Winter wheat drought early warning method and system based on multi-source data, electronic equipment and storage medium

Abstract

The invention belongs to the field of crop disaster early warning, and discloses a winter wheat drought early warning method and system based on multi-source data, electronic equipment and a storage medium, and the method comprises the steps: obtaining a satellite remote sensing vegetation index, meteorological parameters and soil moisture data, and carrying out the time sequence synchronization processing of the obtained data through a preset time window, obtaining a multi-source data set with unified time resolution; performing dimension reduction processing on the multi-source data set by using a principal component analysis method to obtain a feature vector set after dimension reduction; performing growth stage division on the feature vector set through a preset winter wheat growth stage division rule to obtain a staged feature sequence; modeling the water demand characteristic change trend of the characteristic sequence of each growth stage by using a long-short term memory network to obtain a prediction parameter set reflecting a dynamic water demand relationship; and extracting a water demand change trend from the prediction parameter set, extracting drought characteristics from the water demand change trend, carrying out drought grading, and sending out a drought early warning according to a drought grading result.

Description

Technical Field

[0001] The present invention relates to the technical field of crop disaster early warning, and in particular to a winter wheat drought early warning method, system, electronic equipment and storage medium based on multi-source data. Background Art

[0002] As one of the most threatening natural disasters in agricultural production, drought poses a severe challenge to food security, especially for winter wheat, an important food crop, whose yield and quality are directly related to the stability of global food supply. As climate change intensifies, the frequency and intensity of droughts continue to increase, and the study of effective drought early warning methods has become an urgent need to ensure the sustainable development of agriculture. Winter wheat has a long growth cycle and is sensitive to water. Timely and accurate drought early warning can not only reduce economic losses, but also provide support for drought resistance decisions. However, current research and practice still face many difficulties in this regard.

[0003] Existing drought warning methods mostly rely on a single data source, such as precipitation records from weather stations or satellite remote sensing vegetation indices, which are difficult to fully reflect the actual drought conditions of winter wheat. The limitations of a single data set lead to insufficient warning accuracy, especially in complex terrain and regional climate differences, where misjudgments and missed judgments occur frequently. In addition, traditional methods often lack dynamic analysis of the water demand characteristics of wheat during its growth stage, and the warning results are out of touch with actual agricultural needs, limiting their practicality. These defects make it difficult for existing solutions to meet the requirements of modern agriculture for precision and intelligence.

[0004] The core challenge in the field of winter wheat drought early warning is how to integrate multi-source data and achieve accurate analysis. First, the heterogeneity of multi-source data such as remote sensing vegetation index, meteorological parameters and soil moisture data makes fusion analysis face technical barriers. Secondly, the response of wheat to water varies significantly at different growth stages, and it is difficult for existing models to accurately describe this dynamic relationship, resulting in insufficient timeliness of early warning. Finally, how to convert multidimensional data into a unified drought assessment indicator and maintain the objectivity of the assessment in a complex environment is a difficult problem that needs to be broken through. These technical factors have not been resolved, which directly leads to the difficulty of the early warning system to adapt to regional differences and crop characteristics, affecting the scientific nature of drought resistance decisions.

[0005] Therefore, how to build a comprehensive assessment model that can dynamically reflect the water requirement characteristics and drought response of winter wheat by integrating multi-source data such as satellite remote sensing, meteorological observations and soil moisture, and realize standardized graded judgment under a multi-dimensional indicator system has become a key issue in improving the accuracy of winter wheat drought warning. Summary of the invention

[0006] In order to solve the above technical problems, the present invention provides a winter wheat drought early warning method based on multi-source data, the method comprising:

[0007] Obtain satellite remote sensing vegetation index, meteorological parameters and soil moisture data, and perform time series synchronization processing on the obtained data through a preset time window to obtain a multi-source dataset with a unified time resolution;

[0008] Use the principal component analysis method to perform dimensionality reduction processing on the multi-source dataset to obtain a set of feature vectors after dimensionality reduction;

[0009] Through the preset winter wheat growth stage division rules, divide the set of feature vectors to obtain a staged feature sequence;

[0010] Use the long short-term memory network to model the changing trend of water demand characteristics of the feature sequences in each growth stage to obtain a set of prediction parameters reflecting the dynamic water demand relationship;

[0011] Extract the water demand change trend from the set of prediction parameters, extract drought characteristics from the water demand change trend and perform drought grading, and issue a drought warning according to the result of drought grading.

