Multi-source meteorological fusion-based low-temperature cold damage early warning method and system for rice

Abstract

The application discloses a rice low-temperature cold damage early warning method and system based on multi-source meteorological data fusion, and relates to the technical field of data processing.The method comprises the following steps: collecting remote sensing data and meteorological data in a rice area, wherein the meteorological data comprises meteorological temperature and meteorological humidity, and the remote sensing data comprises remote sensing feature data and remote sensing content data; identifying temperature parameters and chlorophyll parameters in the rice area, and inversely deriving remote sensing temperature precision and remote sensing chlorophyll precision according to the remote sensing feature data and the meteorological data; combining the remote sensing temperature precision with the meteorological data to obtain fused temperature through fusion processing, and combining the meteorological humidity to obtain an environmental low-temperature parameter; and predicting rice low-temperature cold damage parameters according to the environmental low-temperature parameter, the chlorophyll parameter and the remote sensing chlorophyll precision, and performing early warning discrimination.The method solves the technical problem that the existing rice low-temperature cold damage early warning method only relies on a single data source, resulting in insufficient early warning precision and difficulty in effectively preventing rice low-temperature cold damage.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data processing, in particular to a rice low-temperature cold damage early warning method and system based on multi-source meteorological data fusion. BACKGROUND

[0002] Rice is one of the important food crops in China, and is prone to be affected by low-temperature cold damage during its growth process. In particular, the frequent low-temperature weather events in the high-latitude areas such as the Northeast China tend to have a significant adverse impact on the growth of rice, and even cause a reduction in yield or even a complete loss in severe cases. Therefore, early and accurate monitoring and early warning of low-temperature cold damage risks are important guarantees for improving rice yield and ensuring the safety of agricultural production. Traditional methods for monitoring rice low-temperature cold damage mainly rely on manual observation or single ground meteorological station observation. The above schemes have great limitations in actual use, for example: manual observation is time-consuming and laborious, and has a small monitoring range and poor timeliness; ground meteorological station observation data is accurate but has insufficient spatial distribution density, and it is difficult to accurately capture the differences in cold damage risks at the regional scale. In addition, in recent years, remote sensing technology has developed rapidly, and remote sensing platforms such as satellites and unmanned aerial vehicles can quickly, widely and frequently obtain surface vegetation and environmental information such as surface temperature, vegetation index, chlorophyll content and other vegetation state parameters. However, remote sensing technology also has its own problems in practical application, such as insufficient accuracy and data interference by cloudy and foggy weather. Therefore, it is difficult to ensure the accuracy of monitoring and early warning by relying solely on remote sensing data.

[0003] Therefore, the rice low-temperature cold damage early warning method in the prior art has the technical problem of insufficient early warning accuracy due to reliance on a single data source, which makes it difficult to effectively prevent rice low-temperature cold damage. SUMMARY

[0004] The present application provides a rice low-temperature cold damage early warning method and system based on multi-source meteorological data fusion, which solves the technical problem of insufficient early warning accuracy due to reliance on a single data source in the prior art, which makes it difficult to effectively prevent rice low-temperature cold damage. By fusing remote sensing data and meteorological data, and considering the plant state and remote sensing recognition accuracy, the accuracy and reliability of rice low-temperature cold damage early warning are significantly improved.

[0005] The application provides a rice low-temperature cold damage early warning method based on multi-source meteorological data fusion, which comprises the following steps: collecting remote sensing data and meteorological data in a rice area, wherein the meteorological data comprises meteorological temperature and meteorological humidity, and the remote sensing data comprises remote sensing feature data and remote sensing content data; identifying temperature parameters and chlorophyll parameters in the rice area according to the remote sensing content data, and inversely deriving remote sensing temperature precision and remote sensing chlorophyll precision according to the remote sensing feature data and the meteorological data; combining the remote sensing temperature precision with the meteorological data to obtain fusion temperature through fusion processing, and combining the fusion temperature with the meteorological humidity to obtain an environmental low-temperature parameter; and predicting a rice low-temperature cold damage parameter according to the environmental low-temperature parameter, the chlorophyll parameter and the remote sensing chlorophyll precision, and performing early warning discrimination.

[0006] In an implementation, the remote sensing data and the meteorological data in the rice area are collected, including: collecting remote sensing data in the rice area, wherein the remote sensing data comprises remote sensing feature data and remote sensing content data, and the remote sensing feature data comprises remote sensing resolution of different wave bands; and collecting meteorological data in the rice area, wherein the meteorological data comprises meteorological temperature and meteorological humidity.

[0007] In an implementation, the temperature parameters and the chlorophyll parameters in the rice area are identified according to the remote sensing content data, including: processing the temperature parameters in the rice area according to thermal infrared wave band radiation values in the remote sensing content data; and identifying the chlorophyll parameters in the rice area according to the remote sensing content data.

[0008] In an implementation, the chlorophyll parameters in the rice area are identified according to the remote sensing content data, including: calculating a normalized vegetation index according to the remote sensing content data; and inputting the normalized vegetation index into a chlorophyll classification table to obtain the chlorophyll parameters, wherein the chlorophyll classification table is constructed based on a mapping relationship between a sample normalized vegetation index set and a sample chlorophyll parameter set of the remote sensing content data of the rice area.

