An UAV-based drought warning system and method for red soil regions

Abstract

The present invention relates to the technical field of drought early warning, and discloses a drone-based drought early warning system and method for red soil regions. The system comprises: a collection layer configured to determine an initial drought early warning value for a monitored area based on historical meteorological data and real-time soil parameters; a judgment layer configured to extract features from real-time meteorological data to obtain real-time meteorological characteristic values; and to determine whether to adjust the initial drought early warning value based on the real-time meteorological characteristic values; a processing layer configured to, when determining that the initial drought early warning value should be adjusted, obtain high-resolution remote sensing images of the monitored area via a drone platform, determine an adjustment coefficient for the initial drought early warning value based on the high-resolution remote sensing images, and obtain a final drought early warning value; and an early warning layer configured to issue a drought early warning for the monitored area based on the final drought early warning value and determine the warning level. The present invention enables accurate early warning of drought conditions in the monitored area.

Description

Technical Field

[0001] The present invention relates to the technical field of drought early warning, and in particular to a drought early warning system and method for red soil areas based on an unmanned aerial vehicle (UAV). Background Art

[0002] Traditional drought early warning methods rely primarily on monitoring data from ground-based observation stations and information provided by meteorological authorities. While this method can provide early warnings to a certain extent, it is often time-consuming and requires significant investment in manpower and resources. More importantly, due to its reliance on fixed observation points and limited meteorological data, this method often struggles with real-time monitoring and is limited in accuracy and coverage, making it difficult to fully and accurately reflect drought conditions.

[0003] Therefore, it is necessary to design a drought early warning system and method based on drones in red soil areas to solve the problems existing in current technologies. Summary of the Invention

[0004] In view of this, the present invention proposes a drought early warning system and method in red soil areas based on drones, aiming to solve the problem that current technology is difficult to reflect drought conditions in real time and accurately.

[0005] In one aspect, the present invention provides a drought early warning system for red soil regions based on drones, comprising:

[0006] A collection layer is configured to collect historical meteorological data and real-time soil parameters corresponding to the area to be monitored; and determine an initial drought warning value for the area to be monitored based on the historical meteorological data and real-time soil parameters;

[0007] a judgment layer configured to collect real-time meteorological data of the area to be monitored, perform feature extraction on the real-time meteorological data, and obtain real-time meteorological characteristic values; and determine whether to adjust the initial drought warning value according to the real-time meteorological characteristic values;

[0008] a processing layer configured to, when determining that the initial drought warning value needs to be adjusted, obtain a high-resolution remote sensing image of the monitored area through an unmanned aerial vehicle platform, determine an adjustment coefficient for the initial drought warning value based on the high-resolution remote sensing image, and obtain a final drought warning value;

[0009] The early warning layer is configured to issue a drought warning to the monitored area according to the final drought warning value and determine the warning level.

[0010] Furthermore, when the collection layer determines the initial drought warning value of the monitored area based on the historical meteorological data and the real-time soil parameters, it includes:

[0011] Performing feature extraction on the historical meteorological data to obtain historical meteorological feature values;

[0012] Analyzing the real-time soil parameters to obtain effective soil water content;

[0013] Determining a basic drought warning value for the area to be monitored based on the historical meteorological characteristic values;

[0014] Determining whether to optimize the basic drought warning value according to the effective soil moisture content;

[0015] If yes, then collecting crop data of the area to be monitored and calculating the crop water stress index based on the crop data;

[0016] The optimization coefficient of the basic drought warning value is determined according to the crop water stress index, and the initial drought warning threshold is obtained.

[0017] Furthermore, when the collection layer determines the basic drought warning value of the area to be monitored based on the historical meteorological characteristic values, it includes:

[0018] The historical meteorological characteristic values include the standard deviation of rainfall in the same period of history, the actual rainfall in history and the average rainfall in the same period of history;

[0019] Calculate the rainfall deficit index based on the historical rainfall standard deviation, historical actual rainfall and historical average rainfall for the same period;

[0020] comparing the rainfall deficit index with a first rainfall deficit index and a second rainfall deficit index, and determining the basic drought warning value according to the comparison result; wherein the first rainfall deficit index is less than the second rainfall deficit index;

[0021] When the rainfall deficit index is less than or equal to the first rainfall deficit index, determining the basic drought warning value to be a first drought warning value;

[0022] When the rainfall deficit index is greater than the first rainfall deficit index and less than or equal to the second rainfall deficit index, determining the basic drought warning value to be a second drought warning value;

[0023] When the rainfall deficit index is greater than the second rainfall deficit index, the basic drought warning value is determined to be a third drought warning value.