[0012] Preferably, the method for obtaining a multi-source dataset with a unified time resolution includes:

[0013] Obtain satellite remote sensing data, extract vegetation index, meteorological parameters and soil moisture data from the satellite remote sensing data to obtain an initial multi-source data set;

[0014] Divide the initial multi-source data set through a preset time window to obtain data subsets within a time period;

[0015] For the data subsets within the time period, use the time series alignment method to align the vegetation index, meteorological parameters and soil moisture data therein to obtain a synchronized data sequence;

[0016] Calculate the unified time resolution and perform time resolution alignment on the synchronized data sequence to obtain the multi-source dataset.

[0017] Preferably, the method for obtaining a set of feature vectors after dimensionality reduction includes:

[0018] Use the principal component analysis method to perform dimensionality reduction processing on the satellite remote sensing, meteorological data and soil moisture in the multi-source dataset to obtain a preliminary set of feature vectors;

[0019] Extract the feature vectors with the principal component contribution rate greater than the preset value from the preliminary set of feature vectors to obtain a refined set of feature vectors;

[0020] For the refined feature vector group, calculate the correlation of each feature vector in the group using the covariance matrix, and judge the independence degree between each feature vector. If the correlation between features exceeds the preset threshold, orthogonalize the feature vector group through linear transformation to obtain an orthogonal feature set;

[0021] Construct a data set after dimensionality reduction according to the orthogonal feature set, obtain the low-dimensional representation form of heterogeneous features, and obtain the feature vector set after dimensionality reduction.

[0022] Preferably, the method for obtaining the staged feature sequence includes:

[0023] Judge the feature distribution corresponding to the growth stage of winter wheat through the preset rules for dividing the growth stages of winter wheat to obtain a preliminary stage division result;

[0024] Use the K-means clustering algorithm to cluster the preliminary stage division result to determine the boundary of the staged features and obtain an optimized staged feature sequence;

[0025] Through the optimized staged feature sequence, obtain the complete feature distribution of the growth stage of winter wheat to obtain the staged feature sequence.

[0026] Preferably, the method for obtaining the prediction parameter set includes:

[0027] Use a long short-term memory network to model the feature sequence to obtain the water requirement characteristics of each growth stage;

[0028] Determine the trend analysis result by analyzing the change trend of the water requirement characteristics to obtain the updated water requirement characteristics;

[0029] Generate a dynamic relationship description according to the updated water requirement characteristics, and obtain prediction parameters through the dynamic relationship description to obtain the prediction parameter set.

[0030] Preferably, the method for issuing a drought warning according to the result of drought grading includes:

[0031] Extract the water requirement change data from the prediction parameter set, and use statistical methods to calculate the change trend to obtain the water requirement change trend;

[0032] Obtain drought-related features from the water requirement change trend, and grade the drought severity based on the drought-related features using a logistic regression algorithm to obtain a preliminary grading result;

[0033] For the preliminary grading result, obtain the severity quantification value and perform standardization processing through a preset threshold to obtain a standardized evaluation grade;

[0034] According to the standardized evaluation level, determine whether the severity of the drought reaches the warning condition. If it does, output an alarm signal through the warning mechanism.

[0035] The present invention also provides a winter wheat drought warning system based on multi-source data. The system is used to implement the method described in any one of the above, and includes: a multi-source data acquisition module, a feature vector acquisition module, a feature sequence acquisition module, a prediction parameter set calculation module, and a drought warning module.

[0036] The multi-source data acquisition module is used to acquire satellite remote sensing vegetation indices, meteorological parameters, and soil moisture data, and perform time series synchronization processing on the obtained data through a preset time window to obtain a multi-source data set with a unified time resolution.

[0037] The feature vector acquisition module uses the principal component analysis method to perform dimensionality reduction processing on the multi-source data set to obtain a set of feature vectors after dimensionality reduction.

