference Figure 3 , Figure 3 This is a schematic diagram of the functional modules of an embodiment of the crop low-temperature freezing disaster identification device that integrates meteorological and remote sensing data according to this application.

[0174] The crop low-temperature freezing disaster identification device that integrates meteorological and remote sensing data includes:

[0175] The first determining module 310 is used to determine the duration of crop freezing damage, the coefficient of change of air temperature, the coefficient of change of chlorophyll index, the coefficient of change of vegetation index and the coefficient of change of backscattering coefficient in the target area.

[0176] The transmission module 320 is used to input the duration, the air temperature variation coefficient, the chlorophyll index variation coefficient, the vegetation index variation coefficient, and the backscattering coefficient variation coefficient into the target model to obtain the crop yield data output by the target model; wherein, the target model is used for crop yield prediction.

[0177] The second determining module 330 is used to determine the amount of crop loss in the target area based on the crop yield data.

[0178] The crop low-temperature freezing disaster identification device that combines meteorological and remote sensing data provided in this application can accurately determine crop yield data based on the target model by determining the duration of crop freezing disaster, air temperature change coefficient, chlorophyll index change coefficient, vegetation index change coefficient, and backscattering coefficient change coefficient in the target area. Based on the crop yield data, it can accurately determine the crop loss in the target area, thereby avoiding the deficiency of insufficient accuracy in disaster information acquisition and improving the accuracy of crop loss assessment.

[0179] In one embodiment, the first determining module 310 is specifically used for:

[0180] Determine the air temperature of the region where a pixel is located in a remote sensing image of the target area;

[0181] Based on the air temperature and the air temperature threshold of the crop corresponding to the pixel, the variation coefficient of crop air temperature in the target area is determined.

[0182] In one embodiment, the first determining module 310 is further configured to:

[0183] Determine the chlorophyll index of the crop corresponding to the pixel in the remote sensing image of the target area during its current growth stage; wherein the chlorophyll index is determined based on the remote sensing reflectance of the green band and the remote sensing reflectance of the red band of the target area;

[0184] Based on the chlorophyll index and the chlorophyll index threshold of the crop corresponding to the pixel in the current growth stage, the change coefficient of crop chlorophyll index in the target area is determined.

[0185] In one embodiment, the first determining module 310 is further configured to:

[0186] Determine the vegetation index of the crop corresponding to a pixel in the remote sensing image of the target area during its current growth stage. Based on the vegetation index and a threshold value for the vegetation index of the crop corresponding to the pixel during its current growth stage, determine the variation coefficient of the crop vegetation index in the target area. The vegetation index is determined based on the remote sensing reflectance in the near-infrared band and the remote sensing reflectance in the red band of the target area.

[0187] The improved soil-atmosphere correction index (PAC) for crops in the current growth stage is determined in the remote sensing image of the target area. Based on the PAC and the PAC threshold for the crop in the current growth stage, the variation coefficient of the crop vegetation index in the target area is determined. The PAC is determined based on the remote sensing reflectance of the blue band, the near-infrared band, and the red band of the target area.

[0188] In one embodiment, the first determining module 310 is further configured to:

[0189] Determine the backscattering coefficient value of the crop corresponding to the pixel in the remote sensing image of the target area during the current growth stage;

[0190] Based on the backscattering coefficient value and the backscattering coefficient threshold of the crop corresponding to the pixel in the current growth period, the variation coefficient of the crop backscattering coefficient in the target area is determined.

[0191] In one embodiment, the second determining module 330 is used to:

[0192] Determine the pixel area of ​​the remote sensing image of the target area and the number of pixels of damaged crops in the remote sensing image;

[0193] Based on the crop yield data, the pixel area, and the number of pixels, the crop loss in the target area is determined.

