Crop water stress diagnosis method and system based on fuzzy discrimination theory and multi-source information fusion

Abstract

The invention provides a crop moisture stress diagnosis method and system based on a fuzzy discrimination theory and multi-source information fusion, and the method comprises the steps: collecting multi-source monitoring data which comprises but is not limited to soil moisture, crop physiology, remote sensing images and meteorological parameters; preprocessing the data, and calculating at least one moisture related index; constructing a membership function of the indexes based on a fuzzy set theory; fusing the indexes by adopting a fuzzy comprehensive discrimination method to obtain a comprehensive moisture stress index; and outputting a mild, moderate or severe crop water stress level according to the index. The system comprises a data acquisition, data processing, fuzzy discrimination and result display modular structure according to the method. By fusing multi-source monitoring data and adopting a fuzzy set theory and a fuzzy comprehensive discrimination method, accurate diagnosis and graded output of crop water stress are realized, the utilization efficiency of water resources is improved, and the method has remarkable water-saving and efficiency-increasing values.

Description

TECHNICAL FIELD

[0001] The application belongs to the field of agricultural informatization and intelligent irrigation, and particularly relates to a crop water stress diagnosis method and system based on fuzzy discrimination theory and multi-source information fusion. BACKGROUND

[0002] Crops are extremely sensitive to water conditions during growth, and if they are in a state of water deficiency or excessive irrigation for a long time, it will lead to yield reduction, quality reduction and waste of water resources. Therefore, how to accurately and timely diagnose the degree of crop water stress is an important problem in agricultural modernization management and efficient use of water resources.

[0003] The prior art mainly includes the following types: 1. Monitoring method based on soil moisture sensor: soil water content is obtained by laying tension meter, volumetric water content sensor, etc. This method responds quickly, but can only reflect local point information, has poor spatial representativeness, and is difficult to cover the entire field.

[0004] 2. Inversion method based on remote sensing vegetation index: NDVI, GNDVI and other vegetation indexes are calculated by using unmanned aerial vehicle or satellite image to judge the crop water condition. This method has a large coverage range, but is limited by imaging conditions and time, has insufficient data continuity, and there is redundancy and noise between indexes.

[0005] 3. Discrimination method based on single threshold: soil water content or vegetation index threshold is set to divide the water stress grade. This method is simple to implement, but it is difficult to reflect the fuzziness and gradualness of crop water state, and the diagnosis result is prone to misjudgment.

[0006] In summary, the prior art still has the following problems in diagnosing crop water stress: single data source, poor stability of diagnosis result; scale difference and noise between data, direct use leading to result deviation; lack of effective processing of gradualness and uncertainty of crop water stress. Therefore, there is an urgent need for a crop water stress diagnosis method and system that can fuse multi-source monitoring data and solve the problem of uncertainty. SUMMARY

[0007] Therefore, the present application provides a crop water stress diagnosis method and system based on fuzzy discrimination theory and multi-source information fusion, which realizes accurate diagnosis and grading output of crop water stress by fusing multi-source monitoring data and using fuzzy set theory and fuzzy comprehensive discrimination method.

[0008] In the first aspect, the present application provides a crop water stress diagnosis method based on fuzzy discrimination theory and multi-source information fusion, comprising the following steps: Collecting multi-source data: the multi-source data includes soil moisture data, crop physiological data, remote sensing image data, and meteorological parameter data; According to the multi-source data, data preprocessing and index calculation are implemented: the data preprocessing and index calculation include remote sensing image preprocessing, data quality control, calculation of typical indexes related to crop water status, and normalization processing; According to the fuzzy set theory, different membership functions are established for different indexes to describe the relationship between a water-related index and the degree of crop water stress; According to the membership function, fuzzy comprehensive discrimination is performed on the water membership value to obtain a comprehensive water stress index; According to the comprehensive water stress index, a mild or moderate or severe crop water stress level is output.

