A method for evaluating drought stress of alfalfa based on unmanned aerial vehicle canopy temperature and vegetation index fusion

Abstract

The application discloses a kind of alfalfa drought stress evaluation methods based on unmanned aerial vehicle canopy temperature and vegetation index fusion, it is related to agricultural remote sensing monitoring technical field.The unmanned aerial vehicle thermal infrared image and unmanned aerial vehicle multispectral image with position corresponding relationship are acquired, and the evaluation unit in alfalfa planting area is determined;Based on the effective canopy area of alfalfa, canopy temperature value and vegetation index value are obtained, canopy temperature stress index and vegetation index deviation index are determined;According to the binary combination determination rule formed by temperature stress threshold and growth potential deviation threshold, the drought stress correlation state of each evaluation unit is determined, and the drought stress zoning evaluation result is formed.The application can solve the problem that canopy temperature response and vegetation index deviation response are difficult to distinguish state, and realize the zoning evaluation of drought stress correlation state.

Description

Technical Field

[0001] This invention relates to the field of agricultural remote sensing monitoring technology, and in particular to a method for assessing drought stress in alfalfa based on the fusion of canopy temperature and vegetation index from unmanned aerial vehicles (UAVs). Background Technology

[0002] Alfalfa is a perennial leguminous forage crop characterized by its multi-year utilization, multiple harvests, and regeneration after harvest. Field moisture conditions affect alfalfa's greening, branching, regeneration rate after harvest, and forage stability. Traditional soil moisture measurement typically relies on fixed-point sensors or manual sampling to obtain point-based soil moisture data. However, soil texture, terrain undulation, irrigation uniformity, and regeneration progress after harvest are not entirely consistent within alfalfa fields, making point-based data insufficient to represent the spatial distribution of water stress within the field. Existing literature, "Research Progress on the Application of UAV Remote Sensing Technology in Precision Irrigation," records that UAV remote sensing can acquire information such as crop canopy temperature, canopy structure, and leaf pigment content, which can be used to perceive the distribution of crop water stress. For alfalfa planting areas requiring zoned irrigation or zoned water and fertilizer management, UAV thermal infrared and multispectral imagery can provide canopy state information corresponding to the spatial location of the field.

[0003] In existing UAV remote sensing drought monitoring methods, canopy temperature is often used to characterize crop transpiration and cooling status, while vegetation indices are often used to characterize crop canopy cover and growth status. Existing literature, "Research on Inversion of Farmland Surface Soil Moisture Using an Improved Temperature-Vegetation Drought Index," describes the construction of a temperature-vegetation drought index based on the feature space of surface temperature and vegetation indices, which can be used for inversion of farmland surface soil moisture. Patent document CN114778476A discloses a UAV remote sensing-based soil moisture monitoring model for alfalfa-cotton fields. This scheme collects UAV remote sensing data and ground survey data, identifies the distribution areas of alfalfa, cotton, and bare soil, calculates vegetation indices, vegetation cover, water stress index, canopy-bare soil temperature difference, and vegetation cover-temperature difference index, and establishes a soil moisture monitoring model for alfalfa-cotton fields.

[0004] The aforementioned methods can utilize remote sensing parameters such as canopy temperature and vegetation index to retrieve soil moisture content or determine water shortage status. However, the results are mostly expressed as numerical values ​​of soil moisture content, overall water shortage degree, or single stress level. Within alfalfa planting areas, changes in canopy temperature and vegetation index correspond to different canopy status information, and the two indicators may exhibit inconsistent performance in different assessment units. Existing methods do not address the combined relationship between canopy temperature stress response and vegetation index deviation response, nor do they differentiate between areas with significant temperature stress but insignificant growth deviation, significant temperature stress and growth deviation, significant growth deviation but insignificant temperature stress, and insignificant growth deviation. This makes it difficult for field zoning assessment results to reflect the differences in stress performance across different areas. Summary of the Invention

[0005] The purpose of this invention is to provide a method for assessing drought stress in alfalfa based on the fusion of canopy temperature and vegetation index from unmanned aerial vehicles (UAVs). This method can solve the problem of difficulty in distinguishing the state of canopy temperature response and vegetation index deviation response in existing alfalfa drought monitoring, and achieve zonal assessment of drought stress-related states.

[0006] To achieve the above objectives, the present invention provides the following solution: A method for assessing drought stress in alfalfa based on the fusion of canopy temperature and vegetation indices from unmanned aerial vehicles (UAVs), comprising: Acquire UAV thermal infrared and UAV multispectral images with location correspondence, and determine the assessment units within the alfalfa planting area according to the preset assessment scale; The effective canopy area of ​​alfalfa in each assessment unit was determined based on UAV multispectral imagery, and the canopy temperature and vegetation index values ​​corresponding to the effective canopy area of ​​alfalfa were obtained. The canopy temperature stress index for each assessment unit is determined based on the canopy temperature value, and the vegetation index deviation index for each assessment unit is determined based on the vegetation index value. According to the binary combination judgment rule formed by the temperature stress threshold and the growth deviation threshold, the canopy temperature stress index and vegetation index deviation index are combined for judgment to determine the drought stress association status of alfalfa in each assessment unit. The binary combination judgment rule includes four types of judgment results: initial drought stress, continuous drought stress, abnormal growth deviation and no obvious drought stress. Based on the drought stress correlation status of alfalfa in each assessment unit, the drought stress zoning assessment results of alfalfa planting areas are formed.

[0007] Preferably, acquiring UAV thermal infrared images and UAV multispectral images with location correspondence, and determining assessment units within the alfalfa planting area according to a preset assessment scale, including: Determine the common coverage area of ​​UAV thermal infrared imagery and UAV multispectral imagery within the alfalfa planting area; The common coverage area is divided into multiple assessment units according to the preset assessment criteria; Within each evaluation unit, the thermal infrared image region corresponding to the UAV thermal infrared image and the multispectral image region corresponding to the UAV multispectral image are determined respectively.

[0008] Preferably, within each evaluation unit, the thermal infrared image region corresponding to the UAV thermal infrared image and the multispectral image region corresponding to the UAV multispectral image are determined, including: Based on field boundary features, row boundary features, or ground control points, establish the positional correspondence between UAV thermal infrared images and UAV multispectral images; Based on the location correspondence, each evaluation unit is assigned to a thermal infrared image region and a multispectral image region, respectively; The thermal infrared image region and the multispectral image region correspond to the same alfalfa canopy region.

