Method and system for early warning of sclerotinia stem rot of rape based on remote sensing characterization data

Abstract

This invention discloses a method and system for early warning of rapeseed sclerotinia stem rot risk based on remote sensing characterization data, belonging to the field of agricultural remote sensing monitoring technology. The method includes: acquiring multi-temporal remote sensing images of rapeseed planting areas to construct rapeseed remote sensing time-series status data; determining the level of petal infection exposure; identifying canopy wetting events, determining the canopy's adaptability to infection based on the hydrothermal response recovery process after wetting, and matching it with the petal infection exposure time interval; identifying abnormal disease responses that are time-related to the infection process based on the reference response of normal phenological stages; and determining the sclerotinia stem rot risk level by comprehensively considering petal infection exposure data, canopy adaptability to infection data, and abnormal disease response data, thus obtaining a rapeseed sclerotinia stem rot risk warning result. This invention can distinguish between normal flowering period changes and sclerotinia stem rot infection-related processes, and characterize the sustained ability of canopy wetting conditions, improving the advance warning and reliability of rapeseed sclerotinia stem rot risk warning.

Description

Technical Field

[0001] This invention relates to the field of agricultural remote sensing monitoring technology, specifically to a method and system for early warning of rapeseed sclerotinia stem rot risk based on remote sensing characterization data. Background Technology

[0002] Sclerotinia rot is a common disease in rapeseed production. Its occurrence and development are closely related to the rapeseed's growth stage, the state of the infection vector, and the local hydrothermal environment of the canopy. After sclerotinia rot occurs, it can damage the rapeseed stems, leaves, and pods, affecting normal plant growth and yield formation. Therefore, monitoring the risk of sclerotinia rot and providing early warnings in rapeseed planting areas is of great significance for the timely implementation of disease control measures.

[0003] With the development of remote sensing technology, technologies such as multispectral remote sensing, thermal infrared remote sensing, and UAV remote sensing are gradually being applied to crop disease monitoring. Existing methods for monitoring rapeseed sclerotinia stem rot typically extract remote sensing information such as vegetation indices, spectral reflectance characteristics, canopy temperature, or water content to identify color changes, tissue damage, or canopy physiological abnormalities in plants after disease occurrence. Some methods also combine environmental data such as rainfall, air humidity, and temperature to assess the risk of sclerotinia stem rot occurrence.

[0004] However, the infection and formation of rapeseed sclerotinia stem rot is a distinct process, especially during the rapeseed flowering period. Petal shedding and its retention within the canopy can create carriers for the pathogen, while a persistently moist canopy condition affects the formation and maintenance of infection conditions. Existing remote sensing monitoring methods primarily focus on abnormal responses to existing diseases, lacking continuous characterization of the exposure process of infection carriers and the suitable canopy environment before disease formation, which can easily lead to delayed risk identification.

[0005] On the other hand, rapeseed exhibits significant phenological changes in flower quantity, canopy color, and spectral response during flowering. Normal remote sensing changes during the transition from full bloom to petal shedding can easily be confused with abnormal responses caused by diseases. Furthermore, relying solely on rainfall, air humidity, or canopy moisture status at a single time phase is insufficient to accurately reflect the actual sustained capacity of different rapeseed canopies to maintain suitable infection conditions after wetting. Therefore, how to utilize multi-temporal remote sensing data to extract effective characteristics related to the formation process of sclerotinia stem rot infection from normal rapeseed flowering changes, and how to combine this with the sustained state of canopy moisture conditions and subsequent abnormal disease responses for risk assessment, has become a key technical problem to be solved in the remote sensing early warning process for rapeseed sclerotinia stem rot. Summary of the Invention

[0006] In view of the above situation and to overcome the shortcomings of the prior art, the technical solution adopted by the present invention is as follows: The present invention provides a method for early warning of rapeseed sclerotinia stem rot risk based on remote sensing characterization data, the method comprising the following steps:

[0007] Step S1: Rapeseed remote sensing state construction;

[0008] Step S2: Characterization of petal infection exposure;

[0009] Step S3: Assessment of canopy adaptation and maintenance;

[0010] Step S4: Identification of abnormal disease responses;

[0011] Step S5: Early warning of sclerotinia stem rot risk.

[0012] Further, in step S1, the rapeseed remote sensing state construction is used to form a basic remote sensing characterization of the rapeseed planting area under different observation time phases. Specifically, it involves acquiring multi-temporal remote sensing images of the target rapeseed planting area, performing radiometric correction, geometric correction, spatial registration, and rapeseed planting area extraction on the multi-temporal remote sensing images, dividing the rapeseed planting area into spatial units based on a set spatial scale, extracting the flowering period characterization, canopy structure characterization, spectral moisture characterization, and thermal response characterization corresponding to each spatial unit, and associating them according to the observation time phase to obtain rapeseed remote sensing time-series state data.

[0013] Further, in step S2, the petal infection exposure characterization is used to identify the petal shedding process related to the exposure of sclerotinia rot infection vector from the continuous change of rapeseed flowering period, and to determine the degree to which different spatial units have the conditions for infection vector exposure. Specifically, based on the rapeseed remote sensing time-series state data, the continuous change characteristics of flower quantity in each spatial unit are extracted, the flowering period transition position where the flower quantity changes from increasing to decreasing is determined, the petal reduction process after the flowering period transition is analyzed in time series, and combined with the spatial carrying capacity of the canopy for the fallen petals, the degree to which the fallen petals form the conditions for infection vector exposure inside the canopy is determined, and petal infection exposure data is obtained.

[0014] The petal infection exposure characterization includes the following steps:

[0015] Step S21: Construction of flower quantity time-series response, used to characterize the continuous change process of rapeseed flower quantity with observation time phase in different spatial units. Specifically, based on the rapeseed remote sensing time-series state data, flower period features corresponding to the spectral response of rapeseed petals are extracted, and the petal coverage degree in the spatial unit is determined according to the flower period features corresponding to each observation time. The petal coverage degree under continuous observation time phases is arranged in chronological order to obtain flower quantity time-series response data.