[0009] Further, the soil moisture data collection method is: collecting the water content of 0-20 cm, 20-40 cm, and 40-60 cm soil layers in the crop root zone 0-60 cm; The crop physiological data collection method is: periodically measuring leaf area index, leaf water content, and plant height; The remote sensing image data collection method is: using multi-spectral and thermal infrared data to obtain high-resolution images at the field scale; The meteorological parameter data collection method is: collecting air temperature, humidity, wind speed, rainfall, and solar radiation, and further calculating the vapor pressure difference.

[0010] Further, the remote sensing image preprocessing specifically includes: Geometric correction to eliminate spatial distortion; Radiometric calibration is performed through standard reflectance calibration panels and camera calibration coefficients to obtain normalized reflectance; Multiple images are spliced to generate a complete field orthographic image; The data quality control specifically includes: IQR method is used to remove outliers; Linear interpolation or Kriging method is used to fill in missing values; Soil moisture and meteorological data are time-synchronized; The calculation of typical indexes related to crop water status specifically includes: normalized difference vegetation index (NDVI), green normalized difference vegetation index (GNDVI), crop water stress index (CWSI), and vapor pressure difference (VPD).

[0011] Further, the membership function construction method is: Threshold interval setting: according to historical test data and expert experience, the mild, moderate, and severe intervals of water content, NDVI, CWSI, and VPD are set respectively; Function form selection: select trigonometric function or trapezoidal function or Gaussian function.

[0012] Further, when the index value falls in the mild interval, the membership degree is close to 0, indicating no obvious stress; When the index value gradually enters the moderate interval, the membership degree gradually increases; When the index reaches the extremely severe interval, the membership degree is close to 1, indicating severe water stress.

[0013] Further, the method for fuzzy comprehensive discrimination of the membership value is: Through a pre-set rule base, the membership degrees of multiple indexes are logically combined; According to the fuzzy rule, the logical combination is inferred, and the comprehensive water stress index is obtained through fuzzy set synthesis and defuzzification calculation.

[0014] Further, the method for fuzzy comprehensive discrimination of the membership value can also be: Assign appropriate weights to the water content, NDVI, CWSI, and VPD; After determining the weights, the membership values of each index are fused in a weighted manner to obtain the comprehensive water stress index SI, and the calculation formula is as follows: ; Wherein: n is the number of indexes participating in discrimination; w i is the weight of the i-th index, satisfying =1; μi(x) is the membership value corresponding to the i-th index.

[0015] In the second aspect, a crop water stress diagnosis system based on fuzzy discrimination theory and multi-source information fusion is provided, comprising: A data acquisition module configured to acquire multi-source data: the multi-source data includes soil moisture data, crop physiological data, remote sensing image data, and meteorological parameter data; A data processing module configured to implement data preprocessing and index calculation according to the multi-source data: the data preprocessing and index calculation includes remote sensing image preprocessing, data quality control, calculation of typical indexes related to crop water status, and normalization processing; A fuzzy discrimination module configured to establish different membership functions for different indexes according to fuzzy set theory, for describing the relationship between a water-related index and the degree of crop water stress; according to the membership function, the water membership value is fuzzyly comprehensively discriminated to obtain a comprehensive water stress index; A result display module configured to output mild, moderate, or severe crop water stress grades according to the comprehensive water stress index and provide visual display.

[0016] Furthermore, the data acquisition module includes a soil moisture sensor, a meteorological sensor, and a UAV remote sensing device.

[0017] Furthermore, the data processing module includes a geometric correction unit and a radiometric calibration unit for remote sensing image correction.

[0018] Furthermore, the fuzzy discrimination module includes a membership degree calculation unit and a fuzzy comprehensive operation unit.

[0019] Furthermore, the result display module includes a visualization interface for displaying spatial distribution maps and grade results, and is capable of communicating with the irrigation control terminal to achieve linkage between diagnostic results and irrigation scheduling.

[0020] The beneficial effects of this invention are: Compared with the prior art, the present invention has the following beneficial effects: 1. Multi-source information fusion: Simultaneously utilize soil moisture, crop physiology, remote sensing imagery, and meteorological parameters to avoid the limitations of a single data source and improve the comprehensiveness and representativeness of diagnostic results.