[0009] Preferably, the effective canopy area of ​​alfalfa within each assessment unit is determined based on UAV multispectral imagery, and the canopy temperature value and vegetation index value corresponding to the effective canopy area of ​​alfalfa are obtained, including: Calculate the initial values ​​of vegetation index in each assessment unit based on UAV multispectral imagery; Alfalfa canopy pixels within each assessment unit were selected based on the initial values ​​of the vegetation index. Connectivity region screening was performed on alfalfa canopy pixels to obtain the effective canopy region of alfalfa in each evaluation unit; Based on the location of the effective canopy area of ​​alfalfa in UAV thermal infrared and UAV multispectral images, the canopy temperature value and vegetation index value of the corresponding evaluation unit are extracted respectively.

[0010] Preferably, connected component screening is performed on alfalfa canopy pixels to obtain the effective alfalfa canopy region within each evaluation unit, including: The area of ​​the connected regions formed by alfalfa canopy pixels is used for screening, and connected regions that do not meet the preset area conditions are removed. Calculate the area percentage of the retained connected regions in the corresponding evaluation unit; When the area ratio meets the preset validity conditions, the retained connected area will be regarded as the effective canopy area of ​​alfalfa in the corresponding evaluation unit; When the area proportion does not meet the preset validity conditions, the corresponding assessment unit will be excluded from the determination process of alfalfa drought stress association status.

[0011] Preferably, the canopy temperature stress index for each assessment unit is determined based on the canopy temperature value, and the vegetation index deviation index for each assessment unit is determined based on the vegetation index value, including: Based on the canopy temperature values ​​of each assessment unit, determine the representative canopy temperature value for each assessment unit; Based on the representative values ​​of canopy temperature of each assessment unit within the same data collection period, determine the low temperature reference value and the high temperature reference value; Based on the normalization relationship between the representative value of canopy temperature and the low-temperature reference value and the high-temperature reference value, the canopy temperature stress index of each assessment unit is determined.

[0012] Preferably, the canopy temperature stress index for each assessment unit is determined based on the canopy temperature value, and the vegetation index deviation index for each assessment unit is determined based on the vegetation index value, including: Based on the vegetation index values ​​of each assessment unit, determine the representative value of the vegetation index for each assessment unit. Based on the representative values ​​of vegetation indices of each assessment unit within the same data collection period, the reference values ​​of vegetation indices are determined. The vegetation index deviation index for each assessment unit is determined based on the degree of deviation of the representative value of the vegetation index from the reference value of the vegetation index.

[0013] Preferably, the canopy temperature stress index and vegetation index deviation index are combined and judged according to the binary combination judgment rule formed by the temperature stress threshold and the growth deviation threshold to determine the drought stress association status of alfalfa in each assessment unit, including: Based on the comparison results between the canopy temperature stress index and the temperature stress threshold, the temperature stress judgment result of each assessment unit is determined. The temperature stress judgment result includes reaching the temperature stress threshold and not reaching the temperature stress threshold. Based on the comparison results between the vegetation index deviation index and the growth deviation threshold, the growth deviation judgment result of each assessment unit is determined. The growth deviation judgment result includes reaching the growth deviation threshold and not reaching the growth deviation threshold. Based on the binary combination of temperature stress determination results and growth deviation determination results, the drought stress association status of alfalfa in each assessment unit is determined.

[0014] Preferably, the drought stress correlation status of alfalfa in each assessment unit is determined based on the binary combination of the temperature stress determination result and the growth deviation determination result, including: When the temperature stress assessment result indicates that the canopy temperature stress index has reached the temperature stress threshold, and the growth deviation assessment result indicates that the vegetation index deviation index has not reached the growth deviation threshold, the corresponding assessment unit will be identified as the initial drought stress. When the temperature stress assessment result indicates that the canopy temperature stress index has reached the temperature stress threshold, and the growth deviation assessment result indicates that the vegetation index deviation index has reached the growth deviation threshold, the corresponding assessment unit will be identified as being under continuous drought stress. When the temperature stress assessment result indicates that the canopy temperature stress index has not reached the temperature stress threshold, and the growth deviation assessment result indicates that the vegetation index deviation index has reached the growth deviation threshold, the corresponding assessment unit will be identified as having abnormal growth deviation. When the temperature stress assessment result indicates that the canopy temperature stress index has not reached the temperature stress threshold, and the growth deviation assessment result indicates that the vegetation index deviation index has not reached the growth deviation threshold, the corresponding assessment unit will be determined as having no obvious drought stress.

[0015] Preferably, based on the drought stress association status of alfalfa in each assessment unit, a drought stress zoning assessment result for the alfalfa planting area is formed, including: By merging assessment units that have the same alfalfa drought stress association status and are spatially adjacent, drought stress partitions are obtained. Based on the alfalfa drought stress association status of each assessment unit within the drought stress zoning, the zoning stress type of the drought stress zoning is determined. Based on drought stress zones and zone stress types, the drought stress zone assessment results for alfalfa planting areas are formed. The results of drought stress zoning assessments will be used as the basis for zoning water and fertilizer management operations in alfalfa planting areas.

[0016] The present invention discloses the following beneficial effects: This invention acquires location-corresponding UAV thermal infrared and multispectral images, and determines assessment units within alfalfa planting areas according to a preset assessment scale, thus establishing canopy temperature data and vegetation index data on the same spatial object. Each assessment unit corresponds to a set of canopy temperature values ​​and a set of vegetation index values, avoiding the mixing of canopy temperature responses and growth responses from different locations for the same judgment. This process solves the problem of difficulty in correspondingly expressing spatial differences within alfalfa fields during the assessment process, providing a clear spatial basis for subsequent drought stress assessment.

[0017] This invention uses UAV multispectral imagery to determine the effective canopy area of ​​alfalfa within each evaluation unit, and obtains canopy temperature and vegetation index values ​​based on this effective canopy area. Both canopy temperature and vegetation index values ​​originate from the same canopy object; the canopy temperature stress index is no longer individually affected by the temperature of non-canopy areas, and the vegetation index deviation index is no longer calculated outside the alfalfa canopy range. This processing ensures that the canopy temperature stress index and the vegetation index deviation index share a common object basis, providing a consistent data source for subsequent combination determinations.

