Crop disease severity prediction method, system, storage medium and electronic device

Abstract

The embodiment of the present invention provides a method, system, storage medium and electronic device for predicting the severity of crop diseases. The method includes: analyzing crop information to determine the severity of diseases of multiple first crops in a preset crop planting area; based on the severity of the disease, selecting prediction factors from a factor library; determining the factor weights of each prediction factor based on the prediction factor and the severity of the disease; the factor weights characterize the contribution of the prediction factor to the crop disease; based on the factor weights, calculating the weighted Euclidean distance between the prediction factor of the first crop and the prediction factor of the crop to be predicted, and obtaining the disease severity of the crop to be predicted based on at least the disease severity of the first crop corresponding to the minimum weighted Euclidean distance. The present invention determines the factor weights of each prediction factor and predicts the severity of crop diseases using the contribution of the prediction factor to the crop disease, which is conducive to improving the accuracy of the prediction of the severity of crop diseases.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to a method, system, storage medium and electronic equipment for predicting the severity of crop diseases. Background Art

[0002] The severity of crop diseases indicates the proportion of diseased areas of crops, and it is of great significance to predict the severity of crop diseases. At present, the accuracy of crop disease severity prediction is not high, so it is necessary to improve the accuracy of crop disease severity prediction. Summary of the invention

[0003] The purpose of the embodiments of the present invention is to provide a method, system, storage medium and electronic device for predicting the severity of crop diseases, which can improve the accuracy of predicting the severity of crop diseases. The specific technical solution is as follows:

[0004] The present invention provides a method for predicting the severity of crop diseases, comprising:

[0005] Acquire a first crop information set, wherein the first crop information set includes crop information of a plurality of first crops in a preset crop planting area;

[0006] Analyzing the crop information to determine the severity of diseases of the plurality of first crops in the preset crop planting area;

[0007] Based on the severity of the disease, predictive factors are screened out from a factor library; wherein the factor library includes multiple meteorological factors and multiple remote sensing factors, the meteorological factors reflect the meteorological conditions of the crops during the growth period, and the remote sensing factors reflect the growth status of the crops;

[0008] Determining a factor weight of each of the prediction factors based on the prediction factors and the severity of the disease; the factor weight represents the contribution of the prediction factor to the crop disease;

[0009] Based on the factor weights, a weighted Euclidean distance between the prediction factor of the first crop and the prediction factor of the crop to be predicted is calculated, and the disease severity of the crop to be predicted is obtained based on at least the disease severity of the first crop corresponding to the minimum weighted Euclidean distance.

[0010] Optionally, based on the severity of the disease, predictive factors are screened out from a factor library, including:

[0011] Inputting the severity of the disease and the factors in the factor library into a regularized logistic regression model, screening the factors in the factor library to obtain a screening factor set;

[0012] The Fisher Score algorithm is used to screen the factors in the screening factor set to obtain prediction factors.

[0013] Optionally, determining the factor weight of each of the prediction factors based on the prediction factors and the disease severity comprises:

[0014] generating a predictor matrix based on the predictors;

[0015] The prediction factor matrix is ​​processed to remove collinearity by using a singular value decomposition method to obtain an orthogonal representation matrix of the prediction factors;

[0016] Using a Logistic regression model to fit the orthogonal representation matrix of the prediction factor and the severity of the disease, and obtain the standard regression coefficient of the prediction factor;

[0017] The factor weight of each of the predictive factors is determined based on the standardized regression coefficient.

[0018] Optionally, the Logistic regression model is used to fit the orthogonal representation matrix of the prediction factor and the severity of the disease to obtain the standard regression coefficient of the prediction factor, including:

[0019] Based on the formula determining standardized regression coefficients for the predictors;

[0020] in,

[0021] ω * =(Z T Z) -1 Z T X

[0022]

[0023] In the formula, ε *2 is the standard regression coefficient of the predictor, ω * is the linear regression coefficient, Z is the orthogonal matrix of the predictor, Z T is the transposed matrix of the prediction factor orthogonal representation matrix, X is the prediction factor matrix, is the standard regression coefficient of Logistic regression, b is the regression coefficient obtained by fitting the prediction factor and the severity of the disease using the Logistic regression model, s Z is the standard deviation of Z, R 0 is the square root of the regression coefficient obtained by fitting the prediction factor and the disease severity using a multivariate linear regression model, is the logistic regression predicted value logit(y 1 )’s standard deviation.