[0021] 2. Fuzzy set discrimination: The membership function and fuzzy comprehensive discrimination method are adopted to effectively characterize the gradual change and uncertainty of crop water stress and avoid misjudgment caused by rigid classification of traditional threshold method.

[0022] 3. High diagnostic accuracy: Through weighted fusion of multiple indicators, it can reflect the comprehensive impact of different factors on crop water status, and the diagnostic results are highly consistent with the measured data.

[0023] 4. Systematic Design: The proposed modular structure for data acquisition, data processing, fuzzy discrimination, and result display supports automated operation and visual output, facilitating widespread application.

[0024] 5. Intelligent linkage: The diagnostic results can be integrated with the irrigation control system to form a closed-loop management model of monitoring-diagnosis-control, which improves water resource utilization efficiency and has significant water-saving and efficiency-enhancing value. Attached Figure Description

[0025] Figure 1 Flowchart of the method of this invention; Figure 2 Membership function diagram; Figure 3 System structure diagram. Detailed Implementation

[0026] To make the objectives, advantages and features of the present invention more apparent, the following detailed description of the embodiments further illustrates the present invention.

[0027] like Figure 1 As shown, the present invention provides a method for diagnosing crop water stress, the specific implementation of which includes the following steps: Step S1: Multi-source data acquisition.

[0028] Collect multi-source monitoring data related to crop moisture status, including but not limited to: soil moisture data: deploy FDR or TDR soil moisture sensors in the crop root zone (0-60 cm), collect soil moisture content in the 0-20, 20-40, and 40-60 cm soil layers, with a sampling frequency of 30 minutes and an error controlled within ±0.5%; Crop physiological data: Periodic measurements of leaf area index (LAI), leaf water content, and plant height are used to supplement the reflection of the actual water status of the crop; Remote sensing image data: UAVs equipped with multispectral cameras (blue, green, red, red edge, and near-infrared bands) and thermal infrared cameras fly during key phenological periods to acquire high-resolution images at the field scale; Meteorological parameter data: Field automatic weather stations collect air temperature, humidity, wind speed, rainfall and solar radiation at 10-minute intervals, and further calculate vapor pressure difference (VPD).

[0029] It should be noted that by complementing data from multiple sources, the lack of representativeness or uncertainty caused by a single data source can be avoided, providing a data foundation for subsequent fuzzy fusion.

[0030] Step S2: Data preprocessing and index calculation.

[0031] 1. Remote sensing image preprocessing Geometric correction is performed using ground control points (GCP) and UAV GPS / IMU data to eliminate spatial distortion; Radiometric calibration is performed using a standard reflectivity calibration plate and camera calibration coefficients to obtain the standardized reflectivity; Multiple images are stitched together to generate an orthophoto of the complete field.

[0032] 2. Data quality control Outliers were removed using the IQR method; Use linear interpolation or kriging methods to fill in missing values; Based on the remote sensing flight time, soil moisture and meteorological data are synchronized in time.

[0033] 3. Indicator Calculation Calculate typical indicators related to crop water status: Normalized Difference Vegetation Index (NDVI): ; Wherein: NIR represents the reflectivity of the near-infrared band; Red represents the reflectivity of the red band.

[0034] Green Normalized Difference Vegetation Index (GNDVI): ; Wherein: NIR represents the reflectivity in the near-infrared band; Green represents the reflectivity in the red band.

[0035] Crop Water Stress Index (CWSI): ; in: T c For canopy temperature, T wet This is the wet limit temperature. T dry This is the dry limit temperature.

[0036] Vapor pressure differential (VPD): ; in: Indicates air temperature; T a The corresponding saturated vapor pressure.

[0037] 4. Normalization processing All indicators are mapped to the (0,1) interval using a linear normalization method: ; Where: X represents the original index value; X min X max These represent the minimum and maximum values ​​of the indicator, respectively. This represents the normalized value, used to ensure that data with different dimensions are comparable in subsequent discrimination.

[0038] Step S3: Construct the membership function.