[0018] This invention determines a canopy temperature stress index based on canopy temperature values ​​and a vegetation index deviation index based on vegetation index values. The canopy temperature stress index characterizes the alfalfa canopy temperature response, while the vegetation index deviation index characterizes the alfalfa canopy growth deviation response. The canopy temperature stress index corresponds to the canopy temperature response of the assessment unit, and the vegetation index deviation index corresponds to the canopy growth deviation response of the assessment unit. After the two indices are formed, subsequent judgments can preserve the differences between temperature stress performance and growth deviation performance.

[0019] This invention uses a binary combination judgment rule formed by a temperature stress threshold and a growth deviation threshold to combine and judge the canopy temperature stress index and the vegetation index deviation index. The canopy temperature stress index, relative to the temperature stress threshold, forms two states: "reached" and "not reached." Similarly, the vegetation index deviation index, relative to the growth deviation threshold, also forms two states: "reached" and "not reached." Combining these two binary states results in four judgment outcomes, corresponding to initial drought stress, persistent drought stress, abnormal growth deviation, and no obvious drought stress, respectively. These four judgment outcomes, combined with the two comparison results, can distinguish at the assessment unit scale areas with significant temperature stress but insignificant growth deviation, significant temperature stress and growth deviation, significant growth deviation but insignificant temperature stress, and insignificant conditions for both.

[0020] This invention generates drought stress zoning assessment results for alfalfa planting areas based on the drought stress correlation status of each assessment unit. The assessment results no longer simply represent soil moisture content values ​​or overall water shortage levels, but rather express the drought stress correlation status of different regions based on the assessment units. This approach solves the problem of insufficient state-specificity in existing technologies, enabling the differentiation and representation of areas with initial drought stress, persistent drought stress, abnormal growth deviations, and areas without obvious drought stress within alfalfa planting areas. Attached Figure Description

[0021] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0022] Figure 1 This is a flowchart of a method provided in an embodiment of the present invention.

[0023] Figure 2 This is a schematic diagram of the extraction of the effective canopy region of alfalfa provided in an embodiment of the present invention.

[0024] Figure 3 This is a schematic diagram of binary combination determination provided in an embodiment of the present invention.

[0025] Figure 4 This is a schematic diagram of the system structure provided in an embodiment of the present invention. Detailed Implementation

[0026] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0027] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0028] Figure 1 The method flowchart provided in the embodiments of the present invention is as follows: Figure 1 As shown, this embodiment provides a method for assessing drought stress in alfalfa based on the fusion of canopy temperature and vegetation index from unmanned aerial vehicles (UAVs), including: Step 100: Acquire UAV thermal infrared images and UAV multispectral images with location correspondence, and determine the assessment units in the alfalfa planting area according to the preset assessment scale; Step 200: Determine the effective canopy area of ​​alfalfa in each assessment unit based on UAV multispectral imagery, and obtain the canopy temperature value and vegetation index value corresponding to the effective canopy area of ​​alfalfa; Step 300: Determine the canopy temperature stress index for each assessment unit based on the canopy temperature value, and determine the vegetation index deviation index for each assessment unit based on the vegetation index value. Step 400: According to the binary combination judgment rule formed by the temperature stress threshold and the growth deviation threshold, the canopy temperature stress index and vegetation index deviation index are combined and judged to determine the drought stress association status of alfalfa in each assessment unit; the binary combination judgment rule includes four types of judgment results: initial drought stress, continuous drought stress, abnormal growth deviation and no obvious drought stress. Step 500: Based on the drought stress correlation status of alfalfa in each assessment unit, form the drought stress zoning assessment results for alfalfa planting areas.

[0029] Specifically, in step 100, the UAV thermal infrared image and the UAV multispectral image are preferably acquired within the same acquisition period. When adjacent acquisition periods are used, the adjacent acquisition periods should meet the requirement of consistency in alfalfa canopy status.

[0030] Alfalfa exhibits different canopy cover states during the greening-up stage, rapid branching stage, post-cut regeneration stage, and before and after budding. When conducting drought stress assessments for the same collection batch, UAV thermal infrared and multispectral images correspond to the same growth stage or the same post-cut regeneration stage. When monitoring multiple batches of the same alfalfa planting area, the same preset assessment scale and assessment unit number can be maintained, ensuring spatial comparability of drought stress zoning assessment results from different collection batches.

[0031] The drone thermal infrared imagery and drone multispectral imagery cover the same alfalfa planting area. During data processing, the common coverage area of ​​the drone thermal infrared imagery and drone multispectral imagery within the alfalfa planting area is first determined. Then, the common coverage area is divided into multiple evaluation units according to a preset evaluation scale. Each evaluation unit corresponds to a thermal infrared imagery area in the drone thermal infrared imagery and a multispectral imagery area in the drone multispectral imagery, respectively.

[0032] As an example, in this embodiment, alfalfa is a perennial leguminous forage grass. Its field canopy is typically affected by branching density, post-cutting regeneration progress, localized missing plants, and the degree of bare ground. Within the same field, areas with inconsistent canopy coverage and height are easily formed. UAV thermal infrared imagery reflects canopy surface temperature, while multispectral imagery reflects canopy spectral growth. If the two types of imagery do not spatially correspond to the same evaluation unit, areas with high temperatures in bare soil, low post-cutting coverage, or uneven localized regeneration may be mistakenly used for the same set of index calculations. Therefore, this embodiment uses the evaluation unit as a common spatial object, ensuring that the subsequently extracted canopy temperature and vegetation index values ​​all correspond to the alfalfa canopy within the same area.

[0033] To ensure that canopy temperature values ​​and vegetation index values ​​correspond to the same spatial object, a positional correspondence between UAV thermal infrared and multispectral images can be established based on field boundary features, ground control points, stable ground features, or row boundary features within alfalfa planting areas that can be jointly identified by thermal infrared and multispectral images. After establishing this correspondence, each assessment unit is mapped to both the thermal infrared and multispectral image regions, ensuring that the thermal infrared and multispectral image regions within the same assessment unit correspond to the same alfalfa canopy region. Once the canopy temperature and vegetation index values ​​are correlated at the assessment unit scale, subsequent determinations can establish a correspondence between temperature response and growth response based on the same alfalfa canopy region.