[0038] The feature sequence acquisition module divides the growth stages of the set of feature vectors according to the preset winter wheat growth stage division rules to obtain a staged feature sequence.

[0039] The prediction parameter set calculation module uses a long short-term memory network to model the changing trend of the water demand characteristics of the feature sequences in each growth stage to obtain a prediction parameter set reflecting the dynamic water demand relationship.

[0040] The drought warning module is used to extract the changing trend of water demand from the prediction parameter set, extract drought characteristics from the changing trend of water demand and perform drought grading, and issue a drought warning according to the result of the drought grading.

[0041] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the program, it implements the winter wheat drought warning method based on multi-source data described in any one of the above.

[0042] The present invention also provides a computer-readable storage medium. The computer-readable storage medium stores a computer program, and when the computer program is executed, it implements the winter wheat drought warning method based on multi-source data described in any one of the above.

[0043] Compared with the prior art, the beneficial effects of the present invention are:

[0044] The present invention obtains satellite remote sensing vegetation index, meteorological parameters and soil moisture data, performs time series synchronization processing and dimensionality reduction to obtain a set of feature vectors with a unified time resolution; then, according to the winter wheat growth stage division rules, the data is segmented into stage feature sequences such as jointing stage, heading stage, filling stage, etc.; a long short-term memory network is used to model the changing trend of water demand characteristics in each growth stage to obtain a set of prediction parameters reflecting the dynamic water demand relationship; finally, the severity of drought is classified and judged through a logistic regression algorithm, and a standardized drought assessment level is obtained and a warning is issued. The present invention realizes the precise monitoring and timely warning of the drought situation during the growth process of winter wheat, and provides strong support for agricultural production management and disaster prevention and control. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] In order to more clearly illustrate the technical solutions of the present invention, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0046] Figure 1 It is a schematic flowchart of the method according to an embodiment of the present invention;

[0047] Figure 2 It is a schematic structural diagram of an electronic device according to an embodiment of the present invention.

[0048] Description of the reference numerals:

[0049] 1010, processor; 1020, memory; 1030, input / output interface; 1040, communication interface; 1050, bus. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0050] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0051] It should be noted that, unless otherwise defined, the technical terms or scientific terms used in the embodiments of the present disclosure should have the ordinary meanings understood by those of ordinary skill in the field to which the present disclosure belongs. The "first", "second" and similar terms used in the embodiments of the present disclosure do not denote any order, quantity or importance, but are only used to distinguish different components. Words such as "including" or "comprising" mean that the elements or objects appearing before the word cover the elements or objects listed after the word and their equivalents, without excluding other elements or objects. Words such as "connected" or "coupled" are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "Up", "down", "left", "right", etc. are only used to indicate relative position relationships, and when the absolute position of the object being described changes, the relative position relationship may also change accordingly.

[0052] Embodiment 1

[0053] In this embodiment, as Figure 1 shown, a winter wheat drought warning method based on multi-source data, the method includes:

[0054] S1. Obtain satellite remote sensing vegetation indices, meteorological parameters, and soil moisture data, and perform time series synchronization processing on the obtained data through a preset time window to obtain a multi-source data set with a unified time resolution.

[0055] The method for obtaining a multi-source data set with a unified time resolution includes: obtaining satellite remote sensing data, extracting vegetation indices, meteorological parameters, and soil moisture data from the satellite remote sensing data to obtain an initial multi-source data set; dividing the initial multi-source data set through a preset time window to obtain data subsets within a time period; for the data subsets within the time period, use a time series method to align the vegetation indices, meteorological parameters, and soil moisture data therein to obtain a synchronized data sequence; calculate the unified time resolution, and perform time resolution alignment on the synchronized data sequence to obtain a multi-source data set.