[0039] Based on fuzzy set theory, membership functions are established for different indicators: 1. Threshold range setting Based on historical test data and expert experience, mild, moderate, and severe ranges were defined, for example: Moisture content: >18% is mild, 12-18% is moderate, and <12% is severe. NDVI: >0.6 indicates mild severity, 0.4–0.6 indicates moderate severity, and <0.4 indicates severe severity. CWSI: <0.4 is mild, 0.4–0.7 is moderate, and >0.7 is severe; VPD: <1.5 kPa is mild, 1.5–2.5 kPa is moderate, and >2.5 kPa is severe.

[0040] 2. Function Form Selection In this invention, the membership function is used to describe the relationship between a certain water-related index and the degree of crop water stress.

[0041] When the indicator value falls within the normal range, the membership degree is close to 0, indicating that there is no obvious stress. As the indicator value gradually enters the stress interval, the membership degree gradually increases; When the index reaches the extreme stress value, the membership degree is close to 1, indicating severe water stress.

[0042] Preferably, the function can be in the following form: Trigonometric functions: suitable for describing indicators sensitive to a single central value (such as NDVI), with a simple structure that can reflect the continuous transition from "mild to moderate to severe".

[0043] ; Trapezoid function: suitable for describing indicators with stable intervals (such as soil moisture content), and can reflect the gradual change on both sides of the threshold and the middle stable region.

[0044] ; Parameter settings: Function parameters a, b, c, and d can be determined through regression analysis of field measured data and crop yield, or adjusted based on expert experience for different crops and phenological stages.

[0045] Step S4: Fuzzy comprehensive discrimination and information fusion.

[0046] After constructing the membership functions of various water-related indicators, this invention performs fuzzy comprehensive discrimination on the membership values ​​to achieve unified integration of multi-source information and overall evaluation of crop water stress.

[0047] To ensure that the contribution of different indicators in the comprehensive judgment is reasonably reflected, corresponding weights are assigned to each indicator. These weights can be determined in various ways, including but not limited to: determining the correlation strength between each indicator and crop moisture status or yield based on correlation analysis; determining the amount of information contained in each indicator based on entropy weighting; and subjective weighting based on expert experience. In a further preferred embodiment, a combination of objective and subjective weighting can be used to improve the scientific rigor and applicability of the weight allocation.

[0048] After determining the weights, the membership degree values of each index are fused in a weighted manner to obtain the comprehensive moisture stress index (Stress Index, SI), and its calculation formula is as follows: ; where: n is the number of indexes participating in the discrimination; w i is the weight of the i-th index, satisfying = 1; μi(x) is the membership degree value corresponding to the i-th index.

[0049] In another embodiment, the present invention can also adopt a rule-based fuzzy inference method. This method logically combines the membership degrees of multiple indexes through a preset rule base, for example, "if NDVI is low and CWSI is high, then it is determined as severe moisture stress". After receiving the input, the system performs inference according to the fuzzy rules, and obtains the comprehensive moisture stress index through fuzzy set synthesis and defuzzification calculation (such as the centroid method). This method can effectively handle the non-linear relationship and interaction effect between indexes.

[0050] To improve the robustness of the diagnosis, the present invention introduces a mechanism for handling missing data and controlling outliers in the comprehensive discrimination process. When some index data is missing, the weights of the remaining indexes are renormalized to ensure the effectiveness of weighted fusion; when there are abnormal extreme values in the input data, methods such as interval clipping or confidence correction can be used to reduce their impact on the discrimination result.

[0051] Finally, the comprehensive moisture stress index SI output by the fuzzy comprehensive discrimination module has a value range of 0, 1. The larger the index value, the more severe the crop moisture stress. The comprehensive moisture stress index will be used as the input for subsequent hierarchical discrimination and result output, realizing the mapping and conversion from multi-source indexes to a single diagnosis result.

[0052] Step S5: Fuzzy comprehensive discrimination and information fusion.

[0053] After obtaining the comprehensive moisture stress index SI, the present invention performs hierarchical discrimination on it and outputs the diagnosis result.