[0024] Furthermore, when the collection layer determines whether to optimize the basic drought warning value according to the effective soil moisture content, it includes:

[0025] Comparing the effective soil moisture content with a soil effective soil moisture content threshold, and determining whether to optimize the basic drought warning value based on the comparison result;

[0026] When the effective soil moisture content is within the effective soil moisture content threshold, determining not to optimize the basic drought warning value, and using the basic drought warning value as the initial drought warning value;

[0027] When the effective soil moisture content is outside the effective soil moisture content threshold, it is determined that the basic drought warning value is to be optimized.

[0028] Furthermore, when the collection layer determines the optimization coefficient of the basic drought warning value according to the crop water stress index and obtains the initial drought warning threshold, it includes:

[0029] comparing the crop water stress index with a first crop water stress index and a second crop water stress index, and determining an optimization coefficient of the basic drought warning value according to the comparison results; wherein the first crop water stress index is less than the second crop water stress index;

[0030] When the crop water stress index is less than or equal to the first crop water stress index, determining the optimization coefficient to be a first optimization coefficient;

[0031] When the crop water stress index is greater than the first crop water stress index and less than or equal to the second crop water stress index, determining the optimization coefficient to be a second optimization coefficient;

[0032] When the crop water stress index is greater than the second crop water stress index, determining the optimization coefficient to be a third optimization coefficient;

[0033] The product of the optimization coefficient and the basic drought warning value is used as the initial drought warning threshold.

[0034] Furthermore, when the judgment layer judges whether to adjust the initial drought warning value according to the real-time meteorological characteristic value, it includes:

[0035] Obtain the real-time meteorological standard value corresponding to the real-time meteorological characteristic value;

[0036] Calculating the difference between the real-time meteorological standard value and the real-time meteorological characteristic value, and recording it as the meteorological difference;

[0037] Comparing the meteorological difference with a meteorological difference threshold, and determining whether to adjust the initial drought warning value according to the comparison result;

[0038] When the meteorological difference is within the meteorological difference threshold, determining not to adjust the initial drought warning value, and using the initial drought warning value as the final drought warning value;

[0039] When the meteorological difference is outside the meteorological difference threshold, it is determined that the initial drought warning value is to be adjusted.

[0040] Furthermore, when the processing layer determines the adjustment coefficient of the initial drought warning value based on the high-resolution remote sensing image and obtains the final drought warning value, it includes:

[0041] Performing image recognition on the high-resolution remote sensing image to obtain terrain slope information of the area to be monitored;

[0042] Dividing the area to be monitored into a plurality of monitoring sub-areas, and extracting a terrain slope characteristic value in each of the monitoring sub-areas;

[0043] Calculating the average value of the terrain slope of the area to be monitored based on all the terrain slope characteristic values;

[0044] Collecting the frequency of crop irrigation in the area to be monitored;

[0045] constructing an adjustment feature set based on the average terrain slope and the crop irrigation frequency;

[0046] Comparing the adjustment feature set with a historical adjustment set, and determining an adjustment coefficient for the initial drought warning value based on the comparison result;

[0047] When there is a historical adjustment feature set identical to the adjustment feature set in the historical adjustment group, using a historical adjustment coefficient corresponding to the historical adjustment feature set as the adjustment coefficient, and using a product of the adjustment coefficient and the initial drought warning value as the final drought warning value;

[0048] When there is no historical adjustment feature set identical to the adjustment feature set in the historical adjustment group, the correlation between the adjustment feature set and the historical adjustment group is calculated based on the Euclidean distance, and the maximum correlation is extracted; the adjustment coefficient of the initial drought warning value is determined according to the maximum correlation, and the product of the adjustment coefficient and the initial drought warning value is used as the final drought warning value.

[0049] Furthermore, when the processing layer determines the adjustment coefficient of the initial drought warning value according to the maximum correlation, it includes:

[0050] Comparing the maximum correlation with a preset adjustment coefficient mapping table to obtain a target adjustment coefficient that matches the maximum correlation;

[0051] The target adjustment coefficient is used as the adjustment coefficient.

[0052] Furthermore, the early warning layer performs drought warning on the monitored area according to the final drought warning value and determines the warning level, including:

[0053] Comparing the final drought warning value with a drought warning level threshold, and determining a warning level for the monitored area based on the comparison result; wherein the drought warning threshold includes a first drought warning level threshold, a second drought warning level threshold, and a third drought warning level threshold;

[0054] When the final drought warning value is within the first drought warning level threshold, determining the warning level of the monitored area as a first level warning;

[0055] When the final drought warning value is within the second drought warning level threshold, determining the warning level of the monitored area to be a second level warning;

[0056] When the final drought warning value is within the third drought warning level threshold, the warning level of the to-be-monitored area is determined to be the third level warning.