[0024] Optionally, determining the factor weight of each of the prediction factors based on the standard regression coefficient includes:

[0025] Based on the formula Determine the factor weights for the predictors;

[0026] In the formula, ω i is the factor weight of the i-th predictor, ε *2 i is the standardized regression coefficient of the ith predictor, and m is the total number of predictors.

[0027] Optionally, the calculating, based on the factor weights, a weighted Euclidean distance between the prediction factor of the first crop and the prediction factor of the crop to be predicted includes:

[0028] Based on the formula

[0029]

[0030] Calculate the weighted Euclidean distance between the predictor of the first crop and the predictor of the crop to be predicted;

[0031] Where, d sj is the weighted Euclidean distance, a si is the i-th predictor of the s-th first crop, b ji is the i-th prediction factor of the j-th crop to be predicted, ω i is the factor weight of the ith predictor, and m is the total number of predictors.

[0032] Optionally, analyzing the crop information to determine the severity of diseases of the plurality of first crops in the preset crop planting area includes:

[0033] Analyzing the crop information to obtain the number of diseased crops in the preset crop planting area and the total number of the first crops;

[0034] The ratio of the number of diseased crops to the total number of the first crops is used as the severity of the disease of the multiple first crops in the preset crop planting area.

[0035] The present invention also provides a crop disease severity prediction system, comprising:

[0036] A crop information acquisition module, used to acquire a first crop information set, wherein the first crop information set includes crop information of a plurality of first crops in a preset crop planting area;

[0037] An analysis module, used to analyze the crop information to determine the severity of diseases of the plurality of first crops in the preset crop planting area;

[0038] A factor screening module, used to screen out prediction factors from a factor library based on the severity of the disease; wherein the factor library includes multiple meteorological factors and multiple remote sensing factors, the meteorological factors reflect the meteorological conditions of the crop during the growth period, and the remote sensing factors reflect the growth status of the crop;

[0039] A factor weight determination module, used to determine the factor weight of each of the prediction factors based on the prediction factors and the disease severity; the factor weight represents the contribution of the prediction factor to the crop disease;

[0040] The prediction module is used to calculate the weighted Euclidean distance between the prediction factor of the first crop and the prediction factor of the crop to be predicted based on the factor weights, and obtain the disease severity of the crop to be predicted based on at least the disease severity of the first crop corresponding to the minimum weighted Euclidean distance.

[0041] The present invention also provides a computer-readable storage medium, on which a program is stored, and when the program is executed by a processor, the above-mentioned crop disease severity prediction method is implemented.

[0042] The present invention further provides an electronic device, characterized in that it comprises:

[0043] At least one processor, and at least one memory and bus connected to the processor;

[0044] The processor and the memory communicate with each other via the bus; the processor is used to call program instructions in the memory to execute the above-mentioned crop disease severity prediction method.

[0045] The embodiments of the present invention provide a crop disease severity prediction method, system, storage medium and electronic device. By determining the factor weights of the prediction factors, the severity of the crop diseases can be predicted using the contribution of the prediction factors to the crop diseases, which is beneficial to improving the accuracy of the crop disease severity prediction.

[0046] Of course, it is not necessary to achieve all of the advantages described above at the same time to implement any product or method of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0048] Figure 1A flow chart of a method for predicting the severity of crop diseases provided by an embodiment of the present invention;

[0049] Figure 2 A structural diagram of a crop disease severity prediction system provided by an embodiment of the present invention;

[0050] Figure 3 A schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0051] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0052] The present invention provides a method for predicting the severity of crop diseases. Figure 1 As shown, the method includes:

[0053] Step 101: Acquire a first crop information set, where the first crop information set includes crop information of a plurality of first crops in a preset crop planting area.