[0054] The grading is based on a preset threshold interval to divide the index values into different moisture stress levels. The grading rules can be: When SI ≤ 0.33, it is determined as mild moisture stress; When 0.33 < SI ≤ 0.66, it is determined as moderate moisture stress; When SI > 0.66, it is determined as severe moisture stress.

[0055] Wherein: SI represents the comprehensive water stress index, with a value range of 0 to 1; the grading threshold can be determined based on historical monitoring data, expert experience, or crop yield-water response curves, and can be adjusted for different crops and different phenological stages.

[0056] The diagnostic results are output through the results display module, and can be selected in the following formats: 1. Spatial distribution map: Visually displays areas of mild, moderate, and severe water stress at the field scale; 2. Statistical Reports: Output the area ratio or crop number ratio corresponding to each level; 3. Time series curves: Tracking changes in water stress in the same region throughout the entire growth period.

[0057] The results display module can communicate with the irrigation control terminal to convert diagnostic results into irrigation scheduling criteria. For example, when the system determines that the stress is moderate, it can recommend appropriate water replenishment; when the system determines that the stress is severe, it can directly trigger zonal irrigation commands.

[0058] Through the above-described grading and output process, this invention achieves a mapping from the comprehensive water stress index to actual management decisions. For example... Figure 2 This demonstrates the core ideas of fuzzy set theory in the diagnosis of crop water stress: Map index values ​​(such as soil moisture content, NDVI, CWSI, VPD, etc.) to membership degrees between 0 and 1; Instead of using the traditional "hard threshold" for division, a "gradual range" is allowed to reflect the ambiguity of water stress; For example, when the moisture content is between 12 and 18%, it may fall under moderate stress, or even partially mild or severe stress. This ambiguity can be quantified using a membership function.

[0059] It should be noted that the figure illustrates that the present invention is not a simple threshold method, but rather introduces fuzzy mathematics to incorporate uncertainty and gradual change into the diagnostic model, thereby improving the scientific nature and accuracy of the diagnosis.

[0060] Based on the same inventive concept, such as Figure 3 As shown, the present invention also provides a crop water stress diagnosis system based on fuzzy discrimination theory and multi-source information fusion, comprising the following modules: 1. Data Acquisition Module This module is used to collect multi-source monitoring data, including soil moisture, crop physiology, remote sensing imagery, and meteorological parameters. Preferably, the module includes a soil moisture sensor unit, a UAV remote sensing unit, and an automatic weather station unit.

[0061] 2. Data Processing Module It is used to standardize and calculate indicators for collected data, and preferably includes a geometric correction unit, a radiometric calibration unit, an indicator calculation unit, and a normalization processing unit.

[0062] 3. Fuzzy discrimination module This is used to fuse multi-source indicators based on fuzzy set theory, preferably including a membership calculation unit and a fuzzy comprehensive operation unit, to output a comprehensive water stress index.

[0063] 4. Result Display Module It is used to output diagnostic results and provide visualization, preferably including a visualization interface and a communication interface unit to achieve linkage between the result display and the irrigation control system.

[0064] Through the modular design of the above system, the entire process from data acquisition to diagnostic output can be automated, supporting real-time operation and historical data backtracking, and is applicable to a variety of crops and regions.

[0065] Specifically, the data acquisition module consists of a soil moisture sensor, a drone remote sensing system, and an automatic weather station, enabling the acquisition of multi-dimensional data on soil, crops, and the environment. The data processing module includes a geometric correction unit, a radiometric calibration unit, an index calculation unit, and a normalization unit, ensuring the automation and standardization of data processing; The fuzzy discrimination module consists of a membership degree calculation unit and a fuzzy comprehensive operation unit. It is the core innovative part of this system and can fuse multi-source information based on fuzzy mathematics to generate a comprehensive water stress index. The results display module includes a visualization interface and a communication interface unit, supporting the output of spatial distribution maps, reports, and irrigation linkage.