[0009] In an implementation, the remote sensing temperature precision and the remote sensing chlorophyll precision are inversely derived according to the remote sensing feature data and the meteorological data, including: calling a remote sensing precision classifier, wherein the remote sensing precision classifier is constructed based on a sample remote sensing feature data set, a sample meteorological data set, a sample remote sensing temperature precision set and a sample remote sensing chlorophyll precision set, the sample remote sensing temperature precision comprises a similarity between a labeled remote sensing identified temperature and an actual temperature, and the sample remote sensing chlorophyll precision comprises a similarity between a labeled remote sensing identified chlorophyll parameter and an actual chlorophyll parameter; and inputting the remote sensing feature data and the meteorological data into the remote sensing precision classifier to output the remote sensing temperature precision and the remote sensing chlorophyll precision.

[0010] In an implementation, the environment low temperature parameter is obtained by fusing the remote sensing temperature precision and the meteorological data, and combining the meteorological humidity, including: correcting a preset remote sensing temperature weight by the remote sensing temperature precision to obtain a remote sensing temperature weight, and determining a meteorological temperature weight; performing weighted calculation on the temperature parameter and the meteorological temperature by using the remote sensing temperature weight and the meteorological temperature weight to obtain a fusion temperature; and obtaining the environment low temperature parameter by combining the fusion temperature and the meteorological humidity.

[0011] In an implementation, the environment low temperature parameter, the chlorophyll parameter and the remote sensing chlorophyll precision are used to predict a rice low temperature cold damage parameter, and to perform early warning discrimination, including: configuring a cold damage demand coefficient according to the chlorophyll parameter and the remote sensing chlorophyll precision; obtaining a rice low temperature cold damage predictor group, and configuring a selected rice low temperature cold damage predictor sequence according to the cold damage demand coefficient, wherein a sample environment low temperature parameter set, a sample chlorophyll parameter set and a sample rice low temperature cold damage parameter set are collected, and multiple training data are divided to train the rice low temperature cold damage predictor group; the environment low temperature parameter and the chlorophyll parameter are combined and input into the rice low temperature cold damage predictor sequence respectively, the mean of an output result sequence is calculated to obtain the rice low temperature cold damage parameter, and early warning discrimination is performed, wherein the rice low temperature cold damage parameter includes a rice low temperature cold damage grade.

[0012] In an implementation, the cold damage demand coefficient is configured according to the chlorophyll parameter and the remote sensing chlorophyll precision, including: obtaining a chlorophyll parameter peak value of rice growth; calculating a ratio of the chlorophyll parameter and the chlorophyll parameter peak value to obtain a first cold damage demand coefficient; calculating a second cold damage coefficient according to the remote sensing chlorophyll precision; and calculating the cold damage demand coefficient according to the first cold damage coefficient and the second cold damage coefficient.

[0013] The application also provides a rice low temperature cold damage early warning system based on multi-source meteorological data fusion, including: a data acquisition module, configured to acquire remote sensing data and meteorological data in a rice area, wherein the meteorological data includes meteorological temperature and meteorological humidity, and the remote sensing data includes remote sensing feature data and remote sensing content data; an accuracy acquisition module, configured to identify a temperature parameter and a chlorophyll parameter in the rice area according to the remote sensing content data, and to obtain a remote sensing temperature precision and a remote sensing chlorophyll precision by back calculation according to the remote sensing feature data and the meteorological data; an environment low temperature parameter acquisition module, configured to obtain a fusion temperature by fusing the remote sensing temperature precision and the meteorological data, and to obtain an environment low temperature parameter by combining the meteorological humidity; and an early warning discrimination module, configured to predict a rice low temperature cold damage parameter according to the environment low temperature parameter, the chlorophyll parameter and the remote sensing chlorophyll precision, and to perform early warning discrimination.

[0014] The rice low-temperature cold damage early warning method and system provided by the application can collect remote sensing data and meteorological data in a rice area, wherein the meteorological data includes meteorological temperature and meteorological humidity, and the remote sensing data includes remote sensing feature data and remote sensing content data; according to the remote sensing content data, temperature parameters and chlorophyll parameters in the rice area are identified, and remote sensing temperature precision and remote sensing chlorophyll precision are obtained according to remote sensing feature data and meteorological data; according to the remote sensing temperature precision and meteorological data, fusion processing is performed to obtain a fusion temperature, and an environmental low-temperature parameter is obtained in combination with the meteorological humidity; according to the environmental low-temperature parameter, the chlorophyll parameter and the remote sensing chlorophyll precision, a rice low-temperature cold damage parameter is predicted to perform early warning discrimination. The technical problem that the existing rice low-temperature cold damage early warning method only relies on a single data source, resulting in insufficient early warning precision and difficulty in effectively preventing rice low-temperature cold damage, is solved. By fusing remote sensing data and meteorological data and comprehensively considering the plant state and remote sensing recognition precision, the accuracy and reliability of rice low-temperature cold damage early warning are significantly improved. BRIEF DESCRIPTION OF DRAWINGS

[0015] In order to more clearly illustrate the technical solutions of the embodiments of the present disclosure, the drawings of the embodiments of the present disclosure will be briefly introduced below. In the present application, a flowchart is used to illustrate the operations performed by the system according to the embodiments of the present application. It should be understood that the foregoing or the following operations are not necessarily performed in sequence. On the contrary, various steps can be processed in reverse order or simultaneously as needed. Meanwhile, other operations can be added to these processes, or a step or several steps can be removed from these processes.

[0016] Figure 1 A process schematic diagram of the rice low-temperature cold damage early warning method provided by the embodiments of the present application is provided.