[0056] In this embodiment, Landsat or MODIS satellites are used to obtain data in bands such as red light and near-infrared light covering a certain area, and the normalized difference vegetation index (NDVI) is extracted from them to reflect the vegetation growth status through the reflectance difference between red light and near-infrared light; meteorological parameters such as rainfall and temperature can be obtained from the meteorological sensors carried by the satellites; soil moisture is estimated through microwave remote sensing data. After obtaining the initial multi-source data set, the multi-source data set is divided by a preset time window. With a 30-day window, the annual data is divided into 12 subsets. For the data subsets within the time period, since the time resolutions of different data sources may be different, a time series synchronization method is used to align the data, and at the same time, a unified time resolution is calculated, and the time resolution of the synchronized data sequence is aligned to obtain a multi-source data set.

[0057] S2. Use the principal component analysis method to perform dimensionality reduction processing on the multi-source data set to obtain a set of feature vectors after dimensionality reduction.

[0058] The method for obtaining the set of feature vectors after dimensionality reduction includes: using the principal component analysis method to perform dimensionality reduction processing on satellite remote sensing, meteorological data, and soil moisture in the multi-source data set to obtain a preliminary set of feature vectors; extracting feature vectors with a principal component contribution rate greater than a preset value from the preliminary set of feature vectors to obtain a refined set of feature vectors; for the refined set of feature vectors, calculate the correlation between each feature vector in the group using the covariance matrix to judge the degree of independence between each feature vector. If the correlation between features exceeds a preset threshold, the feature vector group is orthonormalized through linear transformation to obtain an orthonormal feature set; a data set after dimensionality reduction is constructed according to the orthonormal feature set to obtain a low-dimensional representation form of heterogeneous features, and a set of feature vectors after dimensionality reduction is obtained.

[0059] In this embodiment, before performing principal component analysis, it is first necessary to preprocess the multi-source dataset. This includes standardizing the data to eliminate the dimension and dimension differences between different variables. After that, the principal component analysis method is used to perform preliminary feature extraction on the preprocessed multi-source dataset: calculate the covariance matrix of the dataset to understand the correlation between each variable; by solving the eigenvalues and eigenvectors of the covariance matrix, determine the direction of the principal components, and according to the size of the eigenvalues, select the first k principal components whose cumulative contribution rate reaches a certain threshold (such as 95%), and these principal components are the preliminary feature vector set. Screen out the feature vectors with a principal component contribution rate greater than a preset value (such as 1%) from the preliminary feature vector set to obtain a refined feature vector group. This step can further remove redundant information and retain the features with strong ability to explain data variation. For the refined feature vector group, calculate the correlation between each feature vector in the group using the covariance matrix to judge the independence degree between each feature vector. If the correlation between features exceeds the preset threshold (such as 0.9), then perform orthogonalization processing on the feature vector group through linear transformation to obtain an orthogonal feature set. This step can ensure the independence of feature vectors and avoid problems caused by feature correlation in subsequent analysis. Construct a reduced-dimensional data set according to the orthogonal feature set, obtain a low-dimensional representation form of heterogeneous features, and obtain a reduced-dimensional feature vector set.

[0060] S3. Divide the growth stages of the feature vector set through a preset winter wheat growth stage division rule to obtain a staged feature sequence.

[0061] The method for obtaining the staged feature sequence includes: judging the feature distribution corresponding to the winter wheat growth stage through a preset winter wheat growth stage division rule to obtain a preliminary stage division result; using the K-means clustering algorithm to cluster the preliminary stage division result to determine the boundary of the staged features and obtain an optimized staged feature sequence; through the optimized staged feature sequence, obtain the complete feature distribution of the winter wheat growth stage to obtain the staged feature sequence.

[0062] In this embodiment, the preliminary stage division results include: if the feature distribution in the preliminary stage division results conforms to the jointing stage features, the data is segmented through preset rules to obtain the stage sequence of the jointing stage; if the feature distribution in the preliminary stage division results conforms to the heading stage features, the data is segmented through preset rules to obtain the stage sequence of the heading stage; if the feature distribution in the preliminary stage division results conforms to the filling stage features, the data is segmented through preset rules to obtain the stage sequence of the filling stage. When using the K-means clustering algorithm to process each stage sequence, the number of clusters can be set to 3, corresponding to the typical features of the jointing stage, heading stage, and filling stage respectively. For example, the data points in the jointing stage may be clustered into groups with high near-infrared values and low temperatures; the heading stage is concentrated in areas with high vegetation indices and moderate rainfall; the filling stage is characterized by a decrease in spectral values and a reduction in humidity. The method for determining the boundaries of the stage features includes: assuming that the clustering center of the data points in the jointing stage is a near-infrared value of 0.8 and a humidity of 22%, the stage boundaries can be optimized accordingly to obtain the optimized stage feature sequence. Finally, when obtaining the complete feature distribution of the winter wheat growth stage through the optimized stage feature sequence, the clustering results can be combined with the time axis to obtain the final stage feature sequence.