[0066] The diagram visually illustrates how the system achieves a closed-loop structure of "data acquisition - data processing - fuzzy discrimination - result output". The fuzzy discrimination module combines fuzzy mathematics theory with multi-source information fusion, forming a key innovation that distinguishes it from traditional diagnostic systems.

[0067] It should be understood that the following embodiments are for illustrative purposes only and are not intended to limit the scope of protection of the present invention.

[0068] Example 1: Validation of a crop water stress diagnosis method based on field trials This embodiment was conducted in the field maize planting area of ​​the Hetao Irrigation District in Inner Mongolia. The experimental field area was about 50 mu, and an irrigation method combining canal water diversion and sprinkler irrigation was adopted. Three different irrigation treatments were set up: sufficient irrigation, moderate water shortage and severe water shortage, to simulate different water supply levels.

[0069] Soil moisture sensors (Decagon 5TE type) were deployed in each treatment zone at depths of 0–20 cm, 20–40 cm, and 40–60 cm, with a sampling interval of 30 minutes, to monitor the soil moisture content in the root zone in real time. Simultaneously, maize leaves were periodically collected in the field to determine fresh weight and dry weight after drying, calculate leaf water content, and measure leaf area index using a leaf area meter to reflect the physiological state of the crop.

[0070] During the crop's jointing, tasseling, and grain-filling stages, unmanned aerial vehicles (UAVs) were used to acquire remote sensing images. The UAVs were equipped with multispectral cameras to acquire images in the blue, green, red, red-edge, and near-infrared bands, and with thermal infrared cameras to acquire canopy temperature images. The flight altitude was approximately 100 m, and the image resolution was approximately 10 cm. Automatic weather stations were also deployed in the experimental area to collect data on air temperature, humidity, wind speed, solar radiation, and rainfall at 10-minute intervals to calculate vapor pressure differential (VPD).

[0071] After geometric correction, radiometric calibration, and image stitching, the collected multi-source data were used to calculate NDVI, GNDVI, CWSI, and VPD indices, and each index was normalized to the interval of 0, 10, 10, 1. Subsequently, triangular and trapezoidal membership functions were constructed based on fuzzy set theory, and weighted and fused according to the weights of NDVI 0.3, CWSI 0.3, soil moisture content 0.2, and VPD 0.2 to obtain the comprehensive water stress index SI.

[0072] In monitoring during the tasseling stage in July 2024, the SI value for adequate irrigation was 0.28, corresponding to mild water stress; the SI value for moderate water shortage was 0.54, corresponding to moderate water stress; and the SI value for severe water shortage was 0.72, corresponding to severe water stress. These diagnostic results are highly consistent with the observed decreases in measured soil moisture content (below 12% in the severe water shortage treatment) and leaf water content, indicating that the method of this invention can accurately diagnose the degree of crop water stress and has good applicability and reliability.

[0073] Example 2: Crop Water Stress Diagnosis Based on System Platform The crop water stress diagnosis system provided by this invention was constructed and deployed at the Hetao Irrigation District field test station. The system consists of a data acquisition module, a data processing module, a fuzzy discrimination module, and a result display module, and operates in an edge server and web management platform environment.

[0074] During the experiment, the system was connected to 12 sets of soil moisture sensors, an automatic weather station, and a UAV remote sensing platform. The soil moisture data was transmitted in real time through the LoRa wireless network, the meteorological data was uploaded through the 4G module, and the multi-spectral and thermal infrared images collected by the UAV were transmitted to the processing end through the ground base station. The data processing module automatically completed geometric correction, radiometric calibration, and image mosaicking, and calculated the NDVI, GNDVI, CWSI, and VPD indices.

[0075] The fuzzy discrimination module called the membership function library in the background to perform fuzzy processing on each index, and automatically selected the weight combination according to the characteristics of different phenological periods. For example, the weight of NDVI was increased during the jointing stage, and the weight of CWSI was increased during the filling stage. The comprehensive water stress index SI was obtained through weighted fusion, and the grading results were output in real time.