[0017] Figure 2 A structure schematic diagram of the rice low-temperature cold damage early warning system provided by the embodiments of the present application is provided.

[0018] Explanation of reference signs: data acquisition module 11, precision acquisition module 12, environmental low-temperature parameter acquisition module 13, early warning discrimination module 14. DETAILED DESCRIPTION

[0019] The above description is only a summary of the technical solutions of the present application. In order to more clearly understand the technical means of the present application, the embodiments of the present application can be implemented according to the content of the description, and in order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the following specific embodiments of the present application are described.

[0020] In order to make the purposes, technical solutions and advantages of the present application clearer, the following will further describe the present application in conjunction with the accompanying drawings, the described embodiments should not be regarded as limitations to the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0021] In the following description, "some embodiments" are referred to, which describe a subset of all possible embodiments, but it can be understood that "some embodiments" can be the same subset or different subsets of all possible embodiments, and can be combined with each other without conflict, the term "first\second" referred to only distinguishes similar objects, and does not represent a specific order for the objects. The terms "include" and "have" and any variations, are intended to cover non-exclusive inclusion, for example, a process, method, system, product or server including a series of steps or units does not have to be limited to those steps or units clearly listed, but can include other steps or modules not clearly listed or inherent to these processes, methods, products or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as understood by those skilled in the art to which the present application belongs. The terms used herein are only for the purpose of describing the embodiments of the present application.

[0022] The embodiments of the present application provide a rice low-temperature cold damage early warning method and system based on multi-source meteorological fusion, as shown in Figure 1 The method comprises the following steps:

[0023] Collect remote sensing data and meteorological data in a rice area, wherein the meteorological data comprises meteorological temperature and meteorological humidity, and the remote sensing data comprises remote sensing feature data and remote sensing content data; according to the remote sensing content data, identify temperature parameters and chlorophyll parameters in the rice area, and obtain remote sensing temperature precision and remote sensing chlorophyll precision according to the remote sensing feature data and the meteorological data.

[0024] Before performing rice low-temperature cold damage early warning, multi-dimensional data collection needs to be performed on the rice planting area to be early warned, so as to obtain remote sensing data and meteorological data in the rice area. The meteorological data comprises meteorological temperature and meteorological humidity, and the remote sensing data comprises remote sensing feature data and remote sensing content data. The remote sensing feature data is remote sensing resolution corresponding to different wave bands and wave band pairs, and the remote sensing content data is image data collected under corresponding wave bands. Subsequently, according to the remote sensing content data, temperature parameters and chlorophyll parameters in the rice area are identified. And remote sensing temperature precision corresponding to the temperature parameters and remote sensing chlorophyll precision corresponding to the chlorophyll parameters are obtained according to the obtained remote sensing feature data and the meteorological data.

[0025] The method provided in the embodiments of the present application further includes: collecting remote sensing data in the rice area, wherein the remote sensing data includes remote sensing feature data and remote sensing content data, and the remote sensing feature data includes remote sensing resolution of different wave bands; and collecting meteorological data in the rice area, wherein the meteorological data includes meteorological temperature and meteorological humidity.

[0026] The remote sensing data and the meteorological data in the rice area are collected by: obtaining image data in different wave bands, such as a visible light wave band, a near-infrared wave band, a thermal infrared wave band and the like, by a remote sensing platform such as a remote sensing satellite or a drone, to obtain the remote sensing data in the rice area. The remote sensing data includes remote sensing feature data and remote sensing content data, the remote sensing feature data includes remote sensing resolution of different wave bands, and the remote sensing content data is data collected in corresponding wave bands, such as thermal infrared wave band data, visible light wave band data and near-infrared wave band data and the like. Further, the meteorological data in the rice area is collected by a meteorological station, and the meteorological data includes meteorological temperature and meteorological humidity, the meteorological temperature is actually measured air temperature data, the meteorological humidity is actually measured air relative humidity data, and the meteorological temperature and the meteorological humidity both include a time identifier.

[0027] The method provided in the embodiments of the present application further includes: processing the thermal infrared wave band radiation value in the remote sensing content data to obtain a temperature parameter in the rice area; and identifying a chlorophyll parameter in the rice area according to the remote sensing content data.

[0028] The temperature parameter and the chlorophyll parameter in the rice area are identified according to the remote sensing content data, including: converting the thermal infrared wave band radiation value into a ground temperature parameter by a thermal radiation physical formula according to the thermal infrared wave band radiation value in the remote sensing content data, to obtain the temperature parameter in the rice area. The thermal infrared wave band is a long wave band, which can sense the thermal energy of ground radiation, thereby directly reflecting the temperature state of ground objects. Further, the chlorophyll parameter in the rice area is identified according to the remote sensing content data.

[0029] The method provided in the embodiments of the present application further includes: calculating a normalized vegetation index according to the remote sensing content data; and inputting the normalized vegetation index into a chlorophyll classification table to obtain a chlorophyll parameter, wherein the chlorophyll classification table is constructed based on a mapping relationship between a sample normalized vegetation index set and a sample chlorophyll parameter set of the remote sensing content data of the rice area.

[0030] According to the remote sensing content data, a chlorophyll parameter in the rice area is identified, including: according to the remote sensing content data, visible red light band data and near-infrared band data in the remote sensing content data are obtained. Based on a normalized difference vegetation index calculation formula, the normalized difference vegetation index is calculated according to the visible red light band data and the near-infrared band data. Then, the calculated normalized difference vegetation index is input into a chlorophyll classification table to obtain the corresponding chlorophyll parameter. The chlorophyll classification table is a statistical corresponding relationship database established based on a large amount of sample data of historical rice areas, including a large number of NDVI values, i.e., a sample normalized difference vegetation index set and corresponding measured chlorophyll content data. In the chlorophyll classification table, the chlorophyll parameter values corresponding to different NDVI value intervals and vegetation state identification data, such as normal plants, vigorous plants, and weak plants, are recorded.