[0016] Step S22: Flowering period transition location identification, used to determine the key time phase of rapeseed transitioning from full bloom to continuous petal shedding. Specifically, it involves calculating the direction of flower quantity change between adjacent observation time phases based on the flower quantity time sequence response data, identifying candidate time phases where the flower quantity changes from an increasing state to a decreasing state, and determining the corresponding candidate time phase as the flowering period transition location based on the flower quantity time sequence response data.

[0017] Step S23: Petal shedding response extraction, used to extract the petal reduction process that has significance for the exposure of infection carriers after the flowering period change process. Specifically, based on the flowering period change data, the petal shedding analysis interval corresponding to each spatial unit is determined, the amount of petal reduction is determined according to the difference in petal coverage between adjacent observation phases within the analysis interval, and the degree of petal shedding response is determined according to the amount of petal reduction and its rate of change within the continuous observation interval, so as to obtain petal shedding response data.

[0018] Step S24: Canopy bearing capacity assessment, used to determine the spatial conditions under which fallen petals are received and retained by the canopy surface after entering the rapeseed canopy. Specifically, based on the rapeseed remote sensing time-series state data, the canopy coverage and canopy porosity characterization within the petal fall analysis interval are extracted. The canopy surface distribution that can be used to receive fallen petals is determined based on the canopy coverage. The spatial permeability of fallen petals passing through the canopy is determined based on the canopy porosity characterization. The two are combined to determine the relative degree of reception and retention of fallen petals inside the canopy, thus obtaining canopy bearing capacity data.

[0019] Step S25: Infection exposure determination, used to distinguish between a simple decrease in flower quantity and a petal shedding process with exposure conditions for sclerotinia rot infection vectors. Specifically, based on the petal shedding response data and canopy bearing capacity data, the degree of petal shedding response and canopy bearing capacity are correlated. When the petal shedding process and canopy bearing conditions coexist in the same petal shedding analysis interval, the infection exposure level of the corresponding spatial unit is determined based on the degree of petal shedding response and canopy bearing capacity, thus obtaining petal infection exposure data.

[0020] Further, in step S3, the canopy adaptability maintenance assessment is used to determine the ability of the rapeseed canopy to maintain the moist conditions related to sclerotinia infection after the canopy has formed a moist state. Specifically, it identifies the changes in thermal response and spectral moisture characterization before and after the formation of the canopy moist state based on the rapeseed remote sensing time-series state data, establishes the response trajectory of each spatial unit from the moist state to the reference state before the moist state, and determines the canopy moist condition maintenance ability based on the response recovery time, response deviation decay rate and the consistency of the recovery direction, thus obtaining canopy adaptability maintenance data.

[0021] The assessment of canopy adaptability and maintenance includes the following steps:

[0022] Step S31: Canopy wetting event identification, used to determine the wetting process in which the canopy hydrothermal state changes in tandem. Specifically, based on the rapeseed remote sensing time-series state data, thermal response characterization and spectral moisture characterization are obtained under continuous observation time phases. When the canopy temperature decrease and the spectral moisture increase occur in the same observation time phase or adjacent observation time phases, the corresponding change process is identified as a canopy wetting event, and canopy wetting event data is obtained.

[0023] Step S32: Determine the wet reference state, which is used to establish a comparative benchmark for the recovery process after the canopy wets. Specifically, based on the canopy wet event data, determine the thermal response characterization and spectral moisture characterization of multiple consecutive observation phases before the wet event occurs. Based on the response change range of the multiple consecutive observation phases, determine the reference state before wets and the corresponding reference fluctuation range. Then, determine the first observation phase after the wet event response reaches the extreme value as the starting phase of the recovery analysis to obtain the wet reference state data.

[0024] Step S33: Canopy recovery trajectory construction, used to characterize the continuous change process of canopy hydrothermal response recovering to the pre-wetting reference state after the formation of the wet state. Specifically, starting from the initial phase of the recovery analysis, the thermal response characterization and spectral moisture characterization of the corresponding spatial unit in the subsequent observation phases are extracted, the degree of response deviation of each observation phase relative to the pre-wetting reference state is calculated, and they are correlated in chronological order to obtain canopy recovery trajectory data.

[0025] Step S34: Extraction of invasion-resistant features to distinguish between short-term and continuous wet states. Specifically, based on the canopy recovery trajectory data, the time required for the thermal response characterization and spectral moisture characterization to recover to the reference fluctuation range is determined. Based on the response deviation decay rate during the recovery process and the consistency of the recovery directions of the two types of responses, the continuous wet state characteristics are determined to obtain invasion-resistant feature data.

[0026] Step S35: Determine the ability to maintain canopy moisture related to sclerotinia infection. Specifically, determine the moisture maintenance level of each spatial unit based on the aforementioned ability to maintain canopy moisture. Match the moisture maintenance time interval with the petal infection exposure time interval determined in step S2. When the two time intervals overlap, determine the degree of canopy moisture maintenance of the corresponding spatial unit based on the overlap duration and the moisture maintenance level to obtain canopy moisture maintenance data.

[0027] Further, in step S4, the disease abnormal response identification is used to identify abnormal remote sensing responses of different spatial units relative to the normal growth changes of rapeseed. Specifically, it involves selecting reference spatial units whose phenological stage differences are within a set matching range based on the rapeseed remote sensing time-series state data, establishing reference response intervals for corresponding phenological stages based on the spectral moisture characterization, canopy structure characterization, and thermal response characterization of the reference spatial units, comparing the remote sensing response of the target spatial unit with the reference response intervals, and determining the corresponding response as a candidate abnormal response of disease based on the remote sensing response of the target spatial unit for two consecutive observation phases exceeding the corresponding reference response interval. Furthermore, it involves screening candidate abnormal responses of disease that are time-related to the canopy moisture maintenance process after the occurrence of petal infection exposure data and canopy adaptability maintenance data, thereby obtaining disease abnormal response data.

[0028] Further, in step S5, the sclerotinia rot risk warning is used to comprehensively determine the sclerotinia rot occurrence risk in different spatial units within the rapeseed planting area. Specifically, it determines the infection exposure index, the canopy suitability maintenance index, and the disease abnormality index based on the petal infection exposure data, the canopy suitability maintenance data, and the disease abnormality response data, respectively. After normalizing each index, they are combined according to a set weight to obtain the sclerotinia rot risk assessment value. Based on the comparison between the sclerotinia rot risk assessment value and the risk level threshold, the risk level of each spatial unit is determined, generating sclerotinia rot risk spatial distribution information, and obtaining the rapeseed sclerotinia rot risk warning result.