[0057] Compared with the existing technology, the beneficial effects of the present invention are as follows: the drone-based red soil area drought early warning system provided by the present invention can integrate historical meteorological data, real-time soil parameters and real-time meteorological data, and through multi-level analysis and processing, realize accurate early warning of drought conditions in the monitored area; the system not only takes into account the key factor of soil moisture content, but also combines historical rainfall conditions and real-time meteorological conditions, making drought early warning more comprehensive and scientific; the high-resolution remote sensing images obtained by the drone platform further improve the accuracy and timeliness of the early warning; the system can also automatically determine the warning level according to the warning results, and provide timely and reliable drought early warning information to relevant departments, which helps to take measures in advance and reduce losses caused by drought.

[0058] In another aspect, the present invention also proposes a drought early warning method for red soil areas based on drones, comprising the following steps:

[0059] Collecting historical meteorological data and real-time soil parameters corresponding to the area to be monitored; determining an initial drought warning value for the area to be monitored based on the historical meteorological data and real-time soil parameters;

[0060] collecting real-time meteorological data of the area to be monitored, and performing feature extraction on the real-time meteorological data to obtain real-time meteorological characteristic values; and determining whether to adjust the initial drought warning value based on the real-time meteorological characteristic values;

[0061] When it is determined that the initial drought warning value needs to be adjusted, a high-resolution remote sensing image of the area to be monitored is obtained through an unmanned aerial vehicle platform, an adjustment coefficient of the initial drought warning value is determined based on the high-resolution remote sensing image, and a final drought warning value is obtained;

[0062] A drought warning is performed on the monitored area according to the final drought warning value, and a warning level is determined.

[0063] It is understandable that the above-mentioned drone-based red soil drought early warning system and method have the same beneficial effects and will not be repeated here. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] Various other advantages and benefits will become apparent to those skilled in the art upon reading the detailed description of the preferred embodiment below. The accompanying drawings are for illustration purposes only and are not to be considered as limiting the present invention. The same reference symbols are used throughout the drawings to represent the same components. In the drawings:

[0065] Figure 1 A structural block diagram of a drought early warning system for red soil regions based on drones provided in an embodiment of the present invention;

[0066] Figure 2 This is a flow chart of a method for early warning of drought in red soil areas based on drones provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0067] The exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art. It should be noted that, unless there is a conflict, the embodiments of the present disclosure and the features in the embodiments can be combined with each other. The present invention will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0068] See Figure 1 As shown, in some embodiments of the present application, this embodiment provides a drought early warning system for red soil areas based on drones, including:

[0069] A collection layer is configured to collect historical meteorological data and real-time soil parameters corresponding to the area to be monitored; and determine an initial drought warning value for the area to be monitored based on the historical meteorological data and real-time soil parameters;

[0070] a judgment layer configured to collect real-time meteorological data of the area to be monitored, perform feature extraction on the real-time meteorological data, and obtain real-time meteorological characteristic values; and determine whether to adjust the initial drought warning value according to the real-time meteorological characteristic values;

[0071] a processing layer configured to, when determining that the initial drought warning value needs to be adjusted, obtain a high-resolution remote sensing image of the monitored area through an unmanned aerial vehicle platform, determine an adjustment coefficient for the initial drought warning value based on the high-resolution remote sensing image, and obtain a final drought warning value;

[0072] The early warning layer is configured to issue a drought warning to the monitored area according to the final drought warning value and determine the warning level.

[0073] In this embodiment, the high-resolution remote sensing image refers to a remote sensing image with a resolution of 5 meters or less.

[0074] It can be understood that the drone-based red soil area drought early warning system provided in this embodiment can integrate historical meteorological data, real-time soil parameters and real-time meteorological data, and through multi-level analysis and processing, realize accurate early warning of drought conditions in the monitored area; the system not only takes into account the key factor of soil moisture content, but also combines historical rainfall conditions and real-time meteorological conditions, making drought early warning more comprehensive and scientific; the high-resolution remote sensing images obtained by the drone platform further improve the accuracy and timeliness of the early warning; the system can also automatically determine the warning level according to the warning results, and provide timely and reliable drought early warning information to relevant departments, which helps to take measures in advance and reduce losses caused by drought.

[0075] Specifically, when the collection layer determines the initial drought warning value of the monitored area based on the historical meteorological data and real-time soil parameters, it includes:

[0076] Performing feature extraction on the historical meteorological data to obtain historical meteorological feature values;

[0077] Analyzing the real-time soil parameters to obtain effective soil water content;

[0078] Determining a basic drought warning value for the area to be monitored based on the historical meteorological characteristic values;

[0079] Determining whether to optimize the basic drought warning value according to the effective soil moisture content;

[0080] If yes, then collecting crop data of the area to be monitored and calculating the crop water stress index based on the crop data;

[0081] The optimization coefficient of the basic drought warning value is determined according to the crop water stress index, and the initial drought warning threshold is obtained.

[0082] In this embodiment, the effective soil moisture content is the difference between the field holding capacity and the wilting point moisture content; the field holding capacity refers to the maximum effective moisture content that the soil can maintain in the monitored area, and the wilting point moisture content refers to the soil moisture content when crops in the monitored area begin to wilt due to water shortage.