[0031] The method provided by the embodiment of the application further includes: calling a remote sensing accuracy classifier, wherein the remote sensing accuracy classifier is constructed based on a sample remote sensing feature data set, a sample meteorological data set, a sample remote sensing temperature accuracy set, and a sample remote sensing chlorophyll accuracy set. The sample remote sensing temperature accuracy includes the similarity between the labeled remote sensing identified temperature and the actual temperature, and the sample remote sensing chlorophyll accuracy includes the similarity between the labeled remote sensing identified chlorophyll parameter and the actual chlorophyll parameter; and the remote sensing feature data and the meteorological data are input into the remote sensing accuracy classifier to output the obtained remote sensing temperature accuracy and remote sensing chlorophyll accuracy.

[0032] The remote sensing temperature accuracy and the remote sensing chlorophyll accuracy are obtained according to the remote sensing feature data and the meteorological data, including: constructing a remote sensing accuracy classifier, wherein the remote sensing accuracy classifier is constructed based on a random forest algorithm, and the construction data include a sample remote sensing feature data set, a sample meteorological data set, a sample remote sensing temperature accuracy set, and a sample remote sensing chlorophyll accuracy set. The sample remote sensing feature data set is resolution data of different bands obtained from a remote sensing satellite or an unmanned aerial vehicle sensor in the sample area collection process, and is used to describe the remote sensing data features of each sample area. The sample meteorological data set is ground observation meteorological data actually collected in the sample area collection process, including meteorological temperature, meteorological humidity, etc. The sample remote sensing temperature accuracy set is the similarity between the remote sensing identified temperature of the historical sample area and the actual measured temperature, and is used to supervise the temperature accuracy prediction of the classifier. The sample remote sensing chlorophyll accuracy set is the similarity index between the remote sensing identified chlorophyll parameter of the historical sample area and the measured chlorophyll parameter, and is used to supervise the chlorophyll accuracy prediction of the classifier. The sample remote sensing feature data set, the sample meteorological data set, the sample remote sensing temperature accuracy set, and the sample remote sensing chlorophyll accuracy set are input into the random forest algorithm to supervise the training of the model until the accuracy of the remote sensing temperature accuracy and the remote sensing chlorophyll accuracy parameters output by the model meets the preset requirement, and the trained remote sensing accuracy classifier is obtained.

[0033] by calling a remote sensing precision classifier, wherein the remote sensing precision classifier is constructed based on a sample remote sensing feature dataset, a sample meteorological dataset, a sample remote sensing temperature precision set and a sample remote sensing chlorophyll precision set, the sample remote sensing temperature precision includes the similarity between the labeled remote sensing identified temperature and the actual temperature, and the sample remote sensing chlorophyll precision includes the similarity between the labeled remote sensing identified chlorophyll parameter and the actual chlorophyll parameter. Finally, the remote sensing feature data and the meteorological data are input into the remote sensing precision classifier, and remote sensing temperature precision and remote sensing chlorophyll precision are output.

[0034] According to the remote sensing temperature precision combined with meteorological data, a fusion temperature is obtained by fusion processing, and an environmental low-temperature parameter is obtained in combination with the meteorological humidity. According to the environmental low-temperature parameter, the chlorophyll parameter and the remote sensing chlorophyll precision, a rice low-temperature cold damage parameter is predicted and obtained, and a warning judgment is performed.

[0035] According to the remote sensing temperature precision combined with meteorological data, a fusion temperature is obtained by fusion processing, and an environmental low-temperature parameter is obtained in combination with the meteorological humidity. Further, according to the environmental low-temperature parameter, the chlorophyll parameter and the remote sensing chlorophyll precision, a rice low-temperature cold damage predictor sequence is input respectively, the mean value of the output result sequence is calculated, a rice low-temperature cold damage parameter is obtained, a warning judgment is performed on the rice low-temperature cold damage parameter, and a corresponding judgment level is obtained, such as no cold damage, mild, moderate and severe levels. Thus, the technical problem that the existing rice low-temperature cold damage warning method only relies on a single data source, resulting in insufficient warning precision and difficulty in effectively preventing rice low-temperature cold damage, is solved. By fusing remote sensing data and meteorological data and comprehensively considering the plant state and remote sensing recognition precision, the accuracy and reliability of rice low-temperature cold damage warning are significantly improved.

[0036] The method provided by the embodiment of the present application further includes: correcting a preset remote sensing temperature weight by using the remote sensing temperature precision to obtain a remote sensing temperature weight, and determining a meteorological temperature weight; performing weighted calculation on the temperature parameter and the meteorological temperature by using the remote sensing temperature weight and the meteorological temperature weight to obtain a fusion temperature; and obtaining an environmental low-temperature parameter in combination with the fusion temperature and the meteorological humidity.