[0054] There are multiple first crops in the preset crop planting area, and the first crop information set contains crop information of each crop. The crop information may be information recording whether the crop is a diseased crop.

[0055] Optionally, the crop may be wheat, and its crop disease may be wheat fusarium head blight. When the present invention predicts the severity of wheat fusarium head blight, a wheat planting research area may be selected as a preset crop planting area, in which there are multiple plots, and each plot is planted with multiple wheat plants. When acquiring the first crop information set, crop information collection may be performed on wheat in N areas of the same size for each plot, and information on whether the wheat suffers from fusarium head blight may be recorded. The value of N may be determined according to the number of plots, and the size of the area selected in the plot may be determined according to the area of ​​the plot. For example, if the number of plots is 154, the value of N is 5, and the size of the area selected in the plot is 1*1m 2 .

[0056] Step 102: Analyze the crop information to determine the severity of diseases of multiple first crops in a preset crop planting area.

[0057] As an optional implementation, crop information is analyzed to determine the severity of diseases of multiple first crops in a preset crop planting area, including: analyzing crop information to obtain the number of diseased crops in the preset crop planting area and the total number of first crops; and using the ratio of the number of diseased crops to the total number of first crops as the severity of diseases of multiple first crops in the preset crop planting area.

[0058] The crop information can reflect the disease records of each crop in the preset crop planting area. By analyzing the crop information, the total number of crops with statistical disease records, that is, the total number of first crops, can be obtained, and the number of diseased crops and the number of non-disease crops can also be obtained. Since the severity of the disease is the ratio of the number of diseased crops to the total number of crops, the severity of the disease can be calculated for multiple first crops in the preset crop planting area, and the severity of the disease can also be calculated according to multiple plots in the preset crop planting area. Among them, when calculating the severity of the disease according to multiple plots in the preset crop planting area, the severity of the crop disease can be calculated for each plot, and then the disease severity calculated for all plots is averaged to obtain the severity of the disease of multiple first crops in the preset crop planting area.

[0059] Step 103: Based on the severity of the disease, predictive factors are screened out from a factor library; wherein the factor library includes multiple meteorological factors and multiple remote sensing factors, the meteorological factors reflect the meteorological conditions of the crops during the growing period, and the remote sensing factors reflect the growth status of the crops.

[0060] In the first crop information set, in addition to the record of whether the crop is a diseased crop, there can also be a record of the crop's growth cycle. Taking wheat as an example, since the flowering and heading periods of wheat are the peak periods for wheat ergot, the flowering and heading periods of each wheat plant can be recorded in the first crop information set.

[0061] The factor library is related to the growth cycle of crops. The factors in this factor library can reflect the meteorological conditions of crops during the growth period and the growth status of crops. The factor library includes multiple meteorological factors and multiple remote sensing factors. Among them, meteorological factors can be related to temperature, humidity, and rainfall during the crop growth period, and remote sensing factors can be related to leaf area index, plant nitrogen content, plant stress status, plant photosynthesis capacity, and chlorophyll content. Taking wheat as an example, meteorological factors can be the number of rainy days during the flowering period of wheat, the rainfall during the flowering period of wheat, the average temperature and humidity of wheat during the heading period, etc. Remote sensing factors can be enhanced vegetation index, normalized vegetation index, etc.

[0062] As an optional implementation, based on the severity of the disease, predictive factors are screened out from a factor library, including: inputting the severity of the disease and the factors in the factor library into a regularized logistic regression model, screening the factors in the factor library to obtain a screening factor set; and using a Fisher Score algorithm to screen the factors in the screening factor set to obtain predictive factors.