[0034] In step 200, the effective canopy area of ​​alfalfa within each evaluation unit is determined based on UAV multispectral imagery. Combined with... Figure 2 As shown, this embodiment determines the evaluation unit in the UAV multispectral image according to a preset evaluation scale. and in evaluation units Inner pixel Initial value of vegetation index As a basis for canopy screening. Figure 2 In This represents the canopy screening threshold and the initial value of the vegetation index. Pixels that meet the canopy screening criteria are included in the candidate canopy pixel set. Candidate canopy pixel set The dataset may contain scattered noisy pixels or discontinuous small patches, therefore the candidate canopy pixel set... Connectivity region screening was performed to retain continuously distributed canopy regions that met the area requirements, thus obtaining the effective canopy regions of alfalfa. . Figure 2 In Indicates the effective canopy area of ​​alfalfa In the evaluation unit The area proportion in Used to judge the evaluation unit Whether there is sufficient effective canopy area to participate in the subsequent determination of alfalfa drought stress association status. Each evaluation unit is denoted as ,Will The pixels inside are recorded as After reflectivity conversion, the multispectral images from the UAV are processed to obtain pixels. Corresponding red light band reflectivity Red-edge band reflectivity and near-infrared reflectivity Initial value of vegetation index Normalized Difference Vegetation Index (NDVI) can be used. Or red-edged normalized vegetation index . and Calculate according to the following formulas: In the formula, Represents a pixel Normalized Difference Vegetation Index; Represents a pixel Red-edged normalized vegetation index; Represents a pixel Near-infrared reflectivity; Represents a pixel Reflectivity in the red light band; Represents a pixel The red-edge band reflectivity. When the pixel When the band reflectance is missing, the reflectance is abnormal, or the denominator does not meet the effective calculation conditions, the pixel Not included in the initial value of vegetation index The calculation. Suitable for collection stages where there are significant differences in canopy coverage. Suitable for alfalfa with high canopy cover, The collection phase is approaching saturation. The same vegetation index is used as the initial value for the same collection batch. .

[0035] Canopy screening threshold The vegetation index is determined based on the initial set of effective pixels within the alfalfa planting area. The initial set of effective pixel vegetation index values ​​is denoted as... ,right Candidate threshold in Calculate the inter-class variance : In the formula, express The value is not greater than the candidate threshold. Pixel percentage; express Medium greater than the candidate threshold Pixel percentage; express The value is not greater than the candidate threshold. The initial mean value of the vegetation index; express Medium greater than the candidate threshold The initial mean value of the vegetation index. Candidate threshold. Both classification pixel sets should be non-empty. Canopy screening threshold. Take Maximum candidate threshold . The distribution of initial vegetation index values ​​within the same collection batch determines the value and is not used as a fixed empirical threshold.

[0036] No. Each assessment unit Candidate canopy cell set within Determine by the following formula: In the formula, Indicates the first The set of candidate canopy pixels within each evaluation unit; Represents a cell The initial value of the vegetation index; This represents the canopy screening threshold. It applies to the candidate canopy pixel set. Perform connected component filtering to obtain several connected components. Calculate the area of ​​each connected component and remove those that do not meet the preset area criteria. The set of retained connected components is denoted as . . exist Area proportion Calculate using the following formula: In the formula, Indicates the first The percentage of effective canopy area in each assessment unit; Indicates the first The total area of ​​the connected regions retained within each evaluation unit; Indicates the first The area of ​​each evaluation unit. When UAV thermal infrared imagery and UAV multispectral imagery are unified to the same spatial resolution, and They can also be represented by the number of corresponding pixels.

[0037] when When the preset validity conditions are met, As the first The effective canopy area of ​​alfalfa within each assessment unit. When the preset validity conditions are not met, the first Each assessment unit is excluded from the process of determining the drought stress association status of alfalfa. This screening process excludes bare soil, field ridges, areas with missing plants, and scattered noise pixels from subsequent index calculations.

[0038] For those involved in the judgment Each evaluation unit, from Extracting canopy temperature pixel values ​​from corresponding UAV thermal infrared images and from Extracting vegetation index pixel values ​​from corresponding UAV multispectral images Representative value of canopy temperature Representative value of vegetation index Determine them respectively using the following formulas: In the formula, Indicates the first Representative values ​​of canopy temperature for each assessment unit; Indicates the effective canopy area of ​​alfalfa Inner pixel The canopy temperature value; Indicates the first Representative values ​​of vegetation index for each assessment unit; Indicates the effective canopy area of ​​alfalfa Inner pixel The vegetation index value; This represents the median value. In practice, the average value can also be used after removing outliers. and Using the median or the average value after removing outliers can reduce the impact of individual outlier pixels on the statistical results of the evaluation unit.

[0039] In step 300, the canopy temperature stress index for each assessment unit is determined based on the canopy temperature value, and the vegetation index deviation index for each assessment unit is determined based on the vegetation index value. Within the same data collection batch, representative canopy temperature values ​​from all participating assessment units are compiled into a set. .right The canopy temperature values ​​in the data are arranged in ascending order to obtain... , This indicates the number of evaluation units involved in the judgment. percentile Determine by the following formula: In the formula, Indicates the first The ordinal position corresponding to the percentile; Indicates rounding up. Low temperature reference value. and high temperature reference value Determine them respectively using the following formulas: In the formula, Indicates the low temperature reference value; This indicates a high-temperature reference value. express The 10th percentile; express The 90th percentile. When calibrated sample areas or historical observation data exist in the same alfalfa planting area, the low temperature reference value... and high temperature reference value Correction can be performed by combining calibration sample area data or historical observation data. Quantile statistics are used to determine... and At that time, extreme abnormal temperature values ​​are not directly used as reference boundaries.