[0063] S4. Use a long short-term memory network to model the changing trend of the water requirement characteristics of the feature sequences in each growth stage to obtain a set of prediction parameters reflecting the dynamic water requirement relationship.

[0064] The methods for obtaining the set of prediction parameters include: using a long short-term memory network to model the feature sequences to obtain the water requirement characteristics of each growth stage; analyzing the changing trend of the water requirement characteristics to determine the trend analysis result and obtain the updated water requirement characteristics; generating a dynamic relationship description based on the updated water requirement characteristics, and obtaining the prediction parameters through the dynamic relationship description to obtain the set of prediction parameters.

[0065] In this embodiment, a long short-term memory network is used to model the feature sequences to extract the water requirement characteristics of each growth stage. Specifically, the humidity data of the previous 7 days and the current water requirement are input into the long short-term memory network. The network outputs an average daily water requirement of 2.5 mm in the jointing stage by memorizing historical information, while it may increase to 3.2 mm in the heading stage. The changing trend is analyzed through the water requirement characteristics to determine the trend analysis result. If the trend analysis result exceeds the preset threshold, it indicates that the model may not fully consider rainfall or irrigation factors. At this time, the modeling process is adjusted, such as adding rainfall as an input variable. The updated water requirement characteristics may show that the water requirement in the heading stage is stable at 3.8 mm. Based on the updated water requirement characteristics, a dynamic relationship description is generated, and the prediction parameters are obtained through the dynamic relationship description to obtain the parameter set.

[0066] S5. Extract the water demand change trend from the prediction parameter set, extract drought characteristics from the water demand change trend, classify the drought level, and issue a drought warning according to the result of the drought level classification.

[0067] The method for issuing a drought warning according to the result of the drought level classification includes: extracting water demand change data from the prediction parameter set, calculating the change trend by using a statistical method to obtain the water demand change trend; obtaining drought-related characteristics from the water demand change trend, and classifying the drought severity by using a logistic regression algorithm based on the drought-related characteristics to obtain a preliminary classification result; for the preliminary classification result, obtaining a severity quantification value, performing standardization processing through a preset threshold to obtain a standardized evaluation level; according to the standardized evaluation level, determining whether the drought severity reaches the warning condition, and if so, outputting an alarm signal through the warning mechanism.

[0068] It should be noted that the method of the embodiment of the present disclosure can be executed by a single device, such as a computer or a server. The method of this embodiment can also be applied to a distributed scenario and completed by multiple devices cooperating with each other. In this case of a distributed scenario, one of the multiple devices can only execute one or more steps of the method of the embodiment of the present disclosure, and these multiple devices will interact with each other to complete the described method.

[0069] It should be noted that some embodiments of the present disclosure have been described above. Other embodiments are within the scope of the appended claims. In some cases, it should be understood that the size of the sequence numbers of the steps in the above embodiments does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation to the implementation process of the embodiments of the present invention. The actions or steps recorded in the claims can be executed in an order different from that in the above embodiments and still achieve the desired results. In addition, the processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0070] Embodiment 2

[0071] In this embodiment, a winter wheat drought warning system based on multi-source data includes: a multi-source data acquisition module, a feature vector acquisition module, a feature sequence acquisition module, a prediction parameter set calculation module, and a drought warning module.

[0072] The multi-source data acquisition module is used to acquire satellite remote sensing vegetation index, meteorological parameters, and soil moisture data, and perform time series synchronization processing on the obtained data through a preset time window to obtain a multi-source data set with a unified time resolution.

[0073] The feature vector acquisition module uses the principal component analysis method to perform dimensionality reduction processing on the multi-source data set, and obtains a set of feature vectors after dimensionality reduction.