[0029] The rapeseed sclerotinia stem rot risk early warning system based on remote sensing characterization data provided by the present invention includes a remote sensing status construction module, a petal infection and exposure module, a canopy adaptability maintenance module, a disease anomaly identification module, and a risk early warning module;

[0030] The remote sensing status construction module is used to form a basic remote sensing characterization of the rapeseed planting area under different observation time phases. Through the rapeseed remote sensing status construction, rapeseed remote sensing time series status data is obtained, and the rapeseed remote sensing time series status data is sent to the petal infection exposure module.

[0031] The petal infection and exposure module is used to identify the petal shedding process related to the exposure of the sclerotinia rot infection vector. Through petal infection and exposure characterization, petal infection and exposure data are obtained, and the rapeseed remote sensing time-series status data and petal infection and exposure data are sent to the canopy adaptability maintenance module.

[0032] The canopy adaptability maintenance module is used to determine the ability of the canopy to maintain the moist conditions related to sclerotinia infection after the canopy has formed a moist state. Through the canopy adaptability maintenance assessment, canopy adaptability maintenance data is obtained, and the rapeseed remote sensing time-series status data, petal infection exposure data and canopy adaptability maintenance data are sent to the disease anomaly identification module.

[0033] The disease anomaly identification module is used to identify abnormal remote sensing responses that are relative to the normal growth changes of rapeseed and are time-related to the infection process. Through disease anomaly response identification, disease anomaly response data is obtained, and the petal infection exposure data, canopy adaptability maintenance data and disease anomaly response data are sent to the risk warning module.

[0034] The risk warning module is used to comprehensively determine the risk of sclerotinia disease occurrence in different spatial units, obtain the risk warning result of rapeseed sclerotinia disease through sclerotinia disease risk warning, and output the risk warning result of rapeseed sclerotinia disease.

[0035] The beneficial effects achieved by the present invention using the above solution are as follows:

[0036] (1) In view of the problem that the existing remote sensing early warning of sclerotinia stem rot in rapeseed mainly relies on the obvious remote sensing anomalies of the disease or external meteorological risk factors, which leads to the early warning time being too late and easily affected by the normal flowering period, this scheme adopts a risk early warning method that combines the construction of rapeseed remote sensing time sequence status, characterization of infection carrier exposure, assessment of canopy suitability for infection, and identification of abnormal disease response. By extracting the exposure conditions of infection carrier from the rapeseed flowering period, and further combining the canopy moisture retention status and subsequent abnormal remote sensing response for risk judgment, the scheme realizes the transformation from the identification of the obvious disease results to the early warning of the disease formation process, thereby improving the advance and reliability of sclerotinia stem rot risk early warning. For example, in the late flowering stage before obvious lesions have formed, risk spatial units can be identified based on the co-occurrence of continuous petal shedding, canopy acceptance status, and continuous canopy moisture retention.

[0037] (2) In view of the obvious spectral changes in rapeseed during flowering, the decline in flower quantity is easily confused with the changes in canopy caused by diseases, and it is difficult to determine whether the conditions for exposure to sclerotinia rot infection vectors are met by simply using the flowering state. This scheme adopts a petal infection exposure characterization mechanism based on flowering period transition, petal shedding response and canopy bearing capacity. By first determining the transition position from full bloom to shedding stage, and then judging the degree of shedding petals being received and retained in the canopy based on the canopy coverage and canopy porosity characterization, and linking the petal shedding response with the canopy bearing capacity, the ordinary flower quantity change is converted into a process characterization related to the exposure of sclerotinia rot infection vectors, so as to distinguish between the general decline in flower quantity and the process of infection vector exposure. For example, for two spatial units with the same decline in flower quantity but different canopy spatial structures, different petal infection exposure levels can be obtained based on the permeability of the shedding petals through the canopy and the differences in the distribution of the canopy bearing surface.

[0038] (3) To address the problem that traditional methods of using rainfall, air humidity, or single-phase canopy moisture status are insufficient to reflect the actual duration of rapeseed canopy wetness, this scheme adopts an assessment mechanism for canopy wetness maintenance based on the canopy hydrothermal recovery trajectory after a wet event. By establishing a reference state and reference fluctuation range before wetness based on multiple observation phases before the wet event, the scheme continuously tracks the recovery process of thermal response characterization and spectral moisture characterization to the reference fluctuation range. The scheme determines the canopy wetness maintenance capacity based on the recovery time, response deviation decay rate, and the degree of consistency between the two responses. Furthermore, it is matched with the petal infection exposure time interval to distinguish between short-term wetness and continuous wetness. For example, when two spatial units have similar initial wetness responses, different wetness maintenance levels can be determined based on the time required for them to recover to the reference fluctuation range before wetness and the degree of overlap with the petal infection exposure time interval. Attached Figure Description

[0039] Figure 1 A flowchart illustrating the method for early warning of rapeseed sclerotinia stem rot risk based on remote sensing characterization data provided by the present invention;

[0040] Figure 2 A schematic diagram of the rapeseed sclerotinia stem rot risk early warning system based on remote sensing characterization data provided by the present invention;

[0041] Figure 3 This is a schematic diagram of the process for characterizing petal infection and exposure in step S2;

[0042] Figure 4 This is a flowchart illustrating the process for assessing canopy adaptability and maintenance in step S3.

[0043] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used together with the embodiments of the invention to explain the invention and do not constitute a limitation thereof. Detailed Implementation

[0044] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0045] In the description of this invention, it should be understood that the terms "upper", "lower", "front", "rear", "left", "right", "top", "bottom", "inner", "outer", etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this invention.

[0046] Example 1, see Figure 1 The present invention provides a method for early warning of rapeseed sclerotinia stem rot risk based on remote sensing characterization data. The method includes the following steps:

[0047] Step S1: Rapeseed remote sensing state construction;

[0048] Step S2: Characterization of petal infection exposure;

[0049] Step S3: Assessment of canopy adaptation and maintenance;

[0050] Step S4: Identification of abnormal disease responses;

[0051] Step S5: Early warning of sclerotinia stem rot risk.