[0083] It is understood that by introducing the crop water stress index, the drought early warning system of this embodiment can more accurately reflect the drought conditions in the monitored area, improving the accuracy of early warnings. Furthermore, the system can gradually improve its early warning capabilities and accuracy through continuous learning and algorithm optimization, providing more reliable technical support for agricultural production.

[0084] Specifically, when the collection layer determines the basic drought warning value of the monitored area according to the historical meteorological characteristic value, it includes:

[0085] The historical meteorological characteristic values include the standard deviation of rainfall in the same period of history, the actual rainfall in history and the average rainfall in the same period of history;

[0086] Calculate the rainfall deficit index based on the historical rainfall standard deviation, historical actual rainfall and historical average rainfall for the same period;

[0087] comparing the rainfall deficit index with a first rainfall deficit index and a second rainfall deficit index, and determining the basic drought warning value according to the comparison result; wherein the first rainfall deficit index is less than the second rainfall deficit index;

[0088] When the rainfall deficit index is less than or equal to the first rainfall deficit index, determining the basic drought warning value to be a first drought warning value;

[0089] When the rainfall deficit index is greater than the first rainfall deficit index and less than or equal to the second rainfall deficit index, determining the basic drought warning value to be a second drought warning value;

[0090] When the rainfall deficit index is greater than the second rainfall deficit index, the basic drought warning value is determined to be a third drought warning value.

[0091] In this embodiment, the first drought warning value is less than the second drought warning value and less than the third drought warning value.

[0092] In this embodiment, the calculation formula of the rainfall deficit index is:

[0093] ;

[0094] Among them, RDI represents the rainfall deficit index, P represents the average rainfall in the same historical period, Pact represents the actual rainfall in history, and σP represents the standard deviation of rainfall in the same historical period.

[0095] In this embodiment, the first rainfall deficit index and the second rainfall deficit index are obtained by statistically analyzing the correspondence between historical rainfall data and drought occurrence in the monitored area. Specifically, the rainfall deficit indices (RDIs) for multiple historical periods are first calculated and labeled in combination with the actual drought levels of the historical periods. Cluster analysis or distribution analysis is then used to identify the boundary between the rainfall deficit index (RDI) and the drought level, thereby determining the critical values (the first rainfall deficit index and the second rainfall deficit index) corresponding to different drought levels, which serve as a quantitative basis for drought warning classification.

[0096] Specifically, when the collection layer determines whether to optimize the basic drought warning value according to the effective soil moisture content, it includes:

[0097] Comparing the effective soil moisture content with a soil effective soil moisture content threshold, and determining whether to optimize the basic drought warning value based on the comparison result;

[0098] When the effective soil moisture content is within the effective soil moisture content threshold, determining not to optimize the basic drought warning value, and using the basic drought warning value as the initial drought warning value;

[0099] When the effective soil moisture content is outside the effective soil moisture content threshold, it is determined that the basic drought warning value is to be optimized.

[0100] In this embodiment, the effective soil moisture content threshold is preferably 70% of the field holding capacity.

[0101] It can be understood that when the effective soil moisture content is within the threshold, it means that the soil moisture condition is good and there is no need to optimize the basic drought warning value; when the effective soil moisture content is lower or higher than the threshold, it indicates that the soil moisture condition may be unfavorable for crop growth. At this time, the basic drought warning value needs to be optimized to more accurately reflect the drought situation in the monitored area.

[0102] Specifically, when the collection layer calculates the crop water stress index based on the crop data, it includes:

[0103] Analyzing the crop data to obtain a temperature characteristic value of the crop canopy in the area to be monitored;

[0104] Collecting the crop wet reference temperature and the crop dry reference temperature of the area to be monitored;

[0105] Calculating the crop water stress index according to the crop canopy temperature characteristic value, the crop wet reference temperature, and the crop dry reference temperature;

[0106] The crop water stress index is obtained by the following formula:

[0107] ;

[0108] Among them, CWSI represents the crop water stress index; Tcanopy represents the characteristic value of crop canopy temperature; Twet represents the crop wet reference temperature; and Tdry represents the crop dry reference temperature.

[0109] The Crop Water Stress Index (CWSI) is an indicator of crop water stress, calculated by comparing the difference between crop canopy temperature and the moist reference temperature and the dry reference temperature. When a crop is experiencing water stress, its canopy temperature rises, approaching or exceeding the dry reference temperature. At this point, the CWSI is high, indicating a severe drought. Conversely, when the crop is well-watered, its canopy temperature approaches the moist reference temperature, and the CWSI is low.