[0074] The feature sequence acquisition module divides the growth stages of the set of feature vectors according to the preset rules for dividing the growth stages of winter wheat, and obtains a staged feature sequence.

[0075] The prediction parameter set calculation module uses a long short-term memory network to model the changing trend of the water requirement characteristics of the feature sequences in each growth stage, and obtains a prediction parameter set reflecting the dynamic water requirement relationship.

[0076] The drought warning module is used to extract the changing trend of water requirement from the prediction parameter set, extract drought characteristics from the changing trend of water requirement and perform drought grading, and issue a drought warning according to the results of the drought grading.

[0077] The system of the above embodiment is used to implement the corresponding winter wheat drought warning method based on multi-source data in any of the foregoing embodiments, and has the beneficial effects of the corresponding method embodiments, which will not be elaborated here.

[0078] It should be noted that the above winter wheat drought warning system based on multi-source data is embodied in the form of functional units. The term "module" here can be implemented in the form of software and / or hardware, and no specific limitation is made thereto.

[0079] For example, the "module" can be a software program, a hardware circuit or a combination of the two that implements the above functions. The hardware circuit may include an application specific integrated circuit (ASIC), an electronic circuit, a processor (such as a shared processor, a dedicated processor or a group of processors, etc.) for executing one or more software or firmware programs, a memory, a merged logic circuit and / or other suitable components that support the described functions.

[0080] Embodiment III

[0081] Based on the same inventive concept, corresponding to the method of any of the above embodiments, the present invention also provides an electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor, and when the processor executes the program, it implements the winter wheat drought warning method based on multi-source data described in any of the above embodiments.

[0082] Figure 2FIG. 0 shows a more specific schematic diagram of the hardware structure of the electronic device provided in this embodiment. The device may include: a processor 1010, a memory 1020, an input / output interface 1030, a communication interface 1040, and a bus 1050. Among them, the processor 1010, the memory 1020, the input / output interface 1030, and the communication interface 1040 are communicatively connected to each other inside the device through the bus 1050.

[0083] The processor 1010 may be implemented in the form of a general-purpose CPU (Central Processing Unit), a microprocessor, an application-specific integrated circuit (ASIC), or one or more integrated circuits, etc., and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this specification.

[0084] The memory 1020 may be implemented in the form of a ROM (Read Only Memory), a RAM (Random Access Memory), a static storage device, a dynamic storage device, etc. The memory 1020 may store an operating system and other application programs. When implementing the technical solutions provided in the embodiments of this specification through software or firmware, the relevant program codes are stored in the memory 1020 and are called and executed by the processor 1010.

[0085] The input / output interface 1030 is used to connect to an input / output module to implement information input and output. The input / output module may be configured as a component in the device (not shown in the figure) or externally connected to the device to provide corresponding functions. Among them, the input device may include a keyboard, a mouse, a touch screen, a microphone, various sensors, etc., and the output device may include a display, a speaker, a vibrator, an indicator light, etc.

[0086] The communication interface 1040 is used to connect to a communication module (not shown in the figure) to implement communication interaction between this device and other devices. Among them, the communication module may implement communication in a wired manner (such as USB (Universal Serial Bus), network cable, etc.) or in a wireless manner (such as mobile network, WIFI (Wireless Fidelity), Bluetooth, etc.).

[0087] The bus 1050 includes a path for transmitting information between various components of the device (such as the processor 1010, the memory 1020, the input / output interface 1030, and the communication interface 1040).

[0088] It should be noted that although the above device only shows the processor 1010, the memory 1020, the input / output interface 1030, the communication interface 1040, and the bus 1050, in the specific implementation process, the device may also include other components necessary for normal operation. In addition, those skilled in the art can understand that the above device may also only include the components necessary to implement the solution of the embodiments of this specification, and does not necessarily include all the components shown in the figure.

[0089] The system of the above embodiment is used to implement the corresponding winter wheat drought warning method based on multi-source data in any of the foregoing embodiments, and has the beneficial effects of the corresponding method embodiments, which will not be elaborated here.