[0076] The result display module provided visual display through the Web interface. Different water stress levels were marked in green, yellow, and red on the field distribution map. At the same time, statistical reports and time series curves of the area ratios of each level were output. The system ran continuously for two months, and the diagnostic results could be updated daily, realizing the automatic monitoring and visual presentation of the water stress status.

[0077] Example 3: Linked application of diagnostic results and irrigation control In the irrigation scheduling platform of the Hetao Irrigation Area, the diagnostic system of the present invention was docked with the automatic irrigation equipment, realizing the closed-loop control from monitoring and diagnosis to irrigation execution.

[0078] During the operation of the system, the comprehensive water stress index SI was calculated and graded in real time. When SI ≤ 0.33, the system determined it as mild water stress, and only prompted "No irrigation needed" on the platform interface; when 0.33 < SI ≤ 0.66, it was determined as moderate water stress, and the system automatically generated a recommended irrigation amount of 30–40 mm. After the dispatcher confirmed, the command was sent to the sub-area solenoid valve controller; when SI > 0.66, the system determined it as severe water stress, directly triggering the automatic control equipment to implement 50–80 mm of sub-area irrigation.

[0079] In an application in August 2024, a certain field was determined to be severely water stressed, and the system automatically executed an irrigation operation of 60 mm. After the irrigation was completed, the monitoring data showed that the canopy temperature decreased by about 2.8 °C, the vapor pressure deficit decreased by 0.9 kPa, and the soil moisture content increased significantly. This result was consistent with the diagnostic conclusion, verifying the effectiveness and reliability of the present invention in practical applications.

[0080] It should be noted that any process or method description in the above embodiments can be understood as representing a module, segment, or portion of code comprising one or more executable instructions for implementing a specific logical function or process, and the scope of the preferred embodiments of the present invention includes additional implementations in which functions may be performed not in the order shown or discussed, including substantially simultaneously or in reverse order according to the functions involved, as should be understood by those skilled in the art to which the embodiments of the present invention pertain.

[0081] It should be noted that the logic and / or steps in the embodiments, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-included system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of computer-readable media include: an electrical connection having one or more wires (electronic device), a portable computer disk drive (magnetic device), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Alternatively, the computer-readable medium may be paper or other suitable media on which the program can be printed, since the program can be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in a computer memory.

[0082] It should be understood that various parts of the present invention can be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0083] Those skilled in the art will understand that all or part of the steps of the methods described in the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, the program includes one or a combination of the steps of the method embodiments.

[0084] Furthermore, in the embodiments of the present invention, the functional modules can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module. The integrated module can be implemented in hardware or as a software functional module. If the integrated module is implemented as a software functional module and sold or used as an independent product, it can also be stored in a computer-readable storage medium.

[0085] The storage media mentioned above can be read-only memory, disk, or optical disk, etc.

[0086] The above embodiments have provided a detailed description of the technical solution of the present invention. Obviously, the present invention is not limited to the described embodiments. Based on the embodiments of the present invention, those skilled in the art can make various modifications, but any modifications that are equivalent to or similar to the present invention fall within the scope of protection of the present invention.

[0087] The contents not described in detail in this specification are existing technologies known to those skilled in the art.

Claims

1. A method for diagnosing crop water stress based on fuzzy discriminant theory and multi-source information fusion, characterized in that, Includes the following steps: Collect multi-source data: The multi-source data includes soil moisture data, crop physiological data, remote sensing image data, and meteorological parameter data; Based on the multi-source data, data preprocessing and index calculation are performed: the data preprocessing and index calculation include remote sensing image preprocessing, data quality control, calculation of typical indicators related to crop water status, and normalization processing; Based on fuzzy set theory, different membership functions are established for different indicators to describe the relationship between a certain water-related indicator and the degree of crop water stress. Based on the membership function, a fuzzy comprehensive discrimination is performed on the water membership value to obtain the comprehensive water stress index. The crop water stress level is output as mild, moderate, or severe based on the comprehensive water stress index.