[0063] Optionally, the regularized Logistic regression model can be a Logistic+L1 regularized model, which is used to screen the factors in the factor library, and the regularized Logistic regression model is used to fit the factors in the factor library, and the factors corresponding to the coefficients of 0 after fitting are removed, and the factors with coefficients not 0 are retained to obtain the factors in the screening factor set. Since there are a large number of factors in the screening factor set, the factors in the screening factor set can be screened twice, and the Fisher Score algorithm can be used at this time. Specifically, the factors in the screening factor set and the severity of the disease can be used to obtain factors with high scores, and the top M factors with high scores are selected as prediction factors.

[0064] In this embodiment, two groups of prediction factors are obtained based on different time points, as shown in Table 1 and Table 2.

[0065] Table 1 The first group of predictors

[0066]

[0067]

[0068] Table 2 The second group of predictors

[0069]

[0070] The number of prediction factors in Table 1 is 8, of which 6 are meteorological factors and 2 are remote sensing factors. The number of prediction factors in Table 2 is 7, of which 5 are meteorological factors and 2 are remote sensing factors. RH is relative humidity.

[0071] Step 104: Determine the factor weight of each prediction factor based on the prediction factor and the disease severity; the factor weight represents the contribution of the prediction factor to the crop disease.

[0072] As an optional implementation, the factor weight of each prediction factor is determined based on the prediction factor and the disease severity, including:

[0073] Generate a predictor matrix based on the predictors;

[0074] The singular value decomposition method is used to remove the collinearity of the prediction factor matrix to obtain the prediction factor orthogonal representation matrix;

[0075] The Logistic regression model was used to fit the orthogonal matrix of the prediction factors and the severity of the disease to obtain the standard regression coefficients of the prediction factors.

[0076] The factor weights for each predictor were determined based on the standardized regression coefficients.

[0077] In the case of no collinearity among the prediction factors, the Logistic regression model fitted by the prediction factor orthogonal representation matrix formed by the prediction factors reflects the factor weight of the disease occurrence and development mechanism in disease prediction. The factor weight represents the contribution of the prediction factor to crop diseases and can be quantitatively expressed by standardizing the standard regression coefficient of each factor. Since there is usually collinearity between the prediction factors, the collinearity between the prediction factors can be removed by the singular value decomposition method. Assuming that the prediction factor matrix X composed of the prediction factors is I*J in size and the matrix is ​​a column full rank matrix, the prediction factor matrix X is subjected to singular value decomposition: X = PΔQ T , where P is XX T The characteristic matrix of X T The characteristic matrix of X, Δ is the matrix of X T The matrix composed of the square roots of the eigenvalues ​​of X, then the least squares standard orthogonal approximation of the prediction factor matrix X is Z = PQ T , Z is the orthogonal representation matrix of the prediction factors, and there is no collinearity between the characteristic components of Z. The Logistic regression model can be used to fit the standard regression coefficient of Z, and then the relationship between X and Z can be used to convert the standard regression coefficient of Z into the standard regression coefficient of the prediction factor matrix X.

[0078] Specifically, the Logistic regression model is used to fit the orthogonal representation matrix of the prediction factors and the severity of the disease to obtain the standard regression coefficients of the prediction factors, including:

[0079] Based on the formula Determine the standardized regression coefficients of the predictors;

[0080] in,

[0081] ω * =(Z T Z) -1 Z T X

[0082]

[0083] In the formula, ε *2 is the standard regression coefficient of the predictor, ω * is the linear regression coefficient, Z is the orthogonal matrix of the predictor, Z Tis the transposed matrix of the prediction factor orthogonal representation matrix, X is the prediction factor matrix, is the standard regression coefficient of Logistic regression, b is the regression coefficient obtained by fitting the prediction factor and disease severity using the Logistic regression model, s Z is the standard deviation of Z, R 0 is the square root of the regression coefficient obtained by fitting the predictors and disease severity using a multiple linear regression model, is the logistic regression predicted value logit(y 1 )’s standard deviation.