[0052] By performing the above operations, this scheme addresses the problem that existing remote sensing early warning systems for rapeseed sclerotinia stem rot mainly rely on overt remote sensing anomalies or external meteorological risk factors, leading to delayed warning times and susceptibility to changes in normal flowering periods. Instead, it adopts a risk warning approach that combines rapeseed remote sensing temporal state construction, infectious carrier exposure characterization, canopy suitability assessment, and identification of abnormal disease responses. By extracting infectious carrier exposure conditions from the rapeseed flowering period and further combining canopy moisture retention status and subsequent abnormal remote sensing responses for risk assessment, this scheme shifts from identifying overt disease results to early warning of the disease formation process, improving the advance notice and reliability of sclerotinia stem rot risk warnings. For example, in the late flowering stage before obvious lesions form, risk spatial units can be identified based on the co-occurrence of continuous petal shedding, canopy acceptance status, and sustained canopy moisture retention.

[0053] Example 2, see Figure 1 and Figure 2This embodiment is based on the above embodiment. In step S1, the rapeseed remote sensing state construction is used to form a basic remote sensing characterization of the rapeseed planting area under different observation time phases. Specifically, it involves collecting multi-temporal remote sensing images of the target rapeseed planting area, performing radiometric correction, geometric correction, spatial registration, and rapeseed planting area extraction on the multi-temporal remote sensing images, dividing the rapeseed planting area into spatial units based on a set spatial scale, extracting the flowering period characterization, canopy structure characterization, spectral moisture characterization, and thermal response characterization corresponding to each spatial unit, and associating them according to the observation time phase to obtain rapeseed remote sensing time series state data.

[0054] In specific implementation, the multi-temporal remote sensing images utilize multispectral and thermal infrared remote sensing images acquired over continuous observation phases of the same target rapeseed planting area. The multispectral images are used to characterize flowering period, canopy structure, and spectral moisture content, while the thermal infrared images are used to characterize thermal response. Preferably, the multispectral images are used as the spatial registration reference, and the thermal infrared images are registered to the same spatial coordinate system according to corresponding ground control points. The images, after radiometric correction, geometric correction, and spatial registration, are then unified to the same spatial resolution. After the rapeseed planting area is extracted, the rapeseed planting area is divided into sections with a side length of 1... The spatial unit is a square unit of m, the scale of which can be adjusted according to the ground resolution of the remote sensing image, so that each spatial unit contains no less than 25 effective remote sensing pixels. For each spatial unit, the flowering period is characterized by the difference in reflectance of rapeseed flowers in the visible light band, the canopy structure is characterized by the continuity of vegetation distribution within the spatial unit, the spectral moisture is characterized by the difference in reflectance between the near-infrared band and the moisture-sensitive band, and the thermal response is characterized by the average temperature of the corresponding effective pixels in the thermal infrared remote sensing image. Preferably, the same spatial unit boundary and characterization extraction rules are used for continuous observation phases, and the flowering period characterization, canopy structure characterization, spectral moisture characterization and thermal response characterization of the same spatial unit under different observation phases are associated with the observation time as an index to form the rapeseed remote sensing time-series state data.

[0055] Example 3, see Figure 1 , Figure 2 and Figure 3This embodiment is based on the above embodiment. In step S2, the petal infection exposure characterization is used to identify the petal shedding process related to the exposure of sclerotinia rot infection vector from the continuous change of rapeseed flowering period, and to determine the degree to which different spatial units have the conditions for infection vector exposure. Specifically, based on the rapeseed remote sensing time series status data, the continuous change characteristics of flower quantity in each spatial unit are extracted, the flowering period transition position where the flower quantity changes from increasing to decreasing is determined, the petal reduction process after the flowering period transition is analyzed in time series, and combined with the spatial carrying capacity of the canopy for the fallen petals, the degree to which the fallen petals form the conditions for infection vector exposure inside the canopy is determined, and petal infection exposure data is obtained.

[0056] The petal infection exposure characterization includes the following steps:

[0057] Step S21: Construction of flower quantity time-series response, used to characterize the continuous change process of rapeseed flower quantity with observation time phase in different spatial units. Specifically, based on the rapeseed remote sensing time-series state data, flower period features corresponding to the spectral response of rapeseed petals are extracted, and the petal coverage degree in the spatial unit is determined according to the flower period features corresponding to each observation time. The petal coverage degree under continuous observation time phases is arranged in chronological order to obtain flower quantity time-series response data.

[0058] In practice, the same pixel recognition rule is used for the flowering period characteristics corresponding to each observation time. The petal response pixels are determined based on the spectral response differences between rapeseed petal and non-petal vegetation areas. Preferably, the Otsu method is used to determine the segmentation threshold for the flowering period characteristic values ​​within the rapeseed planting area at the current observation time, and the petal coverage is determined according to the ratio of the number of petal response pixels to the number of effective rapeseed pixels in the corresponding spatial unit, specifically calculated using the following formula:

[0059] ;

[0060] In the formula, This indicates that the i-th spatial unit is in the k-th observation phase. The corresponding petal coverage, where i represents the spatial unit number. This indicates the k-th observation phase, where k represents the observation phase number. Indicates the i-th spatial unit at the time of observation The number of petal response pixels within the range, where the subscript p indicates the number of petal response pixels; Indicates the i-th spatial unit at the time of observation The number of effective pixels of rapeseed within the same spatial unit, with the subscript 'o' indicating the effective pixels of rapeseed; the corresponding values ​​of the same spatial unit at each observation time. Arrange them in chronological order to form the flower quantity time-series response sequence of the spatial unit, and summarize the flower quantity time-series response sequences corresponding to each spatial unit to obtain the flower quantity time-series response data;

[0061] Step S22: Flowering period transition location identification, used to determine the key time phase of rapeseed transitioning from full bloom to continuous petal shedding. Specifically, it involves calculating the direction of flower quantity change between adjacent observation time phases based on the flower quantity time sequence response data, identifying candidate time phases where the flower quantity changes from an increasing state to a decreasing state, and determining the corresponding candidate time phase as the flowering period transition location based on the flower quantity time sequence response data.