[0194] Figure 4 An example is a schematic diagram of the physical structure of an electronic device, such as... Figure 4 As shown, the electronic device may include: a processor 410, a communication interface 420, a memory 430, and a communication bus 440, wherein the processor 410, the communication interface 420, and the memory 430 communicate with each other via the communication bus 440. The processor 410 can call a computer program in the memory 430 to execute the steps of a crop low-temperature freezing disaster identification method based on meteorological and remote sensing data, such as:

[0195] Determine the duration of crop freezing damage, the coefficient of change of air temperature, the coefficient of change of chlorophyll index, the coefficient of change of vegetation index, and the coefficient of change of backscattering coefficient in the target area;

[0196] The duration, the coefficient of change of air temperature, the coefficient of change of chlorophyll index, the coefficient of change of vegetation index, and the coefficient of change of backscattering coefficient are input into the target model to obtain the crop yield data output by the target model; wherein, the target model is used for crop yield prediction.

[0197] Based on the crop yield data, the amount of crop loss in the target area is determined.

[0198] Furthermore, the logical instructions in the aforementioned memory 430 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a 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 computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. 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.

[0199] On the other hand, embodiments of this application also provide a storage medium, which is a computer-readable storage medium storing a computer program. The computer program is used to cause a processor to execute the steps of the methods provided in the above embodiments, including, for example:

[0200] Determine the duration of crop freezing damage, the coefficient of change of air temperature, the coefficient of change of chlorophyll index, the coefficient of change of vegetation index, and the coefficient of change of backscattering coefficient in the target area;

[0201] The duration, the coefficient of change of air temperature, the coefficient of change of chlorophyll index, the coefficient of change of vegetation index, and the coefficient of change of backscattering coefficient are input into the target model to obtain the crop yield data output by the target model; wherein, the target model is used for crop yield prediction.

[0202] Based on the crop yield data, the amount of crop loss in the target area is determined.

[0203] The computer-readable storage medium can be any available medium or data storage device that the processor can access, including but not limited to magnetic storage (e.g., floppy disk, hard disk, magnetic tape, magneto-optical disk (MO)), optical storage (e.g., CD, DVD, BD, HVD), and semiconductor storage (e.g., ROM, EPROM, EEPROM, non-volatile memory (NAND FLASH), solid-state drive (SSD)).

[0204] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0205] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0206] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims

1. A method for identifying crop low-temperature freezing disasters using meteorological and remote sensing data in conjunction, characterized in that, include: The duration of crop freezing damage, the coefficient of change of air temperature, the coefficient of change of chlorophyll index, the coefficient of change of vegetation index, and the coefficient of change of backscattering coefficient in the target area are determined. The coefficient of change of air temperature is used to determine whether the crop has suffered freezing damage, the coefficient of change of chlorophyll index is used to determine whether the crop chlorophyll has changed, the coefficient of change of vegetation index is used to determine whether the crop coverage has changed, and the coefficient of change of backscattering coefficient is used to determine whether the crop plant morphology has changed. The duration, the coefficient of change of air temperature, the coefficient of change of chlorophyll index, the coefficient of change of vegetation index, and the coefficient of change of backscattering coefficient are input into the target model to obtain the crop yield data output by the target model; wherein, the target model is used to predict crop yield, and the crop yield data includes crop yield and crop damage degree, and the crop damage degree is divided into multiple levels; Based on the crop yield data, determining the crop loss in the target area includes: determining the pixel area of ​​the remote sensing image of the target area and the number of pixels of damaged crops in the remote sensing image; based on the crop yield data, the pixel area, and the number of pixels, determining the crop loss in the target area using the following expression: ； in, This represents the amount of crop loss, where p represents the pixel area of ​​the remote sensing image. This represents the number of damaged pixels for crop type r. The loss per unit area is represented by the damage level k, which is determined based on the crop yield data and the crop yield threshold; m represents the number of damage levels; and v represents the number of crop types.

2. The method for identifying crop low-temperature freezing disasters using meteorological and remote sensing data synergy according to claim 1, characterized in that, The determination of the coefficient of variation of crop air temperature in the target area includes: Determine the air temperature of the region where a pixel is located in a remote sensing image of the target area; Based on the air temperature and the air temperature threshold of the crop corresponding to the pixel, the variation coefficient of crop air temperature in the target area is determined.