[0040] No. Canopy temperature stress index of each assessment unit Determine by the following formula: In the formula, Indicates the first Canopy temperature stress indices for each assessment unit; Indicates the first Representative values ​​of canopy temperature for each assessment unit; Indicates the low temperature reference value; This indicates a high-temperature reference value. This indicates the preset minimum temperature difference. To avoid and When they are too close, the denominator is too small. and Used to Limited to the range of 0 to 1. The closer to 1, the more... The more significant the temperature increase of each evaluation unit relative to the reference canopy state within the same acquisition batch, the more pronounced the temperature rise.

[0041] Within the same data collection batch, the representative values ​​of vegetation indices from all participating assessment units are compiled into a set. .right The vegetation index values ​​are arranged in ascending order to obtain Vegetation Index Reference Values Determine by the following formula: In the formula, This indicates the reference value for the vegetation index; express The 90th percentile. When calibrated sample areas or historical observation data exist within the same alfalfa planting area, the vegetation index reference value is... It can be corrected by combining calibration sample area or historical observation data.

[0042] No. Vegetation index deviation index of each assessment unit Determine by the following formula: In the formula, Indicates the first Deviation indicators of vegetation index for each assessment unit; Indicates the first Representative values ​​of vegetation index for each assessment unit; This indicates the reference value for the vegetation index; This indicates the preset minimum vegetation index reference value. To avoid Too small a denominator leads to instability. and Used to Limited to the range of 0 to 1. The closer to 1, the more... The more pronounced the decrease in alfalfa canopy growth relative to the reference level in each assessment unit, the more significant the decline.

[0043] Temperature stress threshold and growth deviates from the threshold This can be determined based on the observation results of the calibrated sample areas. The set of sample areas calibrated as having no significant drought stress is denoted as... The sample set identified as exhibiting significant drought stress is denoted as... The sample area set marked as having significant deviation in growth is denoted as Temperature stress threshold and growth deviates from the threshold Determine them respectively using the following formulas: In the formula, Indicates the temperature stress threshold; This indicates that the growth has deviated from the threshold. Indicates sample area The canopy temperature stress index; Indicates sample area The vegetation index deviation indicator; This represents the mean. Calibration sample areas can be derived from historical observation data, concurrent manual survey results, or experimental sample areas within the same alfalfa growing region. If no calibration sample area is available, the index quantile results of the evaluation units within the same collection batch can be used as a temporary threshold, and updated after obtaining the manual survey results. and .

[0044] Sure , , and Then, the second value is determined according to the binary combination determination rule. Drought stress association status of alfalfa in each assessment unit: like and Then the first One assessment unit was identified as experiencing initial drought stress; like and Then the first One assessment unit was identified as experiencing persistent drought stress; like and Then the first One assessment unit was identified as having abnormal growth; like and Then the first The assessment unit was determined to have no significant drought stress.

[0045] Combination Figure 3 As shown, this embodiment uses the canopy temperature stress index As a horizontal criterion, the vegetation index deviation indicator As a longitudinal criterion. Figure 3 In Indicates the temperature stress threshold. This indicates that the growth has deviated from the threshold; the vertical threshold line corresponds to... The horizontal threshold line corresponds to .when Not achieved and Not achieved At that time, the assessment unit corresponded to no obvious drought stress; when achieve and Not achieved At that time, the assessment unit corresponds to the initial drought stress; when achieve and achieve At that time, the assessment unit corresponds to persistent drought stress; when Not achieved and achieve At that time, the growth of the corresponding assessment unit deviated abnormally.

[0046] In the aforementioned binary combination judgment rule, the canopy temperature stress index is used to form the temperature stress judgment result, and the vegetation index deviation index is used to form the growth deviation judgment result. When the canopy temperature stress index reaches the temperature stress threshold and the vegetation index deviation index does not reach the growth deviation threshold, the corresponding assessment unit is classified as initial drought stress; when both the canopy temperature stress index and the vegetation index deviation index reach the growth deviation threshold, the corresponding assessment unit is classified as persistent drought stress; when both the canopy temperature stress index and the vegetation index deviation index reach the growth deviation threshold, the corresponding assessment unit is classified as abnormal growth deviation; when both the canopy temperature stress index and the vegetation index deviation index do not reach the temperature stress threshold, the corresponding assessment unit is classified as having no obvious drought stress. The above judgment results can express the combined state of the temperature stress judgment result and the growth deviation judgment result at the assessment unit scale.

[0047] In this embodiment, Table 1 lists the evaluation parameter values ​​and sources for a single collection batch. The values ​​in Table 1 are used to illustrate the calculation process and do not constitute a limitation on the scope of protection. Tables 1 to 3 use NDVI as the vegetation index value for calculation.

[0048] Table 1. Explanation of Evaluation Parameter Values ​​and Sources

[0049] Table 2 lists the mean values ​​of the calibration sample areas, used to illustrate the temperature stress threshold. and growth deviates from the threshold The calculations in Table 2. , and Statistical results for different sets of calibrated sample areas.

[0050] Table 2 Threshold Calculation Data for Calibration Sample Area

[0051] According to Table 2, , The threshold calculation above uses only the mean values ​​of the indicators from the calibrated sample areas, without setting any artificial fusion weights.

[0052] Table 3 lists the calculation results for five typical evaluation units from the same sampling batch. Low Temperature Reference Value High temperature reference value and vegetation index reference value The statistical results from all evaluation units participating in the judgment in the same collection batch are not directly derived from the statistics of the five evaluation units in Table 3.

[0053] Table 3. Calculation results of indicators and status determination for typical evaluation units.

[0054] Table 3 shows the percentage of effective canopy area for each assessment unit. All three met the preset validity conditions in Table 1; therefore, A1 to A5 were all included in the determination of the drought stress association status of alfalfa. Taking assessment unit A3 as an example: Reaching the temperature stress threshold , The growth deviation threshold was not reached. A3 was identified as the initial drought stress.

[0055] Taking assessment unit A4 as an example: After being limited, the value is taken as 1.00. Reaching the temperature stress threshold , Reaching the growth deviation threshold A4 was identified as persistent drought stress.

[0056] Taking assessment unit A5 as an example: Temperature stress threshold not reached , Reaching the growth deviation threshold A5 was identified as an abnormal growth deviation. An abnormal growth deviation only indicates a significant deviation in vegetation indices but no significant canopy temperature stress; it does not directly determine the cause of the abnormality.