[0084] The factor weights of each predictor were determined based on the standardized regression coefficients, including:

[0085] Based on the formula Determine the factor weights for the predictors;

[0086] In the formula, ω i is the factor weight of the i-th predictor, ε *2 i is the standardized regression coefficient of the ith predictor, and m is the total number of predictors.

[0087] After calculation, the factor weights of the eight predictive factors corresponding to Table 1 are: the factor weight corresponding to rh70_mid_-11 is 0.15, the factor weight corresponding to rh70_-7 is 0.2, the factor weight corresponding to RH_+3 is 0.21, the factor weight corresponding to P_+11 is 0.11, the factor weight corresponding to t1530rh80_mid_-5 is 0.14, the factor weight corresponding to p_-3 is 0.11, the factor weight corresponding to rehbi is 0.07, and the factor weight corresponding to mcari is 0.01.

[0088] The factor weights of the seven predictors corresponding to Table 2 were: 0.16 for T1530RH70_+5, 0.26 for RH70_-7, 0.27 for T1530RH80MID_-7, 0.11 for P_-15, 0.11 for P-7, 0.07 for PSRI, and 0.02 for FVC.

[0089] Step 105: Based on the factor weights, a weighted Euclidean distance between the prediction factor of the first crop and the prediction factor of the crop to be predicted is calculated, and the disease severity of the crop to be predicted is obtained based on at least the disease severity of the first crop corresponding to the minimum weighted Euclidean distance.

[0090] As an optional implementation, based on the factor weights, calculating the weighted Euclidean distance between the prediction factor of the first crop and the prediction factor of the crop to be predicted includes:

[0091] Based on the formula

[0092]

[0093] Calculate the weighted Euclidean distance between the predictor of the first crop and the predictor of the crop to be predicted;

[0094] Where, d sj is the weighted Euclidean distance, a si is the i-th predictor of the s-th first crop, b ji is the i-th prediction factor of the j-th crop to be predicted, ω i is the factor weight of the ith predictor, and m is the total number of predictors.

[0095] If the Euclidean distance is calculated directly:

[0096]

[0097] By calculating the Euclidean distance between the prediction factor of the first crop and the prediction factor of the crop to be predicted, the impact of the disease occurrence and development mechanism on the severity of the disease is not considered when predicting crop diseases, which will bring large errors to the prediction results. Therefore, the present invention can predict the severity of crop diseases by using the contribution of the prediction factors to crop diseases by determining the factor weights of the prediction factors, which is beneficial to improving the accuracy of crop disease severity prediction.

[0098] After obtaining the weighted Euclidean distance, the distance can be sorted from small to large, and the disease severity of the first crop corresponding to the minimum weighted Euclidean distance can be used as the disease severity of the crop to be predicted. Of course, the average of the disease severities of the S first crops corresponding to the first S minimum weighted Euclidean distances can also be selected as the disease severity of the crop to be predicted.

[0099] The present invention also provides a crop disease severity prediction system, such as Figure 2 As shown, the system includes:

[0100] The crop information acquisition module 201 is used to acquire a first crop information set, where the first crop information set includes crop information of a plurality of first crops in a preset crop planting area.

[0101] The analysis module 202 is used to analyze the crop information and determine the severity of diseases of a plurality of first crops in a preset crop planting area.

[0102] The analysis module 202 is specifically used to: analyze the crop information to obtain the number of diseased crops and the total number of the first crop in the preset crop planting area;

[0103] The ratio of the number of diseased crops to the total number of the first crops is used as the disease severity of the multiple first crops in the preset crop planting area.

[0104] The factor screening module 203 is used to screen out prediction factors from the factor library based on the severity of the disease; wherein the factor library includes multiple meteorological factors and multiple remote sensing factors, the meteorological factors reflect the meteorological conditions of the crops during the growing period, and the remote sensing factors reflect the growth status of the crops.

[0105] The factor screening module 203 is specifically used to: input the disease severity and the factors in the factor library into the regularized Logistic regression model, screen the factors in the factor library to obtain a screening factor set; and use the Fisher Score algorithm to screen the factors in the screening factor set to obtain a prediction factor.