[0110] Specifically, when the collection layer determines the optimization coefficient of the basic drought warning value according to the crop water stress index and obtains the initial drought warning threshold, it includes:

[0111] comparing the crop water stress index with a first crop water stress index and a second crop water stress index, and determining an optimization coefficient of the basic drought warning value according to the comparison results; wherein the first crop water stress index is less than the second crop water stress index;

[0112] When the crop water stress index is less than or equal to the first crop water stress index, determining the optimization coefficient to be a first optimization coefficient;

[0113] When the crop water stress index is greater than the first crop water stress index and less than or equal to the second crop water stress index, determining the optimization coefficient to be a second optimization coefficient;

[0114] When the crop water stress index is greater than the second crop water stress index, determining the optimization coefficient to be a third optimization coefficient;

[0115] The product of the optimization coefficient and the basic drought warning value is used as the initial drought warning threshold.

[0116] In this embodiment, the first crop water stress index and the second crop water stress index are used to classify different water stress levels, thereby determining the key thresholds of the optimization coefficient of the basic drought warning value. The method for obtaining them is: based on the historical crop canopy temperature data of the monitored area, combined with the wet reference temperature and dry reference temperature of the corresponding period, the crop water stress index of the historical period is calculated, and the data is labeled in combination with the degree of drought that has occurred in the current period; then, through cluster analysis or probability distribution analysis methods, the dividing point between the CWSI value and the drought level or crop stress state is identified, thereby determining the CWSI critical values corresponding to different water stress levels, respectively serving as the first and second crop water stress indices, for correspondingly selecting different optimization coefficients, and then correcting the basic drought warning value to improve the accuracy and responsiveness of the warning.

[0117] In this embodiment, the magnitude relationship among the first optimization coefficient, the second optimization coefficient, and the third optimization coefficient is: first optimization coefficient < second optimization coefficient < third optimization coefficient.

[0118] It's understandable that when crop water stress is low, a smaller optimization coefficient (the first optimization coefficient) is selected, making the initial drought warning value relatively conservative and avoiding false alarms. As crop water stress increases, the optimization coefficient gradually increases, making the initial drought warning value more sensitive and able to promptly reflect the drought conditions in the monitored area. In this way, the system can dynamically adjust the sensitivity of drought warnings based on the degree of crop water stress, improving the accuracy and practicality of warnings.

[0119] Specifically, when the judgment layer judges whether to adjust the initial drought warning value according to the real-time meteorological characteristic value, it includes:

[0120] Obtain the real-time meteorological standard value corresponding to the real-time meteorological characteristic value;

[0121] Calculating the difference between the real-time meteorological standard value and the real-time meteorological characteristic value, and recording it as the meteorological difference;

[0122] Comparing the meteorological difference with a meteorological difference threshold, and determining whether to adjust the initial drought warning value according to the comparison result;

[0123] When the meteorological difference is within the meteorological difference threshold, determining not to adjust the initial drought warning value, and using the initial drought warning value as the final drought warning value;

[0124] When the meteorological difference is outside the meteorological difference threshold, it is determined that the initial drought warning value is to be adjusted.

[0125] In this embodiment, the real-time meteorological characteristic values include real-time rainfall, real-time temperature, real-time humidity, and real-time wind speed.

[0126] In this embodiment, the real-time meteorological characteristic value is preferably real-time rainfall.

[0127] It's understandable that real-time rainfall is a key factor influencing drought conditions. By comparing real-time rainfall with the real-time meteorological standard value, we can determine whether current meteorological conditions are affecting the drought warning value. When real-time rainfall approaches or reaches the real-time meteorological standard value, it indicates that current meteorological conditions are relatively favorable and no adjustment to the initial drought warning value is necessary. However, when real-time rainfall is far below the real-time meteorological standard value, it indicates that current meteorological conditions may exacerbate drought conditions, and the initial drought warning value needs to be adjusted to more accurately reflect the drought risk in the monitored area.

[0128] Specifically, when the processing layer determines the adjustment coefficient of the initial drought warning value based on the high-resolution remote sensing image and obtains the final drought warning value, it includes:

[0129] Performing image recognition on the high-resolution remote sensing image to obtain terrain slope information of the area to be monitored;

[0130] Dividing the area to be monitored into a plurality of monitoring sub-areas, and extracting a terrain slope characteristic value in each of the monitoring sub-areas;

[0131] Calculating the average value of the terrain slope of the area to be monitored based on all the terrain slope characteristic values;

[0132] Collecting the frequency of crop irrigation in the area to be monitored;

[0133] constructing an adjustment feature set based on the average terrain slope and the crop irrigation frequency;

[0134] Comparing the adjustment feature set with a historical adjustment set, and determining an adjustment coefficient for the initial drought warning value based on the comparison result;

[0135] When there is a historical adjustment feature set identical to the adjustment feature set in the historical adjustment group, using a historical adjustment coefficient corresponding to the historical adjustment feature set as the adjustment coefficient, and using a product of the adjustment coefficient and the initial drought warning value as the final drought warning value;

[0136] When there is no historical adjustment feature set identical to the adjustment feature set in the historical adjustment group, the correlation between the adjustment feature set and the historical adjustment group is calculated based on the Euclidean distance, and the maximum correlation is extracted; the adjustment coefficient of the initial drought warning value is determined according to the maximum correlation, and the product of the adjustment coefficient and the initial drought warning value is used as the final drought warning value.