[0037] According to the remote sensing temperature precision combined with meteorological data, a fusion temperature is obtained by fusion processing, and an environmental low temperature parameter is obtained by combining the meteorological humidity, comprising: correcting a preset remote sensing temperature weight by using the remote sensing temperature precision, the preset remote sensing temperature weight is 0.5, the higher the remote sensing temperature precision, the higher the corrected remote sensing temperature weight, and the highest cannot be higher than the preset remote sensing temperature weight. The corrected remote sensing temperature weight is the product of the preset remote sensing temperature weight and the remote sensing temperature precision. After determining the remote sensing temperature weight, since the sum of the meteorological temperature weight and the remote sensing temperature weight is 1, the meteorological temperature weight can be obtained according to the obtained remote sensing temperature weight. Finally, the temperature parameter and the meteorological temperature are weighted calculated according to the obtained remote sensing temperature weight and the meteorological temperature weight, and then the fusion temperature is obtained. For example, the remote sensing temperature parameter of a certain area in the A area is 14℃ calculated by remote sensing, the actual temperature measured by the local meteorological station is 12℃, the remote sensing temperature weight obtained according to the remote sensing temperature precision is 0.45, and the meteorological temperature weight is 0.55, so the fusion temperature after calculation is 12.9℃. The final obtained fusion temperature 12.9℃ can comprehensively integrate the advantages of remote sensing data and ground observation, and is closer to the real situation. After obtaining the fusion temperature, the environmental low temperature parameter is obtained by further combining the meteorological humidity data, and the environmental low temperature parameter is the calculation result of the simplified formula of the temperature and humidity index THI.

[0038] The method provided by the embodiment of the application further comprises: configuring a cold damage demand coefficient according to the chlorophyll parameter and the remote sensing chlorophyll precision; obtaining a group of rice low temperature cold damage predictors, and configuring a selected sequence of rice low temperature cold damage predictors according to the cold damage demand coefficient, wherein a sample environmental low temperature parameter set, a sample chlorophyll parameter set and a sample rice low temperature cold damage parameter set are collected, a plurality of training data are divided, and a group of rice low temperature cold damage predictors are obtained by training; the environmental low temperature parameter and the chlorophyll parameter are combined and input into the sequence of rice low temperature cold damage predictors respectively, the mean value of the output result sequence is calculated, a rice low temperature cold damage parameter is obtained, and early warning discrimination is performed, wherein the rice low temperature cold damage parameter comprises a rice low temperature cold damage grade.

[0039] According to the environmental low-temperature parameter, the chlorophyll parameter, and the remote sensing chlorophyll accuracy, a rice low-temperature cold damage parameter is predicted to obtain a warning judgment, including: according to the chlorophyll parameter and the remote sensing chlorophyll accuracy, a cold damage demand coefficient is configured. Further, a rice low-temperature cold damage predictor group is obtained, and the rice low-temperature cold damage predictor group includes a plurality of rice low-temperature cold damage predictors. A mapping relationship between a cold damage demand coefficient parameter range and a rice low-temperature cold damage predictor quantity is constructed in advance, and each cold damage demand coefficient can obtain a corresponding cold damage demand coefficient parameter range and a specific configured rice low-temperature cold damage predictor quantity in the mapping relationship. According to the cold damage demand coefficient, the rice low-temperature cold damage predictor quantity is matched in combination with the mapping relationship, a corresponding quantity of rice low-temperature cold damage predictors is randomly selected from the rice low-temperature cold damage predictor group according to the rice low-temperature cold damage predictor quantity, and a rice low-temperature cold damage predictor sequence is obtained.

[0040] The rice low-temperature cold damage predictor is obtained by training a plurality of training data based on a sample environmental low-temperature parameter set, a sample chlorophyll parameter set, and a sample rice low-temperature cold damage parameter set. Specifically, the sample environmental low-temperature parameter set, i.e., the historical environmental low-temperature index data, is obtained from historical data. The sample chlorophyll parameter set is the historical chlorophyll content estimated by remote sensing. The sample rice low-temperature cold damage parameter set is the rice low-temperature cold damage grade parameter evaluated by artificial according to the actual cold damage situation. The rice low-temperature cold damage grade parameter has a value of 1 to 10. Different rice low-temperature cold damage grade parameters correspond to different grades. The lower the cold damage parameter value grade is, the lower the cold damage severity is. The rice low-temperature cold damage grade parameter value can also be mapped to an intuitive evaluation grade, such as no cold damage, mild, moderate, and severe grades, to intuitively present to the user. For example, the no cold damage grade corresponds to a rice low-temperature cold damage grade parameter value of 0 to 2, the mild grade corresponds to a rice low-temperature cold damage grade parameter value of 2 to 5, the moderate grade corresponds to a rice low-temperature cold damage grade parameter value of 5 to 8, and the severe grade corresponds to a rice low-temperature cold damage grade parameter value of 8 to 10. Each value is left open and right closed. All the above sample data is divided into a plurality of training data to supervise the training of a plurality of neural network models. When the model converges, a plurality of trained rice low-temperature cold damage predictors are obtained to form a rice low-temperature cold damage predictor group. Finally, the environmental low-temperature parameter and the chlorophyll parameter are combined and input into the rice low-temperature cold damage predictor sequence to obtain the rice low-temperature cold damage grade parameter output by each rice low-temperature cold damage predictor. The mean value of the output result sequence is calculated to obtain the rice low-temperature cold damage parameter for warning judgment. The rice low-temperature cold damage parameter includes the rice low-temperature cold damage grade.

[0041] The method provided by the embodiment of the application further includes: obtaining a chlorophyll parameter peak value of the rice growth; calculating a ratio of the chlorophyll parameter and the chlorophyll parameter peak value to obtain a first cold damage requirement coefficient; calculating a second cold damage coefficient according to the remote sensing chlorophyll accuracy; and calculating a cold damage requirement coefficient according to the first cold damage coefficient and the second cold damage coefficient.