[0062] In practice, the petal coverage at adjacent observation times is compared chronologically. A higher petal coverage at a later observation time is considered an increasing state, while a lower petal coverage at a later observation time is considered a decreasing state. When an observation time initially shows an increasing state and subsequently changes to a decreasing state, that observation time is designated as a candidate time. The next two consecutive observation intervals following the candidate time are then examined. If both consecutive observation intervals maintain a decreasing state, the corresponding candidate time is designated as the flowering transition position. When multiple candidate time periods meet the conditions, the candidate time period that first meets the above conditions in chronological order is selected. This represents the observation phase of the flowering period transition location corresponding to the i-th spatial unit, and the corresponding spatial units are... The flowering period conversion data is obtained by associating the flowering period data with its spatial unit number;

[0063] Step S23: Petal shedding response extraction, used to extract the petal reduction process that has significance for the exposure of infection carriers after the flowering period change process. Specifically, based on the flowering period change data, the petal shedding analysis interval corresponding to each spatial unit is determined, the amount of petal reduction is determined according to the difference in petal coverage between adjacent observation phases within the analysis interval, and the degree of petal shedding response is determined according to the amount of petal reduction and its rate of change within the continuous observation interval, so as to obtain petal shedding response data.

[0064] In practical implementation, the flowering period transition location of the corresponding spatial unit is used. The starting point of the petal shedding analysis interval is used as the starting point, and the petal coverage is checked sequentially over time. When two consecutive observation intervals no longer show a decreasing trend, the starting observation phase of the first observation interval is determined as the ending point of the petal shedding analysis interval. For the observation intervals within the petal shedding analysis period where petal coverage decreased, the difference between the petal coverage at the previous observation time and the petal coverage at the next observation time was used as the corresponding petal reduction. The petal reduction was then divided by the duration of the corresponding observation interval to obtain the petal reduction rate. The cumulative petal reduction and the average petal reduction rate within the petal shedding analysis period were normalized to the maximum values ​​in the same observation batch, and the petal shedding response was determined with equal weights, specifically calculated using the following formula:

[0065] ;

[0066] In the formula, This represents the petal shedding response level corresponding to the i-th spatial unit, where i represents the spatial unit number; the first 0.5 represents the weight corresponding to the cumulative petal reduction. This represents the sum of the petal reduction amounts corresponding to each descent observation interval within the petal shedding analysis interval for the i-th spatial unit, denoted by the symbol. Indicates the amount of change in the degree of petal coverage; This indicates that all valid spatial units in the same observation batch correspond to The maximum value in the range; the second 0.5 represents the weight corresponding to the average rate of decrease in petals; This represents the average rate of petal reduction for the i-th spatial unit across all observation intervals within the petal shedding analysis interval. The subscript indicates the average rate of petal reduction. This represents the average value; This indicates that all valid spatial units in the same observation batch correspond to The maximum value in; when or When the value is 0, the corresponding normalization term is set to 0; and the corresponding petal shedding analysis interval The petal fall-off response results of the i-th spatial unit are correlated and summarized to obtain the petal fall-off response data.

[0067] Step S24: Canopy bearing capacity assessment, used to determine the spatial conditions under which fallen petals are received and retained by the canopy surface after entering the rapeseed canopy. Specifically, based on the rapeseed remote sensing time-series state data, the canopy coverage and canopy porosity characterization within the petal fall analysis interval are extracted. The canopy surface distribution that can be used to receive fallen petals is determined based on the canopy coverage. The spatial permeability of fallen petals passing through the canopy is determined based on the canopy porosity characterization. The two are combined to determine the relative degree of reception and retention of fallen petals inside the canopy, thus obtaining canopy bearing capacity data.

[0068] In practical implementation, the petal shedding analysis interval for the corresponding spatial unit is determined from the petal shedding response data, and the effective canopy area is determined based on the canopy structure characterization within this interval. The canopy coverage is determined by the ratio of the area of ​​the effective canopy area to the area of ​​the corresponding spatial unit. Within the spatial outer envelope of the effective canopy area, the ratio of the area of ​​the non-canopy region continuously connected to the external background region to the area of ​​the spatial outer envelope is determined as the canopy porosity characterization. The average canopy coverage and average canopy porosity characterization within the petal shedding analysis interval are calculated respectively, and the canopy bearing capacity is determined according to the following formula:

[0069] ;

[0070] In the formula, This represents the canopy bearing capacity corresponding to the i-th spatial unit, where i represents the spatial unit number. This represents the average canopy coverage of the i-th spatial unit within the petal shedding analysis interval, with the subscript avg indicating the average value; the constant 1 represents the upper bound of the normalized range of canopy porosity characterization values; The value 'avg' represents the average canopy porosity of the i-th spatial unit within the petal shedding analysis interval; both canopy coverage and canopy porosity are represented by area ratios, with values ​​ranging from 0 to 1. The canopy bearing capacity results for each spatial unit are associated with the corresponding spatial unit number and petal shedding analysis interval to form the canopy bearing capacity results for that spatial unit. The canopy bearing capacity results for each spatial unit are then summarized to obtain the canopy bearing capacity data.

[0071] Step S25: Infection exposure determination, used to distinguish between a simple decrease in flower quantity and a petal shedding process with exposure conditions for sclerotinia rot infection vectors. Specifically, based on the petal shedding response data and canopy bearing capacity data, the degree of petal shedding response and canopy bearing capacity are correlated. When the petal shedding process and canopy bearing conditions coexist in the same petal shedding analysis interval, the infection exposure level of the corresponding spatial unit is determined based on the degree of petal shedding response and canopy bearing capacity to obtain petal infection exposure data.

[0072] In practical implementation, based on the spatial unit number and the petal shedding analysis interval, the degree of petal shedding response in the petal shedding response data is determined. Canopy bearing capacity in the aforementioned canopy bearing capacity data To correspond; when Greater than 0 and When the value is greater than 0, it is determined that the petal shedding process and the canopy bearing conditions of the corresponding spatial unit coexist within the same petal shedding analysis interval, and the infection exposure level is determined according to the following formula:

[0073] ;

[0074] In the formula, This represents the infection exposure level corresponding to the i-th spatial unit, where i represents the spatial unit number; This indicates the petal shedding response level corresponding to the i-th spatial unit in the petal shedding response data; This represents the canopy bearing capacity corresponding to the i-th spatial unit in the canopy bearing capacity data; a product relationship is used to make the petal shedding response and canopy bearing capacity jointly affect the infection exposure level; when 0 or When it is 0, the corresponding spatial unit will be The value is determined to be 0; for Spatial units greater than 0 are used to determine the corresponding petal shedding analysis interval as the petal infection exposure time interval, and the infection exposure level is then defined. The petal infection exposure time interval and the corresponding spatial unit number are associated to form the petal infection exposure result of the corresponding spatial unit. The petal infection exposure results of each spatial unit are summarized to obtain the petal infection exposure data.