[0106] The factor weight determination module 204 is used to determine the factor weight of each prediction factor based on the prediction factor and the disease severity; the factor weight represents the contribution of the prediction factor to the crop disease.

[0107] The factor weight determination module 204 is specifically used to: generate a prediction factor matrix based on the prediction factors; use the singular value decomposition method to remove the collinearity of the prediction factor matrix to obtain the prediction factor orthogonal representation matrix; use the Logistic regression model to fit the prediction factor orthogonal representation matrix and the disease severity to obtain the standard regression coefficient of the prediction factor; determine the factor weight of each prediction factor based on the standard regression coefficient.

[0108] Optional, based on formula Determine the standardized regression coefficients of the predictors;

[0109] in,

[0110] ω * =(Z T Z) -1 Z T X

[0111]

[0112] In the formula, ε *2 is the standard regression coefficient of the predictor, ω * is the linear regression coefficient, Z is the orthogonal matrix of the predictor, Z T is the transposed matrix of the prediction factor orthogonal representation matrix, X is the prediction factor matrix, is the standard regression coefficient of Logistic regression, b is the regression coefficient obtained by fitting the prediction factor and disease severity using the Logistic regression model, s Z is the standard deviation of Z, R 0 is the square root of the regression coefficient obtained by fitting the predictors and disease severity using a multiple linear regression model, is the logistic regression predicted value logit(y 1 )’s standard deviation.

[0113] Based on the formula Determine the factor weights for the predictors;

[0114] In the formula, ω i is the factor weight of the i-th predictor, ε *2 i is the standardized regression coefficient of the ith predictor, and m is the total number of predictors.

[0115] The prediction module 205 is used to calculate the weighted Euclidean distance between the prediction factor of the first crop and the prediction factor of the crop to be predicted based on the factor weights, and obtain the disease severity of the crop to be predicted based on at least the disease severity of the first crop corresponding to the minimum weighted Euclidean distance.

[0116] Prediction module 205 is specifically used for: based on the formula

[0117]

[0118] Calculate the weighted Euclidean distance between the predictor of the first crop and the predictor of the crop to be predicted;

[0119] Where, d sj is the weighted Euclidean distance, a si is the i-th predictor of the s-th first crop, b ji is the i-th prediction factor of the j-th crop to be predicted, ω i is the factor weight of the ith predictor, and m is the total number of predictors.

[0120] An embodiment of the present invention provides a computer-readable storage medium having a program stored thereon, and when the program is executed by a processor, the above-mentioned crop disease severity prediction method is implemented.

[0121] An embodiment of the present invention provides an electronic device, such as Figure 3As shown, the electronic device 30 includes at least one processor 301, at least one memory 302 and a bus 303 connected to the processor 301; wherein the processor 301 and the memory 302 communicate with each other through the bus 303; the processor 301 is used to call the program instructions in the memory 302 to execute the above-mentioned crop disease severity prediction method. The electronic device in this article can be a server, a PC, a PAD, a mobile phone, etc.

[0122] The present application also provides a computer program product, which, when executed on a data processing device, is suitable for executing a program that initializes the steps included in the above-mentioned crop disease severity prediction method.

[0123] The present application is described with reference to the flowcharts and / or block diagrams of the methods, systems and computer program products according to the embodiments of the present application. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of the processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0124] In a typical configuration, the device includes one or more processors (CPU), memory and bus. The device may also include input / output interface, network interface and the like.

[0125] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM, and the memory includes at least one memory chip. The memory is an example of a computer-readable medium.

[0126] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0127] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0128] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. It should also be noted that the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0129] Each embodiment in this specification is described in a related manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0130] The above are only embodiments of the present application and are not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included within the scope of the claims of the present application.