[0090] Embodiment 4

[0091] Based on the same inventive concept, corresponding to the method of any of the above embodiments, the present invention also provides a non-transitory computer-readable storage medium storing computer instructions for causing the computer to execute the winter wheat drought warning method based on multi-source data as described in any of the above embodiments.

[0092] The computer-readable medium of this embodiment includes permanent and non-permanent, removable and non-removable media, and information storage can be implemented by any method or technology. The 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 technologies, compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassette tapes, magnetic tape magnetic disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information accessible by a computing device.

[0093] The computer instructions stored in the storage medium of the above embodiment are used to cause the computer to execute the winter wheat drought warning method based on multi-source data as described in any of the above embodiments, and have the beneficial effects of the corresponding method embodiments, which will not be elaborated here.

[0094] Those of ordinary skill in the art should understand that any discussion of the above embodiments is merely exemplary and is not intended to imply that the scope of the present disclosure (including the claims) is limited to these examples; under the concept of the present disclosure, the technical features in the above embodiments or different embodiments can also be combined, the steps can be implemented in any order, and there are many other variations in different aspects of the embodiments of the present disclosure as described above, and for the sake of brevity, they are not provided in detail.

[0095] In addition, for the sake of simplicity of description and discussion, and in order not to make the embodiments of the present disclosure difficult to understand, the well-known power / ground connections to integrated circuit (IC) chips and other components may or may not be shown in the accompanying drawings. Further, the devices may be shown in block diagram form in order to avoid making the embodiments of the present disclosure difficult to understand, and this also takes into account the fact that the details of the implementation of these block diagram devices are highly dependent on the platform on which the embodiments of the present disclosure are to be implemented (i.e., these details should be entirely within the understanding of those skilled in the art). In cases where specific details (such as circuits) are set forth to describe exemplary embodiments of the present disclosure, it will be apparent to those skilled in the art that the embodiments of the present disclosure can be implemented without these specific details or with variations of these specific details. Therefore, these descriptions should be considered illustrative rather than restrictive.

[0096] Although the present disclosure has been described in connection with specific embodiments thereof, many alternatives, modifications, and variations of these embodiments will be apparent to those of ordinary skill in the art in light of the foregoing description. For example, other memory architectures (such as dynamic RAM (DRAM)) may be used with the embodiments discussed.

[0097] Therefore, the units of the examples described in the embodiments of the present application can be implemented in electronic hardware, or in a combination of computer software and electronic hardware. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods for each specific application to implement the described functions, but such implementation should not be considered to exceed the scope of the present application.

[0098] The embodiments of the present disclosure are intended to cover all such alternatives, modifications, and variations that fall within the broad scope of the appended claims. Therefore, any omissions, modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the embodiments of the present disclosure shall be included within the protection scope of the present disclosure.

Claims

1. A winter wheat drought warning method based on multi-source data, characterized in that, The method includes: Obtain satellite remote sensing vegetation indices, meteorological parameters, and soil moisture data, and perform time series synchronization processing on the obtained data through a preset time window to obtain a multi-source dataset with a unified time resolution; Use the principal component analysis method to perform dimensionality reduction processing on the multi-source dataset to obtain a set of feature vectors after dimensionality reduction; Through a preset winter wheat growth stage division rule, divide the set of feature vectors to obtain a staged feature sequence; Use a long short-term memory network to model the changing trend of water demand characteristics of the feature sequences in each growth stage to obtain a prediction parameter set reflecting the dynamic water demand relationship; Extract the water demand change trend from the prediction parameter set, extract drought characteristics from the water demand change trend and perform drought grading, and issue a drought warning according to the result of the drought grading.

2. The winter wheat drought warning method based on multi-source data according to claim 1, wherein The method for obtaining a multi-source dataset with a unified time resolution includes: Obtain satellite remote sensing data, extract vegetation indices, meteorological parameters, and soil moisture data from the satellite remote sensing data to obtain an initial multi-source data set; Divide the initial multi-source data set through a preset time window to obtain data subsets within a time period; For the data subsets within the time period, use a time series alignment method to align the vegetation indices, meteorological parameters, and soil moisture data therein to obtain a synchronized data sequence; Calculate the unified time resolution and perform time resolution alignment on the synchronized data sequence to obtain the multi-source dataset.