[0057] In step 500, drought stress zoning assessment results for the alfalfa planting area are formed based on the alfalfa drought stress association status of each assessment unit. Specifically, assessment units with the same alfalfa drought stress association status and spatial adjacency are first merged to obtain drought stress zoning. Then, based on the alfalfa drought stress association status of each assessment unit within the drought stress zoning, the zoning stress type of the drought stress zoning is determined. Finally, based on the drought stress zoning and zoning stress type, the drought stress zoning assessment results for the alfalfa planting area are formed.

[0058] The drought stress zoning assessment results can include the spatial extent and stress type of each drought stress zoning zone. Stress types include initial drought stress, persistent drought stress, abnormal growth deviation, and no obvious drought stress. Assessment units that did not participate in the determination of alfalfa drought stress association status can be separately identified as unassessed areas in the drought stress zoning assessment results to prevent assessment units with insufficient effective canopy area from being included in the drought stress zoning.

[0059] In one possible implementation, drought stress zoning assessment results can serve as the basis for zoning water and fertilizer management operations in alfalfa planting areas. Initial drought stress zones can serve as areas of focus for water status verification and irrigation regulation assessment; persistent drought stress zones can serve as key water management areas; and zones with abnormal growth deviations can be further verified in conjunction with soil nutrient, pest and disease surveys, or post-harvest regeneration status. The above zoning criteria represent the drought stress correlation status and spatial relationship between different regions, and do not directly limit irrigation or fertilization amounts. Irrigation or fertilization amounts can be determined separately based on soil moisture, soil nutrients, historical fertilization records, and management objectives.

[0060] The drought stress assessment method for alfalfa provided in this embodiment ensures that the canopy temperature response in UAV thermal infrared imagery and the vegetation index deviation response in UAV multispectral imagery correspond to the same assessment unit and the same effective alfalfa canopy area. The canopy temperature stress index characterizes the canopy temperature response, while the vegetation index deviation index characterizes the canopy growth deviation response. A binary combination judgment rule is used to distinguish between initial drought stress, persistent drought stress, abnormal growth deviation, and no obvious drought stress. This method can generate drought stress zoning assessment results for alfalfa planting areas without relying on complex models and artificial weight fusion.

[0061] This embodiment also provides a drought stress assessment system for alfalfa based on the fusion of canopy temperature and vegetation index from UAVs. This system can be deployed in computer equipment, edge computing devices, agricultural remote sensing processing platforms, or UAV data processing terminals. After receiving thermal infrared and multispectral images from UAVs, the system generates drought stress zoning assessment results for alfalfa planting areas according to the data processing procedures described in the aforementioned method embodiments.

[0062] like Figure 4 As shown, the system in this embodiment includes an image acquisition module, an evaluation unit determination module, an effective canopy region determination module, an index determination module, a binary combination determination module, and a zoning evaluation module.

[0063] The image acquisition module is used to acquire UAV thermal infrared and multispectral images with location correspondence. The UAV thermal infrared images provide alfalfa canopy temperature data, while the UAV multispectral images provide alfalfa canopy spectral data. The image acquisition module can receive UAV thermal infrared and multispectral images acquired within the same acquisition period; when using images acquired in adjacent acquisition periods, the module receives image data that meets the consistency requirements of alfalfa canopy status. The image acquisition module can also record image acquisition time, spatial range, image resolution, and geolocation information, providing a data foundation for subsequent assessment unit division.

[0064] The assessment unit determination module is connected to the image acquisition module and is used to determine assessment units within the alfalfa planting area according to a preset assessment scale. The module first determines the assessable area of ​​the alfalfa planting area based on the common coverage of UAV thermal infrared and multispectral images, and then divides the assessable area into multiple assessment units according to the preset assessment scale. Each assessment unit corresponds to a thermal infrared image area in the UAV thermal infrared image and a multispectral image area in the UAV multispectral image. The module can also establish a positional correspondence between UAV thermal infrared and multispectral images based on field boundary features, ground control points, stable ground features, or row / ridge boundary features that can be jointly identified by both types of images, ensuring that the thermal infrared and multispectral image areas within the same assessment unit correspond to the same alfalfa canopy area.

[0065] The effective canopy area determination module is connected to the evaluation unit determination module. It is used to determine the effective canopy area of ​​alfalfa within each evaluation unit based on UAV multispectral imagery, and to obtain the corresponding canopy temperature and vegetation index values ​​for the effective canopy area. The effective canopy area determination module can calculate the initial vegetation index value for each pixel within the evaluation unit based on the UAV multispectral imagery, and determine the canopy screening threshold based on the distribution of the initial vegetation index values ​​within the same acquisition batch. The effective canopy area determination module then filters candidate canopy pixels according to the canopy screening threshold, performs connectivity region filtering on the candidate canopy pixels, removes scattered noise pixels and isolated areas that do not meet the area conditions, and obtains the effective canopy area of ​​alfalfa.

[0066] The effective canopy area determination module is also used to calculate the area proportion of the effective canopy area of ​​alfalfa in the corresponding assessment unit. When the area proportion meets the preset validity conditions, the effective canopy area determination module retains the effective canopy area of ​​alfalfa in the corresponding assessment unit and extracts the corresponding canopy temperature value from UAV thermal infrared imagery and the corresponding vegetation index value from UAV multispectral imagery. When the area proportion does not meet the preset validity conditions, the effective canopy area determination module excludes the corresponding assessment unit from the determination process of alfalfa drought stress association status. This module ensures that the canopy temperature value and vegetation index value both originate from the same effective canopy area of ​​alfalfa, reducing the impact of bare soil, field ridges, missing plant areas, and scattered noise pixels on subsequent determination results.

[0067] The index determination module is connected to the effective canopy area determination module. It is used to determine the canopy temperature stress index for each assessment unit based on canopy temperature values, and to determine the vegetation index deviation index for each assessment unit based on vegetation index values. For assessment units participating in the evaluation, the index determination module determines a representative canopy temperature value based on the canopy temperature values ​​within the effective canopy area of ​​alfalfa, and determines low-temperature and high-temperature reference values ​​based on the representative canopy temperature values ​​of each assessment unit within the same collection batch. The index determination module determines the canopy temperature stress index based on the normalized relationship between the representative canopy temperature value and the low-temperature and high-temperature reference values. The canopy temperature stress index is used to characterize the degree of temperature rise of the alfalfa canopy within the assessment unit relative to the reference canopy state.