Claims

1. A method for predicting the severity of crop diseases, It is characterized in that include: Acquire a first crop information set, wherein the first crop information set includes crop information of a plurality of first crops in a preset crop planting area; The first crop information set includes records of diseased crops and records of crop growth cycles; Analyzing the crop information to determine the severity of diseases of the plurality of first crops in the preset crop planting area; Based on the severity of the disease, predictive factors are screened out from a factor library; wherein the factor library includes multiple meteorological factors and multiple remote sensing factors, the meteorological factors reflect the meteorological conditions of the crops during the growth period, and the remote sensing factors reflect the growth status of the crops; Determining a factor weight of each of the prediction factors based on the prediction factors and the severity of the disease; the factor weight represents the contribution of the prediction factor to the crop disease; Based on the factor weights, a weighted Euclidean distance between the prediction factor of the first crop and the prediction factor of the crop to be predicted is calculated, and the disease severity of the crop to be predicted is obtained based on at least the disease severity of the first crop corresponding to the minimum weighted Euclidean distance; Based on the severity of the disease, predictive factors are screened from a factor library, including: Inputting the severity of the disease and the factors in the factor library into a regularized logistic regression model, screening the factors in the factor library to obtain a screening factor set; The Fisher Score algorithm is used to screen the factors in the screening factor set to obtain prediction factors; The step of determining the factor weight of each of the prediction factors based on the prediction factors and the disease severity comprises: generating a predictor matrix based on the predictors; The prediction factor matrix is ​​processed to remove collinearity by using a singular value decomposition method to obtain an orthogonal representation matrix of the prediction factors; Using a Logistic regression model to fit the orthogonal representation matrix of the prediction factor and the severity of the disease, and obtain the standard regression coefficient of the prediction factor; Determining the factor weight of each of the prediction factors based on the standardized regression coefficient; The Logistic regression model is used to fit the orthogonal representation matrix of the prediction factor and the severity of the disease to obtain the standard regression coefficient of the prediction factor, including: Based on the formula determining standardized regression coefficients for the predictors; in, In the formula, is the standardized regression coefficient of the predictor, is the linear regression coefficient, is the orthogonal representation matrix of the predictor, is the transposed matrix of the predictor orthogonal representation matrix, is the predictor matrix, is the standard regression coefficient of Logistic regression, is the regression coefficient obtained by fitting the prediction factor and the severity of the disease using the Logistic regression model, for The standard deviation of is the square root of the regression coefficient obtained by fitting the prediction factor and the disease severity using a multivariate linear regression model, Logistic regression predicted value The standard deviation of The step of calculating the weighted Euclidean distance between the prediction factor of the first crop and the prediction factor of the crop to be predicted based on the factor weights includes: Based on the formula Calculate the weighted Euclidean distance between the predictor of the first crop and the predictor of the crop to be predicted; In the formula, is the weighted Euclidean distance, is the i-th predictor of the s-th first crop, is the i-th prediction factor of the j-th crop to be predicted, is the factor weight of the i-th predictor, and m is the total number of predictors; The step of analyzing the crop information to determine the severity of diseases of the plurality of first crops in the preset crop planting area includes: Analyzing the crop information to obtain the number of diseased crops in the preset crop planting area and the total number of the first crops; Using the ratio of the number of diseased crops to the total number of the first crops as the severity of the disease of the plurality of first crops in the preset crop planting area; When calculating the severity of disease for multiple plots within the preset crop planting area, the severity of crop disease can be calculated for each plot, and then the calculation results of all plots are averaged to obtain the severity of disease for multiple first crops in the preset crop planting area.

2. The method for predicting the severity of crop diseases according to claim 1, It is characterized in that The step of determining the factor weight of each of the prediction factors based on the standard regression coefficient comprises: Based on the formula Determine the factor weights for the predictors; In the formula, is the factor weight of the i-th predictor, is the standardized regression coefficient of the ith predictor, and m is the total number of predictors.