2. The method according to claim 1, characterized in that, The method for collecting soil moisture data is as follows: moisture content of soil layers 0-20 cm, 20-40 cm, and 40-60 cm are collected in the crop root zone 0-60 cm. The method for collecting the crop physiological data is as follows: periodically measuring leaf area index, leaf water content, and plant height; The method for acquiring the remote sensing image data is as follows: using multispectral and thermal infrared data to obtain high-resolution images at the field scale; The meteorological parameter data are collected by collecting air temperature, humidity, wind speed, rainfall and solar radiation, and then calculating the vapor pressure difference.

3. The method according to claim 2, characterized in that, The remote sensing image preprocessing specifically includes: Perform geometric correction to eliminate spatial distortion; Radiometric calibration is performed using a standard reflectivity calibration plate and camera calibration coefficients to obtain the standardized reflectivity; Multiple images are stitched together to generate an orthophoto of the complete field. The data quality control specifically includes: Outliers were removed using the IQR method; Use linear interpolation or kriging methods to fill in missing values; Synchronize soil moisture and meteorological data in real time; The typical indicators related to crop water status are: Normalized Difference Vegetation Index (NDVI), Green Normalized Difference Vegetation Index (GNDVI), Crop Water Stress Index (CWSI), and Vapor Pressure Difference (VPD).

4. The method according to claim 3, characterized in that, The membership function is constructed as follows: Threshold range setting: Based on historical test data and expert experience, mild, moderate and severe ranges for moisture content, NDVI, CWSI and VPD are set respectively; Function form selection: Choose trigonometric function, trapezoidal function, or Gaussian function.

5. The method according to claim 4, characterized in that, When the indicator value falls in the mild range, the membership degree is close to 0, indicating that there is no obvious stress. As the index value gradually enters the medium range, the membership degree gradually increases; When the index reaches the extremely severe range, the membership degree is close to 1, indicating severe water stress.

6. The method according to claim 5, characterized in that, The method for fuzzy comprehensive discrimination of the membership values ​​is as follows: By using a pre-defined rule base, the membership degrees of multiple indicators are logically combined; The logical combination is reasoned according to fuzzy rules, and the comprehensive water stress index is obtained through fuzzy set synthesis and defuzzification calculation.

7. The method according to claim 5, characterized in that, The method for fuzzy comprehensive discrimination of the membership values ​​is as follows: Assign appropriate weights to moisture content, NDVI, CWSI, and VPD; After determining the weights, the membership values ​​of each indicator are fused in a weighted manner to obtain the comprehensive water stress index SI, which is calculated as follows: ； Where: n is the number of indicators involved in the judgment; w i The weight of the i-th indicator satisfies =1; μi(x) is the membership value corresponding to the i-th index.

8. A crop water stress diagnosis system based on fuzzy discriminant theory and multi-source information fusion, characterized in that, include: The data acquisition module is configured to acquire multi-source data, including soil moisture data, crop physiological data, remote sensing image data, and meteorological parameter data. The data processing module is configured to perform data preprocessing and index calculation based on the multi-source data: the data preprocessing and index calculation includes remote sensing image preprocessing, data quality control, calculation of typical indicators related to crop water status, and normalization processing; The fuzzy discrimination module is configured to establish different membership functions for different indicators based on fuzzy set theory, which are used to describe the relationship between a certain water-related indicator and the degree of crop water stress; and to perform fuzzy comprehensive discrimination on the water membership value based on the membership function to obtain the comprehensive water stress index. The results display module is configured to output mild, moderate or severe crop water stress levels based on the comprehensive water stress index and provide a visual display.

9. The system according to claim 8, characterized in that, The data acquisition module includes a soil moisture sensor, a meteorological sensor, and a UAV remote sensing device.

10. The system according to claim 8, characterized in that, The data processing module includes a geometric correction unit and a radiometric calibration unit for remote sensing image correction.

11. The system according to claim 8, characterized in that, The fuzzy discrimination module includes a membership degree calculation unit and a fuzzy comprehensive operation unit.

12. The system according to claim 8, characterized in that, The results display module includes a visualization interface for displaying spatial distribution maps and grade results, and can communicate with the irrigation control terminal to realize the linkage between diagnostic results and irrigation scheduling.

Images