[0137] In this embodiment, the terrain slope characteristic value refers to the terrain slope of each monitoring sub-area in the area to be monitored, and is used to reflect the terrain undulation of the area to be monitored.

[0138] In this embodiment, the adjustment feature set is represented as {average terrain slope, crop irrigation frequency}, denoted by {S, I}. S represents the average terrain slope of the monitored area, and I represents the crop irrigation frequency of the monitored area. The historical adjustment set is represented as {historical average terrain slope, historical crop irrigation frequency; historical adjustment coefficient}, denoted by {HS, HI; Ach}.

[0139] It can be understood that by comparing the current adjustment feature set with the feature sets in the historical adjustment group, the system can find the historical adjustment case that is most similar to the current situation, thereby determining a more reasonable adjustment coefficient. If there is no exactly identical adjustment feature set in the historical adjustment group, the system calculates the correlation to extract the maximum correlation, ensuring that the determination of the adjustment coefficient is scientific and accurate. In this way, the system can comprehensively consider factors such as terrain slope and crop irrigation frequency to make more precise adjustments to the initial drought warning value, further improving the pertinence and practicality of drought warnings.

[0140] In this embodiment, the correlation is obtained by the following formula:

[0141] ;

[0142] Among them, Ri represents the correlation between the current adjustment feature set and the i-th historical adjustment feature set in the historical adjustment group; S represents the average terrain slope of the area to be monitored; HSi represents the historical average terrain slope of the i-th historical adjustment feature set in the historical adjustment group; I represents the crop irrigation frequency of the area to be monitored; HIi represents the historical crop irrigation frequency of the i-th historical adjustment feature set in the historical adjustment group; ε represents an extremely small positive number used to avoid the denominator being zero.

[0143] Specifically, when the processing layer determines the adjustment coefficient of the initial drought warning value according to the maximum correlation, it includes:

[0144] Comparing the maximum correlation with a preset adjustment coefficient mapping table to obtain a target adjustment coefficient that matches the maximum correlation;

[0145] The target adjustment coefficient is used as the adjustment coefficient.

[0146] In this embodiment, a preset adjustment coefficient mapping table establishes a correspondence between maximum correlations and adjustment coefficients, guiding the system to determine appropriate adjustment coefficients based on the current situation. By consulting this table, the system can quickly and accurately find the target adjustment coefficient that best matches the current set of adjustment features, thereby enabling fine-tuning of the initial drought warning value.

[0147] In this embodiment, when the maximum correlation is 0.98, the corresponding target adjustment coefficient is 1.2; when the maximum correlation is 0.85, the corresponding target adjustment coefficient is 0.98; when the maximum correlation is 0.72, the corresponding target adjustment coefficient is 0.8.

[0148] It's understandable that in the preset adjustment coefficient mapping table, each maximum correlation corresponds to an adjustment coefficient, reflecting the reasonable range of adjustment coefficient values under different correlation conditions. In this way, the system can quickly determine the adjustment coefficient based on the maximum correlation, ensuring the accuracy and reliability of the final drought warning value. In actual applications, the system can also adjust and optimize the preset adjustment coefficient mapping table based on user needs and actual conditions to improve the pertinence and practicality of the warning.

[0149] Specifically, the early warning layer performs drought early warning on the monitored area according to the final drought early warning value and determines the early warning level, including:

[0150] Comparing the final drought warning value with a drought warning level threshold, and determining a warning level for the monitored area based on the comparison result; wherein the drought warning threshold includes a first drought warning level threshold, a second drought warning level threshold, and a third drought warning level threshold;

[0151] When the final drought warning value is within the first drought warning level threshold, determining the warning level of the monitored area as a first level warning;

[0152] When the final drought warning value is within the second drought warning level threshold, determining the warning level of the monitored area to be a second level warning;

[0153] When the final drought warning value is within the third drought warning level threshold, the warning level of the to-be-monitored area is determined to be the third level warning.

[0154] In this embodiment, the first, second, and third drought warning level thresholds refer to warning threshold ranges corresponding to different degrees of drought. Specifically, the first drought warning level threshold corresponds to a mild drought. While the drought severity in the monitored area is relatively low, it still warrants attention and preventive measures. The second drought warning level threshold corresponds to a moderate drought. Drought severity in the monitored area is already significant, requiring appropriate drought mitigation measures to mitigate its impact on crop growth. The third drought warning level threshold corresponds to a severe drought. Drought severity in the monitored area is extremely severe, requiring immediate activation of emergency drought mitigation plans to ensure that crop growth is not severely impacted.