[0042] According to the chlorophyll parameter and the remote sensing chlorophyll accuracy, the cold damage requirement coefficient is configured, including: obtaining a highest value of chlorophyll content of the rice plant in a normal growth and development cycle, which is the chlorophyll parameter peak value. Then, a ratio of the obtained chlorophyll parameter and the chlorophyll parameter peak value is calculated to obtain the first cold damage requirement coefficient. The lower the first cold damage requirement coefficient is, the greater the gap between the current plant and the best growth state of the plant itself is, the lower the resistance to cold damage is, and the higher the cold damage risk is. Then, the second cold damage coefficient is obtained according to a result of 1 minus the remote sensing chlorophyll accuracy. The smaller the second cold damage coefficient is, the smaller the influence of the chlorophyll accuracy on the cold damage risk prediction is. The first cold damage coefficient is compensated by the second cold damage coefficient to obtain the final cold damage requirement coefficient. According to the first cold damage coefficient and the second cold damage coefficient, the cold damage requirement coefficient is calculated by multiplying the sum of 1 and the second cold damage coefficient by the first cold damage coefficient. The higher the cold damage requirement coefficient is, the higher the low-temperature sensitivity of the plant is, and the greater the influence of the chlorophyll accuracy on the cold damage risk prediction is. Therefore, a larger number of rice low-temperature cold damage predictors need to be set, and the accuracy of the prediction is improved by comprehensively analyzing a plurality of output results.

[0043] In the foregoing, reference is made to Figure 1 The method for rice low-temperature cold damage early warning based on multi-source meteorological data fusion according to the embodiment of the application is described in detail. Next, the system for rice low-temperature cold damage early warning based on multi-source meteorological data fusion according to the embodiment of the application will be described with reference to the accompanying drawings. Figure 2 The system for rice low-temperature cold damage early warning based on multi-source meteorological data fusion according to the embodiment of the application is described.

[0044] The system for rice low-temperature cold damage early warning based on multi-source meteorological data fusion according to the embodiment of the application solves the technical problem in the prior art that the rice low-temperature cold damage early warning method only depends on a single data source, resulting in insufficient early warning accuracy and difficulty in effectively preventing rice low-temperature cold damage. By fusing remote sensing data and meteorological data and comprehensively considering the plant state and remote sensing recognition accuracy, the accuracy and reliability of the rice low-temperature cold damage early warning are significantly improved. The system for rice low-temperature cold damage early warning based on multi-source meteorological data fusion includes a data acquisition module 11, an accuracy acquisition module 12, an environmental low-temperature parameter acquisition module 13, and an early warning discrimination module 14.

[0045] The data acquisition module 11 is configured to acquire remote sensing data and meteorological data in the rice area, wherein the meteorological data comprises meteorological temperature and meteorological humidity, and the remote sensing data comprises remote sensing feature data and remote sensing content data; the precision acquisition module 12 is configured to identify temperature parameters and chlorophyll parameters in the rice area according to the remote sensing content data, and to obtain remote sensing temperature precision and remote sensing chlorophyll precision according to remote sensing feature data and meteorological data; the environmental low-temperature parameter acquisition module 13 is configured to obtain fused temperature by fusing the remote sensing temperature precision and the meteorological data, and to obtain the environmental low-temperature parameter by combining the fused temperature with the meteorological humidity; and the early warning discrimination module 14 is configured to obtain a rice low-temperature cold damage parameter by prediction according to the environmental low-temperature parameter, the chlorophyll parameter and the remote sensing chlorophyll precision, and to perform early warning discrimination.

[0046] Next, the specific configuration of the data acquisition module 11 will be described in detail. The data acquisition module 11 further comprises: acquiring remote sensing data and meteorological data in the rice area, comprising: acquiring remote sensing data in the rice area, wherein the remote sensing data comprises remote sensing feature data and remote sensing content data, and the remote sensing feature data comprises remote sensing resolution of different wave bands; and acquiring meteorological data in the rice area, wherein the meteorological data comprises meteorological temperature and meteorological humidity.

[0047] Next, the specific configuration of the precision acquisition module 12 will be described in detail. The precision acquisition module 12 further comprises: identifying temperature parameters and chlorophyll parameters in the rice area according to the remote sensing content data, comprising: processing the temperature parameters in the rice area according to thermal infrared wave band radiation values in the remote sensing content data; and identifying chlorophyll parameters in the rice area according to the remote sensing content data.

[0048] Next, the specific configuration of the precision acquisition module 12 will be described in detail. The precision acquisition module 12 further comprises: identifying chlorophyll parameters in the rice area according to the remote sensing content data, comprising: calculating a normalized vegetation index according to the remote sensing content data; and inputting the normalized vegetation index into a chlorophyll classification table to obtain the chlorophyll parameters, wherein the chlorophyll classification table is constructed based on a mapping relationship between a sample normalized vegetation index set and a sample chlorophyll parameter set of the remote sensing content data of the rice area.