3. The winter wheat drought warning method based on multi-source data according to claim 1, characterized in that The method for obtaining a set of feature vectors after dimensionality reduction includes: Use the principal component analysis method to perform dimensionality reduction processing on the satellite remote sensing, meteorological data, and soil moisture in the multi-source dataset to obtain a preliminary set of feature vectors; Extract the feature vectors with the principal component contribution rate greater than a preset value from the preliminary set of feature vectors to obtain a refined set of feature vectors; For the refined set of feature vectors, calculate the correlation between each feature vector in the group using the covariance matrix, judge the degree of independence between each feature vector. If the correlation between features exceeds a preset threshold, perform orthogonalization processing on the set of feature vectors through linear transformation to obtain an orthogonal feature set; Construct a data set after dimensionality reduction according to the orthogonal feature set, obtain a low-dimensional representation form of heterogeneous features, and obtain the set of feature vectors after dimensionality reduction.

4. The winter wheat drought early warning method based on multi-source data according to claim 1, characterized in that The method for obtaining a staged feature sequence includes: Through a preset winter wheat growth stage division rule, judge the feature distribution corresponding to the winter wheat growth stage to obtain a preliminary stage division result; Use the K-means clustering algorithm to perform clustering processing on the preliminary stage division result to determine the boundary of the staged features and obtain an optimized staged feature sequence; Through the optimized staged feature sequence, obtain the complete feature distribution of the winter wheat growth stage to obtain the staged feature sequence.

5. The winter wheat drought warning method based on multi-source data according to claim 1, characterized in that, The method for obtaining a prediction parameter set includes: Use a long short-term memory network to model the feature sequence to obtain the water demand characteristics of each growth stage; Through trend analysis of the water demand characteristics, determine the trend analysis result to obtain updated water demand characteristics. Generate a dynamic relationship description according to the updated water requirement characteristics, and obtain prediction parameters through the dynamic relationship description to obtain the prediction parameter set.

6. The method for winter wheat drought early warning based on multi-source data according to claim 1, characterized in that, The method for issuing a drought warning according to the result of drought grading includes: Extract the water requirement change data from the prediction parameter set, and use statistical methods to calculate the change trend to obtain the water requirement change trend; Obtain drought-related features from the water requirement change trend, and use a logistic regression algorithm to grade the drought severity based on the drought-related features to obtain a preliminary grading result; For the preliminary grading result, obtain a severity quantification value, and perform normalization processing through a preset threshold to obtain a normalized evaluation level; According to the normalized evaluation level, judge whether the drought severity reaches the warning condition. If it does, output an alarm signal through the warning mechanism.

7. A winter wheat drought early warning system based on multi-source data, the system is used to implement the method described in any one of claims 1-6, characterized in that, Including: A multi-source data acquisition module, a feature vector acquisition module, a feature sequence acquisition module, a prediction parameter set calculation module, and a drought warning module; The multi-source data acquisition module is used to acquire satellite remote sensing vegetation indices, meteorological parameters, and soil moisture data, and perform time series synchronization processing on the obtained data through a preset time window to obtain a multi-source data set with a unified time resolution; The feature vector acquisition module uses the principal component analysis method to perform dimensionality reduction processing on the multi-source data set to obtain a reduced-dimensional feature vector set; The feature sequence acquisition module divides the growth stages of the feature vector set according to a preset winter wheat growth stage division rule to obtain a staged feature sequence; The prediction parameter set calculation module uses a long short-term memory network to model the change trend of the water requirement characteristics of the feature sequences in each growth stage to obtain a prediction parameter set reflecting the dynamic water requirement relationship; The drought warning module is used to extract the water requirement change trend from the prediction parameter set, extract drought features from the water requirement change trend and perform drought grading, and issue a drought warning according to the result of the drought grading.

8. An electronic device, characterized in that, It includes a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the program, it implements the winter wheat drought warning method based on multi-source data according to any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, and when the computer program is executed, it implements the winter wheat drought warning method based on multi-source data according to any one of claims 1 to 6.

Images