[0155] As you can see, this approach allows the system to provide more accurate drought warning information for monitored areas, tailored to varying degrees of drought severity. This helps relevant departments and farmers take timely drought mitigation measures, mitigating the impact of drought on agricultural production. Furthermore, the system can continuously optimize and adjust drought warning thresholds based on historical drought data and crop growth conditions, improving the accuracy and practicality of warnings.

[0156] See Figure 2 As shown, in some embodiments of the present application, this embodiment provides a drought early warning method for red soil areas based on a drone, comprising the following steps:

[0157] S100: Collecting historical meteorological data and real-time soil parameters corresponding to the area to be monitored; determining an initial drought warning value for the area to be monitored based on the historical meteorological data and real-time soil parameters;

[0158] S200: collecting real-time meteorological data of the area to be monitored, and performing feature extraction on the real-time meteorological data to obtain real-time meteorological characteristic values; and determining whether to adjust the initial drought warning value based on the real-time meteorological characteristic values;

[0159] S300: When it is determined that the initial drought warning value needs to be adjusted, a high-resolution remote sensing image of the area to be monitored is acquired through an unmanned aerial vehicle platform, an adjustment coefficient of the initial drought warning value is determined based on the high-resolution remote sensing image, and a final drought warning value is obtained;

[0160] S400: Performing a drought warning for the area to be monitored according to the final drought warning value, and determining a warning level.

[0161] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware. Furthermore, the present application may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0162] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems) and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0163] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0164] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0165] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered by the scope of protection of the claims of the present invention.

Claims

1. A drought early warning system for red soil areas based on drones, characterized by: include: The collection layer is configured to collect historical meteorological data and real-time soil parameters corresponding to the area to be monitored; Determining an initial drought warning value for the area to be monitored based on the historical meteorological data and real-time soil parameters; The judgment layer is configured to collect real-time meteorological data of the area to be monitored, and perform feature extraction on the real-time meteorological data to obtain real-time meteorological feature values; Determining whether to adjust the initial drought warning value according to the real-time meteorological characteristic value; a processing layer configured to, when determining that the initial drought warning value needs to be adjusted, obtain a high-resolution remote sensing image of the monitored area through an unmanned aerial vehicle platform, determine an adjustment coefficient for the initial drought warning value based on the high-resolution remote sensing image, and obtain a final drought warning value; an early warning layer, configured to issue a drought early warning to the monitored area according to the final drought early warning value and determine an early warning level; When the processing layer determines the adjustment coefficient of the initial drought warning value based on the high-resolution remote sensing image and obtains the final drought warning value, it includes: Performing image recognition on the high-resolution remote sensing image to obtain terrain slope information of the area to be monitored; Dividing the area to be monitored into a plurality of monitoring sub-areas, and extracting a terrain slope characteristic value in each of the monitoring sub-areas; Calculating the average value of the terrain slope of the area to be monitored based on all the terrain slope characteristic values; Collecting the frequency of crop irrigation in the area to be monitored; constructing an adjustment feature set based on the average terrain slope and the crop irrigation frequency; Comparing the adjustment feature set with a historical adjustment set, and determining an adjustment coefficient for the initial drought warning value based on the comparison result; When there is a historical adjustment feature set identical to the adjustment feature set in the historical adjustment group, using a historical adjustment coefficient corresponding to the historical adjustment feature set as the adjustment coefficient, and using a product of the adjustment coefficient and the initial drought warning value as the final drought warning value; When there is no historical adjustment feature set identical to the adjustment feature set in the historical adjustment group, calculating the correlation between the adjustment feature set and the historical adjustment group based on the Euclidean distance, and extracting the maximum correlation; determining an adjustment coefficient for the initial drought warning value based on the maximum correlation, and using the product of the adjustment coefficient and the initial drought warning value as the final drought warning value; When the processing layer determines the adjustment coefficient of the initial drought warning value according to the maximum correlation, it includes: Comparing the maximum correlation with a preset adjustment coefficient mapping table to obtain a target adjustment coefficient that matches the maximum correlation; The target adjustment coefficient is used as the adjustment coefficient.

2. The UAV-based red soil drought early warning system according to claim 1, characterized in that: When the collection layer determines the initial drought warning value of the monitored area based on the historical meteorological data and the real-time soil parameters, it includes: Performing feature extraction on the historical meteorological data to obtain historical meteorological feature values; Analyzing the real-time soil parameters to obtain effective soil water content; Determining a basic drought warning value for the area to be monitored based on the historical meteorological characteristic values; Determining whether to optimize the basic drought warning value according to the effective soil moisture content; If yes, then collecting crop data of the area to be monitored and calculating the crop water stress index based on the crop data; An optimization coefficient of the basic drought warning value is determined according to the crop water stress index, and the initial drought warning value is obtained.