[0049] Next, the specific configuration of the precision acquisition module 12 will be described in detail. The precision acquisition module 12 further comprises: obtaining remote sensing temperature precision and remote sensing chlorophyll precision according to remote sensing feature data and meteorological data, comprising: calling a remote sensing precision classifier, wherein the remote sensing precision classifier is constructed based on a sample remote sensing feature data set, a sample meteorological data set, a sample remote sensing temperature precision set, and a sample remote sensing chlorophyll precision set. The sample remote sensing temperature precision includes the similarity between the labeled remote sensing identified temperature and the actual temperature, and the sample remote sensing chlorophyll precision includes the similarity between the labeled remote sensing identified chlorophyll parameter and the actual chlorophyll parameter; inputting the remote sensing feature data and the meteorological data into the remote sensing precision classifier to output the remote sensing temperature precision and the remote sensing chlorophyll precision.

[0050] Next, the specific configuration of the environment low-temperature parameter acquisition module 13 will be described in detail. The environment low-temperature parameter acquisition module 13 further comprises: obtaining a fusion temperature by fusion processing according to the remote sensing temperature precision and meteorological data, and obtaining an environment low-temperature parameter by combining the meteorological humidity, comprising: correcting a preset remote sensing temperature weight using the remote sensing temperature precision to obtain a remote sensing temperature weight, and determining a meteorological temperature weight; performing weighted calculation on the temperature parameter and the meteorological temperature using the remote sensing temperature weight and the meteorological temperature weight to obtain a fusion temperature; obtaining an environment low-temperature parameter by combining the fusion temperature and the meteorological humidity.

[0051] Next, the specific configuration of the early warning discrimination module 14 will be described in detail. The early warning discrimination module 14 further comprises: predicting a rice low-temperature cold damage parameter according to the environment low-temperature parameter, the chlorophyll parameter, and the remote sensing chlorophyll precision, and performing early warning discrimination, comprising: configuring a cold damage demand coefficient according to the chlorophyll parameter and the remote sensing chlorophyll precision; obtaining a rice low-temperature cold damage predictor group, and configuring a selected rice low-temperature cold damage predictor sequence according to the cold damage demand coefficient, wherein a sample environment low-temperature parameter set, a sample chlorophyll parameter set, and a sample rice low-temperature cold damage parameter set are collected, and multiple training data are divided to train a rice low-temperature cold damage predictor group; inputting the environment low-temperature parameter and the chlorophyll parameter into the rice low-temperature cold damage predictor sequence respectively, calculating the mean value of the output result sequence to obtain a rice low-temperature cold damage parameter, and performing early warning discrimination, wherein the rice low-temperature cold damage parameter includes a rice low-temperature cold damage grade.

[0052] The specific configuration of the early warning discrimination module 14 will be described in detail below. The early warning discrimination module 14 further includes: configuring a chilling injury demand coefficient based on the chlorophyll parameters and the remote sensing chlorophyll accuracy, including: obtaining the peak value of the chlorophyll parameters during rice growth; calculating the ratio of the chlorophyll parameters to the peak value of the chlorophyll parameters to obtain a first chilling injury demand coefficient; calculating a second chilling injury coefficient based on the remote sensing chlorophyll accuracy; and calculating a chilling injury demand coefficient based on the first chilling injury coefficient and the second chilling injury coefficient.

[0053] The rice low-temperature chilling injury early warning system provided in the embodiments of the present invention can execute the rice low-temperature chilling injury early warning method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the method.

[0054] Although this application makes various references to certain modules in the system according to the embodiments of this application, any number of different modules can be used and run on user terminals and / or servers. The various units and modules included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of each functional unit are only for easy distinction between each other and are not used to limit the scope of protection of this invention.

[0055] The specific embodiments described above do not constitute a limitation on the scope of protection of this application. Those skilled in the art should understand that various modifications, combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A multi-source meteorological fusion method for early warning of low-temperature chilling injury in rice, characterized in that, The method includes: Remote sensing data and meteorological data are collected in the rice-growing area. The meteorological data includes meteorological temperature and meteorological humidity, and the remote sensing data includes remote sensing feature data and remote sensing content data. Based on the remote sensing data, the temperature and chlorophyll parameters within the rice paddy area are identified, and the accuracy of remote sensing temperature and chlorophyll is obtained by reverse calculation based on the remote sensing feature data and meteorological data. Based on the remote sensing temperature accuracy and meteorological data, a fused temperature is obtained by fusion processing, and environmental low temperature parameters are obtained by combining the meteorological humidity. Based on the aforementioned environmental low-temperature parameters, chlorophyll parameters, and remote sensing chlorophyll accuracy, parameters for low-temperature chilling injury in rice are predicted and used for early warning and discrimination, including: Configure the chilling injury demand coefficient based on the chlorophyll parameters and remote sensing chlorophyll accuracy. A rice low-temperature chilling injury predictor group is obtained. According to the chilling injury demand coefficient, the rice low-temperature chilling injury predictor sequence is configured and selected. In this process, a sample environmental low-temperature parameter set, a sample chlorophyll parameter set, and a sample rice low-temperature chilling injury parameter set are collected. Multiple training data are divided to train and obtain the rice low-temperature chilling injury predictor group. The environmental low temperature parameters and chlorophyll parameters are combined and input into the rice low temperature injury predictor sequence, the mean of the output result sequence is calculated, the rice low temperature injury parameters are obtained, and early warning judgment is performed. The rice low temperature injury parameters include the rice low temperature injury level. The chilling injury demand coefficient is configured based on the chlorophyll parameters and the accuracy of remote sensing chlorophyll, including: Obtain the peak values ​​of chlorophyll parameters during rice growth; The ratio of the chlorophyll parameter to the peak chlorophyll parameter is calculated to obtain the first chilling injury coefficient; The second chilling injury coefficient was calculated based on the accuracy of the remote sensing chlorophyll. The cooling injury demand coefficient is calculated based on the first cooling injury coefficient and the second cooling injury coefficient.