[0075] By performing the above operations, this scheme addresses the issues of significant spectral changes during rapeseed flowering, the potential for confusion between decreased flower quantity and canopy changes caused by diseases, and the difficulty in determining whether conditions for sclerotinia rot infection vector exposure exist solely based on flowering stage status. Instead, it employs a petal infection exposure characterization mechanism based on flowering stage transition, petal shedding response, and canopy carrying capacity. This involves first determining the transition point from full bloom to shedding, then assessing the degree to which fallen petals are received and retained within the canopy based on canopy coverage and porosity, and correlating the petal shedding response with canopy carrying capacity. This transforms ordinary flower quantity changes into a process characterization related to sclerotinia rot infection vector exposure, enabling the distinction between general flower quantity decrease and infection vector exposure. For example, for two spatial units with the same flower quantity decrease but different canopy structures, different petal infection exposure levels can be obtained based on the permeability of fallen petals through the canopy and the differences in canopy carrying surface distribution.

[0076] Example 4, see Figure 1 , Figure 2 and Figure 4 This embodiment is based on the above embodiment. In step S3, the canopy adaptability maintenance assessment is used to determine the ability of rapeseed canopy to maintain the moist conditions related to sclerotinia infection after the canopy has formed a moist state. Specifically, it identifies the changes in thermal response and spectral moisture characterization before and after the formation of the canopy moist state based on the rapeseed remote sensing time series state data, establishes the response trajectory of each spatial unit from the moist state to the reference state before the moist state, and determines the canopy moist condition maintenance ability based on the response recovery time, response deviation decay rate and the consistency of recovery direction, and obtains canopy adaptability maintenance data.

[0077] The assessment of canopy adaptability and maintenance includes the following steps:

[0078] Step S31: Canopy wetting event identification, used to determine the wetting process in which the canopy hydrothermal state changes in tandem. Specifically, based on the rapeseed remote sensing time-series state data, thermal response characterization and spectral moisture characterization are obtained under continuous observation time phases. When the canopy temperature decrease and the spectral moisture increase occur in the same observation time phase or adjacent observation time phases, the corresponding change process is identified as a canopy wetting event, and canopy wetting event data is obtained.

[0079] In practice, the decrease in canopy temperature and the increase in spectral moisture characterization between adjacent observation phases are calculated separately. Preferably, the 75th percentile of the positive change amplitude corresponding to all effective spatial units in the same observation batch is used as the temperature change threshold and the moisture change threshold, respectively. When the decrease in canopy temperature reaches the temperature change threshold and the increase in spectral moisture characterization reaches the moisture change threshold, and the two types of changes occur in the same observation phase or adjacent observation phases, a canopy wetting event is determined to have occurred. Observation intervals that continuously meet the above judgment conditions are classified as the same canopy wetting event. The corresponding spatial unit number, canopy wetting event sequence number, and the start and end phases of the canopy wetting event are recorded to obtain the canopy wetting event data.

[0080] Step S32: Determine the wet reference state, which is used to establish a comparative benchmark for the recovery process after the canopy wets. Specifically, based on the canopy wet event data, determine the thermal response characterization and spectral moisture characterization of multiple consecutive observation phases before the wet event occurs. Based on the response change range of the multiple consecutive observation phases, determine the reference state before wets and the corresponding reference fluctuation range. Then, determine the first observation phase after the wet event response reaches the extreme value as the starting phase of the recovery analysis to obtain the wet reference state data.

[0081] In specific implementation, for the j-th canopy wetting event in the i-th spatial unit, the thermal response characterization and spectral moisture characterization of the three consecutive valid observation phases preceding the onset of the corresponding canopy wetting event are extracted. The arithmetic mean of the three response values ​​is taken as the corresponding pre-wetting reference state, and the range between the minimum and maximum response values ​​is taken as the corresponding reference fluctuation range. Within the time interval corresponding to the canopy wetting event, the observation phases corresponding to the lowest value of the thermal response characterization and the highest value of the spectral moisture characterization are determined. The later observation phase is taken as the extreme value completion phase, and the first valid observation phase thereafter is determined as the starting phase of the recovery analysis. If there are less than three valid observation phases before the onset of the canopy wetting event, subsequent recovery analysis is not performed on the canopy wetting event. The corresponding spatial unit number, canopy wetting event sequence number, the two types of pre-wetting reference states and their reference fluctuation ranges are associated with the starting phase of the recovery analysis to obtain the wetting reference state data.

[0082] Step S33: Canopy recovery trajectory construction, used to characterize the continuous change process of canopy hydrothermal response recovering to the pre-wetting reference state after the formation of the wet state. Specifically, starting from the initial phase of the recovery analysis, the thermal response characterization and spectral moisture characterization of the corresponding spatial unit in the subsequent observation phases are extracted, the degree of response deviation of each observation phase relative to the pre-wetting reference state is calculated, and they are correlated in chronological order to obtain canopy recovery trajectory data.