3. The UAV-based red soil drought early warning system according to claim 2, characterized in that: When the collection layer determines the basic drought warning value of the monitored area according to the historical meteorological characteristic value, it includes: The historical meteorological characteristic values include the standard deviation of rainfall in the same period of history, the actual rainfall in history and the average rainfall in the same period of history; Calculate the rainfall deficit index based on the historical rainfall standard deviation, historical actual rainfall and historical average rainfall for the same period; comparing the rainfall deficit index with a first rainfall deficit index and a second rainfall deficit index, and determining the basic drought warning value according to the comparison result; wherein the first rainfall deficit index is less than the second rainfall deficit index; When the rainfall deficit index is less than or equal to the first rainfall deficit index, determining the basic drought warning value to be a first drought warning value; When the rainfall deficit index is greater than the first rainfall deficit index and less than or equal to the second rainfall deficit index, determining the basic drought warning value to be a second drought warning value; When the rainfall deficit index is greater than the second rainfall deficit index, the basic drought warning value is determined to be a third drought warning value.

4. The UAV-based red soil drought early warning system according to claim 3, characterized in that: When the collection layer determines whether to optimize the basic drought warning value according to the effective soil moisture content, it includes: Comparing the effective soil moisture content with a soil effective soil moisture content threshold, and determining whether to optimize the basic drought warning value based on the comparison result; When the effective soil moisture content is within the effective soil moisture content threshold, determining not to optimize the basic drought warning value, and using the basic drought warning value as the initial drought warning value; When the effective soil moisture content is outside the effective soil moisture content threshold, it is determined that the basic drought warning value is to be optimized.

5. The UAV-based red soil drought early warning system according to claim 4, characterized in that: When the acquisition layer determines the optimization coefficient of the basic drought warning value according to the crop water stress index and obtains the initial drought warning value, it includes: comparing the crop water stress index with a first crop water stress index and a second crop water stress index, and determining an optimization coefficient of the basic drought warning value according to the comparison results; wherein the first crop water stress index is less than the second crop water stress index; When the crop water stress index is less than or equal to the first crop water stress index, determining the optimization coefficient to be a first optimization coefficient; When the crop water stress index is greater than the first crop water stress index and less than or equal to the second crop water stress index, determining the optimization coefficient to be a second optimization coefficient; When the crop water stress index is greater than the second crop water stress index, determining the optimization coefficient to be a third optimization coefficient; The product of the optimization coefficient and the basic drought warning value is used as the initial drought warning value.

6. The UAV-based red soil drought early warning system according to claim 5, characterized in that: When the judgment layer judges whether to adjust the initial drought warning value according to the real-time meteorological characteristic value, it includes: Obtain the real-time meteorological standard value corresponding to the real-time meteorological characteristic value; Calculating the difference between the real-time meteorological standard value and the real-time meteorological characteristic value, and recording it as the meteorological difference; Comparing the meteorological difference with a meteorological difference threshold, and determining whether to adjust the initial drought warning value according to the comparison result; When the meteorological difference is within the meteorological difference threshold, determining not to adjust the initial drought warning value, and using the initial drought warning value as the final drought warning value; When the meteorological difference is outside the meteorological difference threshold, it is determined that the initial drought warning value is to be adjusted.

7. The UAV-based red soil drought early warning system according to claim 6, characterized in that: When the early warning layer performs drought early warning on the monitored area according to the final drought early warning value and determines the early warning level, it includes: Comparing the final drought warning value with a drought warning level threshold, and determining a warning level for the monitored area based on the comparison result; wherein the drought warning threshold includes a first drought warning level threshold, a second drought warning level threshold, and a third drought warning level threshold; When the final drought warning value is within the first drought warning level threshold, determining the warning level of the monitored area as a first level warning; When the final drought warning value is within the second drought warning level threshold, determining the warning level of the monitored area to be a second level warning; When the final drought warning value is within the third drought warning level threshold, the warning level of the to-be-monitored area is determined to be the third level warning.

8. A method for early warning of drought in red soil regions based on drones, applied to the early warning system for drought in red soil regions based on drones according to any one of claims 1 to 7, characterized in that: include: Collect historical meteorological data and real-time soil parameters corresponding to the area to be monitored; Determining an initial drought warning value for the area to be monitored based on the historical meteorological data and real-time soil parameters; Collecting real-time meteorological data of the area to be monitored, and performing feature extraction on the real-time meteorological data to obtain real-time meteorological feature values; Determining whether to adjust the initial drought warning value according to the real-time meteorological characteristic value; When it is determined that the initial drought warning value needs to be adjusted, a high-resolution remote sensing image of the area to be monitored is obtained through an unmanned aerial vehicle platform, an adjustment coefficient of the initial drought warning value is determined based on the high-resolution remote sensing image, and a final drought warning value is obtained; A drought warning is performed on the monitored area according to the final drought warning value, and a warning level is determined.

Images