3. The method for identifying crop low-temperature freezing disasters using meteorological and remote sensing data synergy according to claim 1, characterized in that, The determination of the variation coefficient of crop chlorophyll index in the target area includes: Determine the chlorophyll index of the crop corresponding to the pixel in the remote sensing image of the target area during its current growth stage; wherein the chlorophyll index is determined based on the remote sensing reflectance of the green band and the remote sensing reflectance of the red band of the target area; Based on the chlorophyll index and the chlorophyll index threshold of the crop corresponding to the pixel in the current growth stage, the change coefficient of crop chlorophyll index in the target area is determined.

4. The method for identifying crop low-temperature freezing disasters using meteorological and remote sensing data synergy according to claim 3, characterized in that, The chlorophyll index is determined based on the following formula: ； in, The remote sensing reflectance of the green band in the region corresponding to a pixel in the remote sensing image of the target area. This represents the remote sensing reflectance of the red band in the region corresponding to the pixel.

5. The method for identifying crop low-temperature freezing disasters using meteorological and remote sensing data synergy according to claim 1, characterized in that, The determination of the change coefficient of crop vegetation index in the target area includes: The vegetation index of the crop corresponding to a pixel in the remote sensing image of the target area during its current growth stage is determined. Based on the vegetation index and a threshold value for the vegetation index of the crop corresponding to the pixel during its current growth stage, a variation coefficient of the crop vegetation index in the target area is determined. The vegetation index is determined based on the remote sensing reflectance in the near-infrared band and the remote sensing reflectance in the red band of the target area. The improved soil-atmosphere correction index of the crop corresponding to a pixel in the remote sensing image of the target area during its current growth stage is determined. Based on the improved soil-atmosphere correction index and a threshold value for the improved soil-atmosphere correction index of the crop corresponding to the pixel during its current growth stage, a variation coefficient of the crop vegetation index in the target area is determined. The improved soil-atmosphere correction index is determined based on the remote sensing reflectance in the blue band, the remote sensing reflectance in the near-infrared band, and the remote sensing reflectance in the red band of the target area.

6. The method for identifying crop low-temperature freezing disasters using meteorological and remote sensing data synergy according to claim 5, characterized in that, The vegetation index is determined based on the following formula: ； in, Represents the near-infrared reflectance of the region corresponding to a pixel in a remotely sensed image of the target area. This represents the remote sensing reflectance of the red band in the region corresponding to the pixel; The improved soil atmospheric correction index is determined based on the following formula: ； Among them, R b This represents the remote sensing reflectance of the blue band in the region corresponding to the pixel. This represents the remote sensing reflectance in the near-infrared band of the region corresponding to the pixel. The values ​​represent the remote sensing reflectance of the red band in the region corresponding to the pixel; L=1; C1=6; C2=7.5; G=2.

5.

7. The method for identifying crop low-temperature freezing disasters using meteorological and remote sensing data synergy according to claim 1, characterized in that, The determination of the variation coefficient of crop backscattering coefficient in the target area includes: Determine the backscattering coefficient value of the crop corresponding to the pixel in the remote sensing image of the target area during the current growth stage; Based on the backscattering coefficient value and the backscattering coefficient threshold of the crop corresponding to the pixel in the current growth period, the variation coefficient of the crop backscattering coefficient in the target area is determined.

8. The method for identifying crop low-temperature freezing disasters using meteorological and remote sensing data synergy according to claim 1, characterized in that, Determining the crop loss in the target area based on the crop yield data includes: Determine the pixel area of ​​the remote sensing image of the target area and the number of pixels of damaged crops in the remote sensing image; Based on the crop yield data, the pixel area, and the number of pixels, the crop loss in the target area is determined.

9. The method for identifying crop low-temperature freezing disasters using meteorological and remote sensing data synergy according to claim 1, characterized in that, The expression for the target model is as follows: ； in, This represents an ensemble decision tree, where n represents the number of sub-decision trees. A sub-decision tree representing the original output CY given input variable C; ； Wherein, CY represents crop yield; C represents input variables, which include at least the duration of crop freezing damage, the coefficient of change of air temperature, the coefficient of change of chlorophyll index, the coefficient of change of vegetation index, and the coefficient of change of backscattering coefficient; ( () represents a nonlinear function that establishes the relationship between input variables and crop yield; This indicates the error value.

Images