[0083] In practice, starting from the initial phase of the recovery analysis, thermal response characterization and spectral moisture characterization are extracted sequentially to determine the deviation distance between the current response value and the corresponding reference fluctuation range. When the current response value is within the reference fluctuation range, the deviation distance is 0; when the current response value is outside the reference fluctuation range, the absolute value of the difference between the current response value and the nearest range boundary is taken. The deviation distance is normalized according to the width of the corresponding reference fluctuation range to determine the degree of response deviation, specifically calculated according to the following formula:

[0084] ;

[0085] In the formula, This indicates that the i-th spatial unit and the j-th canopy wetting event are at the k-th observation time. The degree of response deviation corresponding to response type x, where i represents the spatial unit number, j represents the canopy wetting event number, and x represents the response type. When x is T, it represents thermal response characterization; when x is W, it represents spectral moisture characterization. This indicates the k-th observation phase, where k represents the observation phase number. This indicates the corresponding response value at the observation time phase. The deviation distance from the corresponding reference fluctuation range; This indicates the width of the corresponding reference fluctuation range. When the width of the reference fluctuation range is 0, the smallest effective resolution unit of the corresponding remote sensing data is used as the reference range. The minimum effective resolution unit is determined based on the sensor output accuracy that acquires the corresponding remote sensing data; the degree of response deviation corresponding to the two types of responses is correlated according to the observation time phase to form the recovery trajectory of the corresponding canopy wetting event, and the recovery trajectories corresponding to each canopy wetting event are summarized to obtain the canopy recovery trajectory data;

[0086] Step S34: Extraction of invasion-resistant features to distinguish between short-term and continuous wet states. Specifically, based on the canopy recovery trajectory data, the time required for the thermal response characterization and spectral moisture characterization to recover to the reference fluctuation range is determined. Based on the response deviation decay rate during the recovery process and the consistency of the recovery directions of the two types of responses, the continuous wet state characteristics are determined to obtain invasion-resistant feature data.

[0087] In practical implementation, when the deviation of the response corresponding to thermal response characterization or spectral moisture characterization is 0 for two consecutive valid observation phases, the first observation phase is determined as the recovery completion phase of the corresponding response, and the duration from the start phase of the recovery analysis to the recovery completion phase is taken as the corresponding response recovery duration. The later of the two recovery completion phases is taken as the recovery end phase, and the duration from the start phase of the recovery analysis to the recovery end phase is taken as the response recovery duration. If any type of response has not been recovered by the last valid observation phase, the last valid observation phase is taken as the phase at the end of recovery and the unrecovered state is recorded. The reduction in the deviation of the two types of responses corresponding to adjacent valid observation phases during the recovery process is calculated, and the reduction is divided by the duration of the corresponding observation interval to obtain the response deviation reduction rate. The arithmetic mean of the response deviation reduction rates corresponding to the two types of responses is taken to determine the response deviation decay rate. The degree of consistency in the recovery direction is determined by the proportion of observation intervals in which the deviation of both types of responses does not increase within the same observation interval, and one type decreases, to the total number of valid observation intervals in the recovery process. ; Corresponding to all valid canopy wetting events in the same observation batch and The maximum value in the range is normalized, and the eigenvalues ​​that are maintained are determined according to the following formula:

[0088] ;

[0089] In the formula, denoted as the persistence characteristic value corresponding to the j-th canopy wetting event in the i-th spatial unit, where i represents the spatial unit number and j represents the canopy wetting event sequence number; 0.4 represents the weight of the normalized response recovery time in the persistence characteristic value, which is used to ensure that the length of time the canopy hydrothermal state deviates from the pre-wetting reference state plays a major role in the evaluation. This represents the normalized response recovery time corresponding to the j-th canopy wetting event in the i-th spatial unit. The symbol ~ indicates that the parameter has been normalized to 0 to 1. The subscript rec indicates that the parameter originates from the recovery process experienced by the thermal response characterization and spectral moisture characterization to the corresponding reference fluctuation range.

[0090] The first 0.3 represents the weight of the normalized response deviation decay rate in the sustained eigenvalues; the constant 1 represents the upper bound of the normalized response deviation decay rate. The normalized response deviation decay rate corresponding to the i-th spatial unit and the j-th canopy wetting event is used to characterize the rate at which the thermal response characterization and spectral moisture characterization decrease in response deviation from the pre-wetting reference state during the recovery process. The symbol ~ indicates that the parameter has been normalized to 0 to 1, and the subscript dec is used to indicate the decay process of the parameter reflecting the degree of response deviation.

[0091] Since a lower response deviation decay rate indicates a slower decline in canopy wetness, therefore... The response deviation decay rate is converted into a parameter that changes in the same direction as the degree of wetness retention;

[0092] The second 0.3 represents the weight of the degree of consistency in the recovery direction in the sustained eigenvalues; This represents the degree of consistency in the recovery direction corresponding to the j-th canopy wetting event in the i-th spatial unit. It is used to characterize whether the thermal response characterization and spectral moisture characterization continuously change towards the corresponding pre-wetting reference state during the recovery process. The subscript "con" indicates that this parameter reflects the consistency relationship between the recovery directions of the two types of responses. The sum of the three weights is 1, which results in a longer response recovery time, a lower response deviation decay rate, and a higher degree of consistency in the recovery direction. The larger; will The response recovery time, the phase at the end of recovery and the unrecovered state are associated with the corresponding spatial unit number and the canopy wetting event number to obtain the invasion-resistance characteristic data.

[0093] Step S35: Determine the ability to maintain the canopy moisture condition related to sclerotinia rot infection. Specifically, determine the moisture maintenance level of each spatial unit based on the aforementioned ability to maintain the canopy moisture condition. Match the moisture maintenance time interval with the petal infection exposure time interval determined in step S2. When the two time intervals overlap, determine the degree of canopy moisture maintenance of the corresponding spatial unit based on the overlap duration and the moisture maintenance level to obtain canopy moisture maintenance data.

[0094] In specific implementation, the persistent retention feature value in the aforementioned intrusion-resistant feature data will be... The wettability retention level for the corresponding canopy wettability event is determined, and the time interval from the start phase to the end phase of recovery of the corresponding canopy wettability event is defined as the wettability retention time interval. Based on the spatial unit number, the wettability retention time interval is matched with the petal infection exposure time interval in the petal infection exposure data obtained in step S2. When the two time intervals overlap, the corresponding overlap duration is calculated. When multiple canopy wettability events satisfying the time interval overlap condition exist in the same spatial unit, the appropriate infection retention level of the corresponding spatial unit is determined according to the following formula:

[0095] ;

[0096] In the formula, This indicates the degree of intrusion preservation corresponding to the i-th spatial unit, where i represents the spatial unit number; This indicates that the maximum value is taken among the canopy wetting events that satisfy the time interval overlap condition within the i-th spatial unit, and j represents the canopy wetting event number; This represents the moisture retention level corresponding to the j-th canopy wetting event in the i-th spatial unit; This indicates the duration of overlap between the time interval corresponding to the wetness retention of the i-th spatial unit and the time interval of petal infection exposure for the j-th canopy wetness event, with the subscript ov indicating time overlap; This represents the duration of the petal infection exposure time interval corresponding to the i-th spatial unit obtained in step S2, with the subscript exp indicating infection exposure; when there is no canopy wetting event satisfying the time interval overlap condition, Set to 0; The corresponding time intervals and spatial unit numbers of the corresponding moisture retention are associated to form the canopy invasion retention results of the corresponding spatial units. The canopy invasion retention results of each spatial unit are then summarized to obtain the canopy invasion retention data.