2. The method for early warning of low-temperature chilling injury in rice using multi-source meteorological fusion as described in claim 1, characterized in that, Collect remote sensing and meteorological data within the rice-growing area, including: Remote sensing data is collected within the rice-growing area. The remote sensing data includes remote sensing feature data and remote sensing content data. The remote sensing feature data includes remote sensing resolutions for different wavebands. Meteorological data was collected from the rice-growing area, including temperature and humidity.

3. The method for early warning of low-temperature chilling injury in rice using multi-source meteorological fusion as described in claim 1, characterized in that, Based on the remote sensing data, temperature parameters and chlorophyll parameters within the rice paddy area are identified, including: Temperature parameters within the rice paddy area are obtained by processing the thermal infrared radiation values ​​within the remote sensing data. Based on the remote sensing data, the chlorophyll parameters within the rice paddy area are identified.

4. The method for early warning of low-temperature chilling injury in rice using multi-source meteorological fusion according to claim 3, characterized in that, Based on the remote sensing data, chlorophyll parameters within the rice paddy area are identified, including: Based on the remote sensing data, the normalized vegetation index was calculated. The normalized vegetation index is input into the chlorophyll classification table to obtain chlorophyll parameters. The chlorophyll classification table is constructed based on the mapping relationship between the sample normalized vegetation index set and the sample chlorophyll parameter set of the rice area remote sensing content data.

5. The method for early warning of low-temperature chilling injury in rice using multi-source meteorological fusion as described in claim 1, characterized in that, The accuracy of remote sensing temperature and remote sensing chlorophyll content is obtained by inverse calculation based on remote sensing feature data and meteorological data, including: The remote sensing accuracy classifier is invoked, wherein the remote sensing accuracy classifier is constructed based on the sample remote sensing feature dataset, the sample meteorological dataset, the sample remote sensing temperature accuracy set, and the sample remote sensing chlorophyll accuracy set. The sample remote sensing temperature accuracy includes the similarity between the labeled remote sensing temperature and the actual temperature, and the sample remote sensing chlorophyll accuracy includes the similarity between the labeled remote sensing chlorophyll parameters and the actual chlorophyll parameters. The remote sensing feature data and meteorological data are input into the remote sensing accuracy classifier, and the remote sensing temperature accuracy and remote sensing chlorophyll accuracy are output.

6. The method for early warning of low-temperature chilling injury in rice using multi-source meteorological fusion according to claim 1, characterized in that, Based on the remote sensing temperature accuracy and meteorological data, a fused temperature is obtained through fusion processing. Environmental low-temperature parameters are obtained by combining the meteorological humidity data, including: The remote sensing temperature accuracy is used to correct the preset remote sensing temperature weights to obtain the remote sensing temperature weights, and the meteorological temperature weights are determined. The temperature parameters and meteorological temperatures are weighted and calculated using the remote sensing temperature weights and meteorological temperature weights to obtain the fused temperature. The environmental low-temperature parameters are obtained by combining the fused temperature and meteorological humidity.

7. A multi-source meteorological fusion early warning system for low-temperature chilling injury in rice, characterized in that: The system is used to perform the method according to any one of claims 1-6, the system comprising: The data acquisition module is used to collect remote sensing data and meteorological data in the rice-growing area. The meteorological data includes meteorological temperature and meteorological humidity, and the remote sensing data includes remote sensing feature data and remote sensing content data. The accuracy acquisition module is used to identify the temperature parameters and chlorophyll parameters in the rice area based on the remote sensing content data, and to infer the remote sensing temperature accuracy and remote sensing chlorophyll accuracy based on the remote sensing feature data and meteorological data. The environmental low temperature parameter acquisition module is used to obtain a fused temperature by combining the remote sensing temperature accuracy with meteorological data and combining the meteorological humidity to obtain environmental low temperature parameters. The early warning and discrimination module is used to predict and discriminate low-temperature chilling injury parameters of rice based on the environmental low-temperature parameters, chlorophyll parameters, and remote sensing chlorophyll accuracy, including: Configure the chilling injury demand coefficient based on the chlorophyll parameters and remote sensing chlorophyll accuracy. A rice low-temperature chilling injury predictor group is obtained. According to the chilling injury demand coefficient, the rice low-temperature chilling injury predictor sequence is configured and selected. In this process, a sample environmental low-temperature parameter set, a sample chlorophyll parameter set, and a sample rice low-temperature chilling injury parameter set are collected. Multiple training data are divided to train and obtain the rice low-temperature chilling injury predictor group. The environmental low temperature parameters and chlorophyll parameters are combined and input into the rice low temperature injury predictor sequence, the mean of the output result sequence is calculated, the rice low temperature injury parameters are obtained, and early warning judgment is performed. The rice low temperature injury parameters include the rice low temperature injury level. The chilling injury demand coefficient is configured based on the chlorophyll parameters and the accuracy of remote sensing chlorophyll, including: Obtain the peak values ​​of chlorophyll parameters during rice growth; The ratio of the chlorophyll parameter to the peak chlorophyll parameter is calculated to obtain the first chilling injury coefficient; The second chilling injury coefficient was calculated based on the accuracy of the remote sensing chlorophyll. The cooling injury demand coefficient is calculated based on the first cooling injury coefficient and the second cooling injury coefficient.

Images