[0068] The indicator determination module also determines representative vegetation index values ​​based on vegetation index values ​​within the effective canopy area of ​​alfalfa, and reference vegetation index values ​​based on representative vegetation index values ​​from each assessment unit within the same collection batch. The module then determines a vegetation index deviation index based on the degree of deviation between the representative vegetation index value and the reference vegetation index value. This deviation index characterizes the extent to which alfalfa canopy growth within the assessment unit decreases relative to the reference level. The module can limit the amplitude of both the canopy temperature stress index and the vegetation index deviation index, keeping them within preset value ranges to prevent calculation instability caused by abnormal pixels or excessively small reference values.

[0069] The binary combination judgment module is connected to the index determination module. It is used to combine and judge the canopy temperature stress index and vegetation index deviation index according to the binary combination judgment rule formed by the temperature stress threshold and the growth deviation threshold, thus determining the drought stress association status of alfalfa in each assessment unit. The binary combination judgment module compares the canopy temperature stress index with the temperature stress threshold to obtain the temperature stress judgment result; it compares the vegetation index deviation index with the growth deviation threshold to obtain the growth deviation judgment result. The temperature stress judgment result includes whether the temperature stress threshold has been reached or not. The growth deviation judgment result includes whether the growth deviation threshold has been reached or not.

[0070] The binary combination judgment module determines the drought stress correlation status of alfalfa in an assessment unit based on the combination of temperature stress judgment results and growth deviation judgment results. When the canopy temperature stress index reaches the temperature stress threshold and the vegetation index deviation index does not reach the growth deviation threshold, the binary combination judgment module identifies the corresponding assessment unit as experiencing initial drought stress. When both the canopy temperature stress index and the vegetation index deviation index reach the growth deviation threshold, the binary combination judgment module identifies the corresponding assessment unit as experiencing persistent drought stress. When both the canopy temperature stress index and the vegetation index deviation index reach the growth deviation threshold, the binary combination judgment module identifies the corresponding assessment unit as experiencing abnormal growth deviation. When both the canopy temperature stress index and the vegetation index deviation index do not reach the growth deviation threshold, the binary combination judgment module identifies the corresponding assessment unit as experiencing no significant drought stress.

[0071] The zoning assessment module is connected to the binary combination determination module. It is used to generate drought stress zoning assessment results for alfalfa planting areas based on the alfalfa drought stress association status of each assessment unit. The zoning assessment module can merge assessment units with the same alfalfa drought stress association status and spatial adjacency to obtain drought stress zoning. The zoning assessment module then determines the zoning stress type of the drought stress zoning based on the alfalfa drought stress association status of each assessment unit within the drought stress zoning. The drought stress zoning assessment results may include the spatial extent of the drought stress zoning, the zoning stress type, and the identifiers of assessment units not involved in the determination.

[0072] In this embodiment, the image acquisition module, evaluation unit determination module, effective canopy area determination module, index determination module, binary combination judgment module, and zoning evaluation module can be software functional modules running on the same computer device, or they can be distributed across the UAV data acquisition terminal, edge computing device, and agricultural remote sensing processing platform. The data transmitted between modules includes UAV thermal infrared imagery, UAV multispectral imagery, evaluation units, effective canopy area of ​​alfalfa, canopy temperature value, vegetation index value, canopy temperature stress index, vegetation index deviation index, alfalfa drought stress correlation status, and drought stress zoning evaluation results. After the above data is transmitted sequentially, the system can generate drought stress zoning evaluation results for alfalfa planting areas without relying on complex models and manual weight fusion.

[0073] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the systems disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the descriptions are relatively simple; relevant parts can be referred to the method section.

[0074] Specific examples have been used to illustrate the principles and implementation methods of this invention. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of this invention. Furthermore, those skilled in the art will recognize that, based on the ideas of this invention, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of this invention.

Claims

1. A method for assessing drought stress in alfalfa based on the fusion of canopy temperature and vegetation index from unmanned aerial vehicles (UAVs), characterized in that, include: Acquire UAV thermal infrared and UAV multispectral images with location correspondence, and determine the assessment units within the alfalfa planting area according to the preset assessment scale; The effective canopy area of ​​alfalfa in each evaluation unit is determined based on the UAV multispectral imagery, and the canopy temperature value and vegetation index value corresponding to the effective canopy area of ​​alfalfa are obtained. The canopy temperature stress index of each assessment unit is determined based on the canopy temperature value, and the vegetation index deviation index of each assessment unit is determined based on the vegetation index value. According to the binary combination judgment rule formed by the temperature stress threshold and the growth deviation threshold, the canopy temperature stress index and the vegetation index deviation index are combined for judgment to determine the drought stress association status of alfalfa in each assessment unit; the binary combination judgment rule includes four types of judgment results: initial drought stress, continuous drought stress, abnormal growth deviation and no obvious drought stress. Based on the drought stress correlation status of alfalfa in each assessment unit, the drought stress zoning assessment results of the alfalfa planting area are formed.

2. The method for assessing drought stress in alfalfa based on the fusion of canopy temperature and vegetation index from unmanned aerial vehicles, as described in claim 1, is characterized in that... Acquire location-corresponding UAV thermal infrared and multispectral images, and determine assessment units within the alfalfa planting area according to a preset assessment scale, including: Determine the common coverage area of ​​the UAV thermal infrared image and the UAV multispectral image within the alfalfa planting area; The common coverage area is divided into multiple evaluation units according to the preset evaluation criteria; Within each evaluation unit, the thermal infrared image region corresponding to the UAV thermal infrared image and the multispectral image region corresponding to the UAV multispectral image are determined respectively.

3. The method for assessing drought stress in alfalfa based on the fusion of canopy temperature and vegetation index from unmanned aerial vehicles, as described in claim 2, is characterized in that... Within each evaluation unit, the thermal infrared image region corresponding to the UAV thermal infrared image and the multispectral image region corresponding to the UAV multispectral image are determined, including: Based on field boundary features, row boundary features, or ground control points, establish the positional correspondence between the UAV thermal infrared image and the UAV multispectral image; According to the aforementioned positional correspondence, each evaluation unit is respectively assigned to the thermal infrared image region and the multispectral image region; The thermal infrared image region and the multispectral image region are made to correspond to the same alfalfa canopy region.