3. A crop disease severity prediction system, It is characterized in that include: A crop information acquisition module, used to acquire a first crop information set, wherein the first crop information set includes crop information of a plurality of first crops in a preset crop planting area; The first crop information set includes records of diseased crops and records of crop growth cycles; An analysis module, used to analyze the crop information to determine the severity of diseases of the plurality of first crops in the preset crop planting area; A factor screening module, used to screen out prediction factors from a factor library based on the severity of the disease; wherein the factor library includes multiple meteorological factors and multiple remote sensing factors, the meteorological factors reflect the meteorological conditions of the crop during the growth period, and the remote sensing factors reflect the growth status of the crop; A factor weight determination module, used to determine the factor weight of each of the prediction factors based on the prediction factors and the disease severity; the factor weight represents the contribution of the prediction factor to the crop disease; A prediction module, configured to calculate a weighted Euclidean distance between a prediction factor of the first crop and a prediction factor of the crop to be predicted based on the factor weights, and obtain a disease severity of the crop to be predicted based at least on a disease severity of the first crop corresponding to a minimum weighted Euclidean distance; Based on the severity of the disease, predictive factors are screened from a factor library, including: Inputting the severity of the disease and the factors in the factor library into a regularized logistic regression model, screening the factors in the factor library to obtain a screening factor set; The Fisher Score algorithm is used to screen the factors in the screening factor set to obtain prediction factors; The step of determining the factor weight of each of the prediction factors based on the prediction factors and the disease severity comprises: generating a predictor matrix based on the predictors; The prediction factor matrix is ​​processed to remove collinearity by using a singular value decomposition method to obtain an orthogonal representation matrix of the prediction factors; Using a Logistic regression model to fit the orthogonal representation matrix of the prediction factor and the severity of the disease, and obtain the standard regression coefficient of the prediction factor; Determining the factor weight of each of the prediction factors based on the standardized regression coefficient; The Logistic regression model is used to fit the orthogonal representation matrix of the prediction factor and the severity of the disease to obtain the standard regression coefficient of the prediction factor, including: Based on the formula determining standardized regression coefficients for the predictors; in, In the formula, is the standardized regression coefficient of the predictor, is the linear regression coefficient, is the orthogonal representation matrix of the predictor, is the transposed matrix of the predictor orthogonal representation matrix, is the predictor matrix, is the standard regression coefficient of Logistic regression, is the regression coefficient obtained by fitting the prediction factor and the severity of the disease using the Logistic regression model, for The standard deviation of is the square root of the regression coefficient obtained by fitting the prediction factor and the disease severity using a multivariate linear regression model, Logistic regression predicted value The standard deviation of The step of calculating the weighted Euclidean distance between the prediction factor of the first crop and the prediction factor of the crop to be predicted based on the factor weights includes: Based on the formula Calculate the weighted Euclidean distance between the predictor of the first crop and the predictor of the crop to be predicted; In the formula, is the weighted Euclidean distance, is the i-th predictor of the s-th first crop, is the i-th prediction factor of the j-th crop to be predicted, is the factor weight of the i-th predictor, and m is the total number of predictors; The step of analyzing the crop information to determine the severity of diseases of the plurality of first crops in the preset crop planting area includes: Analyzing the crop information to obtain the number of diseased crops in the preset crop planting area and the total number of the first crops; Using the ratio of the number of diseased crops to the total number of the first crops as the severity of the disease of the plurality of first crops in the preset crop planting area; When calculating the severity of disease for multiple plots within the preset crop planting area, the severity of crop disease can be calculated for each plot, and then the calculation results of all plots are averaged to obtain the severity of disease for multiple first crops in the preset crop planting area.

4. A computer-readable storage medium, It is characterized in that The computer-readable storage medium stores a program, and when the program is executed by the processor, the method for predicting the severity of crop diseases according to claim 1 or 2 is implemented.

5. An electronic device, It is characterized in that include: At least one processor, and at least one memory and bus connected to the processor; The processor and the memory communicate with each other via the bus; The processor is used to call the program instructions in the memory to execute the crop disease severity prediction method according to claim 1 or 2.

Images