4. The method for assessing drought stress in alfalfa based on the fusion of canopy temperature and vegetation index from unmanned aerial vehicles, as described in claim 1, is characterized in that... The effective canopy area of ​​alfalfa within each evaluation unit is determined based on the UAV multispectral imagery, and the canopy temperature and vegetation index values ​​corresponding to the effective canopy area of ​​alfalfa are obtained, including: Calculate the initial value of the vegetation index in each evaluation unit based on the UAV multispectral imagery; Filter alfalfa canopy pixels in each evaluation unit according to the initial value of the vegetation index; Connectivity region filtering is performed on the alfalfa canopy pixels to obtain the effective canopy region of alfalfa in each evaluation unit; Based on the location of the effective canopy area of ​​alfalfa in the UAV thermal infrared image and the UAV multispectral image, the canopy temperature value and the vegetation index value of the corresponding evaluation unit are extracted respectively.

5. The method for assessing drought stress in alfalfa based on the fusion of UAV canopy temperature and vegetation index as described in claim 4, characterized in that, Connectivity region filtering is performed on the alfalfa canopy pixels to obtain the effective alfalfa canopy region within each evaluation unit, including: The connected regions formed by the alfalfa canopy pixels are filtered by area, and connected regions that do not meet the preset area conditions are removed. Calculate the area percentage of the retained connected regions in the corresponding evaluation unit; When the area ratio meets the preset validity conditions, the retained connected region will be regarded as the effective canopy region of alfalfa in the corresponding evaluation unit; When the area percentage does not meet the preset validity condition, the corresponding evaluation unit will be excluded from the determination process of the alfalfa drought stress association state.

6. The method for assessing drought stress in alfalfa based on the fusion of canopy temperature and vegetation index from unmanned aerial vehicles, as described in claim 1, is characterized in that... Based on the canopy temperature values, canopy temperature stress indices for each assessment unit are determined, and based on the vegetation index values, vegetation index deviation indices for each assessment unit are determined, including: Based on the canopy temperature values ​​of each assessment unit, determine the representative canopy temperature value for each assessment unit; Based on the representative canopy temperature values ​​of each evaluation unit within the same data collection period, determine the low-temperature reference value and the high-temperature reference value; The canopy temperature stress index for each assessment unit is determined based on the normalization relationship between the representative canopy temperature value and the low-temperature reference value and the high-temperature reference value.

7. The method for assessing drought stress in alfalfa based on the fusion of canopy temperature and vegetation index from unmanned aerial vehicles, as described in claim 1, is characterized in that... Based on the canopy temperature values, canopy temperature stress indices for each assessment unit are determined, and based on the vegetation index values, vegetation index deviation indices for each assessment unit are determined, including: Based on the vegetation index values ​​of each assessment unit, determine the representative value of the vegetation index for each assessment unit. Based on the representative values ​​of the vegetation index of each assessment unit within the same data collection period, determine the reference value of the vegetation index; The vegetation index deviation index for each assessment unit is determined based on the degree of deviation of the representative value of the vegetation index from the reference value of the vegetation index.

8. The method for assessing drought stress in alfalfa based on the fusion of canopy temperature and vegetation index from unmanned aerial vehicles, as described in claim 1, is characterized in that... According to the binary combination judgment rule formed by the temperature stress threshold and the growth deviation threshold, the canopy temperature stress index and the vegetation index deviation index are combined for judgment to determine the drought stress association status of alfalfa in each assessment unit, including: Based on the comparison results between the canopy temperature stress index and the temperature stress threshold, the temperature stress judgment result of each evaluation unit is determined. The temperature stress judgment result includes reaching the temperature stress threshold and not reaching the temperature stress threshold. Based on the comparison results between the vegetation index deviation index and the growth deviation threshold, the growth deviation judgment result of each assessment unit is determined. The growth deviation judgment result includes reaching the growth deviation threshold and not reaching the growth deviation threshold. Based on the binary combination of the temperature stress determination result and the growth deviation determination result, the drought stress correlation status of alfalfa in each evaluation unit is determined.

9. The method for assessing drought stress in alfalfa based on the fusion of canopy temperature and vegetation index from unmanned aerial vehicles, as described in claim 8, is characterized in that... Based on the binary combination of the temperature stress determination result and the growth deviation determination result, the drought stress association status of alfalfa in each evaluation unit is determined, including: When the temperature stress determination result indicates that the canopy temperature stress index has reached the temperature stress threshold, and the growth deviation determination result indicates that the vegetation index deviation index has not reached the growth deviation threshold, the corresponding assessment unit is determined to be under initial drought stress. When the temperature stress determination result indicates that the canopy temperature stress index has reached the temperature stress threshold, and the growth deviation determination result indicates that the vegetation index deviation index has reached the growth deviation threshold, the corresponding assessment unit will be identified as under continuous drought stress. When the temperature stress determination result indicates that the canopy temperature stress index has not reached the temperature stress threshold, and the growth deviation determination result indicates that the vegetation index deviation index has reached the growth deviation threshold, the corresponding evaluation unit will be determined as having abnormal growth deviation. When the temperature stress determination result indicates that the canopy temperature stress index has not reached the temperature stress threshold, and the growth deviation determination result indicates that the vegetation index deviation index has not reached the growth deviation threshold, the corresponding evaluation unit is determined to be without obvious drought stress.

10. The method for assessing drought stress in alfalfa based on the fusion of canopy temperature and vegetation index from unmanned aerial vehicles, as described in claim 1, is characterized in that... Based on the drought stress correlation status of alfalfa in each assessment unit, the drought stress zoning assessment results of the alfalfa planting area are formed, including: By merging assessment units that have the same alfalfa drought stress association status and are spatially adjacent, drought stress partitions are obtained. Based on the alfalfa drought stress association status of each assessment unit within the drought stress zone, the zone stress type of the drought stress zone is determined. Based on the drought stress zones and the stress types of the zones, the drought stress zone assessment results for the alfalfa planting area are formed. The drought stress zoning assessment results will be used as the zoning basis for water and fertilizer management operations in the alfalfa planting area.

Images