[0097] By performing the above operations, this scheme addresses the problem that traditional methods using rainfall, air humidity, or single-phase canopy moisture status are insufficient to reflect the actual duration of canopy wetness in rapeseed. Instead, it employs an assessment mechanism based on the canopy's hydrothermal recovery trajectory after a wet event to maintain suitability for invasion. This involves establishing a pre-wetting reference state and reference fluctuation range based on multiple observation phases prior to the wet event, continuously tracking the recovery process of thermal response characterization and spectral moisture characterization towards the reference fluctuation range, and determining the canopy's ability to maintain wet conditions based on recovery time, response deviation decay rate, and the degree of consistency between the two responses. Furthermore, this is matched with the petal infection exposure time interval to distinguish between short-term and continuous wet states. For example, when two spatial units have similar initial wet responses, different levels of suitability for invasion can be determined based on the time required for recovery to the pre-wetting reference fluctuation range and the degree of overlap with the petal infection exposure time interval.

[0098] Example 5, see Figure 1 , Figure 2 This embodiment is based on the above embodiment. In step S4, the disease abnormal response identification is used to identify abnormal remote sensing responses of different spatial units relative to the normal growth changes of rapeseed. Specifically, it selects reference spatial units whose phenological stage differences are within a set matching range based on the rapeseed remote sensing time series state data. It establishes a reference response interval for the corresponding phenological stage based on the spectral moisture characterization, canopy structure characterization, and thermal response characterization of the reference spatial units. It compares the remote sensing response of the target spatial unit with the reference response interval. When the remote sensing response of the target spatial unit exceeds the corresponding reference response interval for two consecutive observation phases, the corresponding response is determined as a candidate abnormal response of disease. Based on the petal infection exposure data and canopy adaptability maintenance data, it filters candidate abnormal responses of disease that are time-related to the canopy moistening maintenance process after the infection exposure occurs, and obtains disease abnormal response data.

[0099] In practical implementation, phenological stage matching is performed based on the flowering period characteristics in the rapeseed remote sensing time-series data. The absolute value of the difference in flowering period characteristics between the target spatial unit and the candidate reference spatial units is set as no more than 10% of the range of flowering period characteristics in the same observation phase. Spatial units meeting this condition constitute a reference spatial unit set. The 25th and 75th percentiles are determined based on the spectral response, canopy state, and thermal response of the reference spatial units. The lower bound of the reference response interval is the 25th percentile minus 1.5 interquartile range, and the upper bound is the 75th percentile plus 1.5 interquartile range. The degree of anomaly is calculated based on the extent to which each response of the target spatial unit exceeds the corresponding reference response interval.

[0100] ;

[0101] In the formula, This indicates that the i-th spatial unit is in the k-th observation phase. The corresponding anomaly level, where i represents the spatial unit number, This indicates the kth observation phase, where k represents the observation phase number; 1 / 3 represents the average coefficient of the normalized superboundary distance corresponding to the three responses. This represents the distance the spectral response exceeds the corresponding reference response range, where S represents the spectral response. This represents the width of the reference response interval for spectral moisture characterization established for the i-th spatial unit; This indicates the distance the canopy state exceeds the corresponding reference response interval, where C represents the canopy state; This indicates the width of the reference response interval corresponding to the canopy state; This represents the distance the thermal response exceeds the corresponding reference response range, where T represents the thermal response. This indicates the width of the reference response interval corresponding to the thermal response;

[0102] When the corresponding response is within the reference response interval, its over-limit distance is 0. When the width of the reference response interval is 0, the smallest effective resolution unit of the corresponding response is used instead.

[0103] When two or three of the three responses exceed the corresponding reference response interval for two consecutive observation phases, candidate abnormal responses for the disease are identified. Based on the spatial unit number, the candidate abnormal responses for the disease are correlated with the petal infection exposure data and the canopy suitability maintenance data. Candidate abnormal responses for the disease that occur after the start phase of the petal infection exposure time interval and are located within the corresponding wet maintenance time interval or within the first valid observation interval after its end are retained. The corresponding abnormality degree, abnormal observation phase, and spatial unit number are associated to obtain the abnormal response data for the disease.

[0104] Example 6, see Figure 1 and Figure 2 This embodiment is based on the above embodiment. In step S5, the sclerotinia disease risk warning is used to comprehensively determine the sclerotinia disease occurrence risk of different spatial units in the rapeseed planting area. Specifically, it determines the infection exposure index, the canopy suitability maintenance index, and the disease abnormality index based on the petal infection exposure data, the canopy suitability maintenance data, and the disease abnormality response data, respectively. After normalizing each index, they are combined according to the set weights to obtain the sclerotinia disease risk assessment value. Based on the comparison result between the sclerotinia disease risk assessment value and the risk level threshold, the risk level of each spatial unit is determined, the spatial distribution information of sclerotinia disease risk is generated, and the rapeseed sclerotinia disease risk warning result is obtained.

[0105] In specific implementation, the infection exposure level in the petal infection exposure data is determined as the infection exposure index, the degree of canopy adaptability maintenance in the canopy adaptability maintenance data is determined as the adaptability maintenance index, and the maximum degree of anomaly in the corresponding spatial unit in the disease anomaly response data is determined as the disease anomaly index. When there is no disease anomaly response in the corresponding spatial unit, the disease anomaly index is determined to be 0. The minimum and maximum values ​​of each index in the same observation batch are used for normalization. When the maximum and minimum values ​​of the same index are the same, the corresponding normalization result is determined to be 0.

[0106] Calculate the sclerotinia disease risk assessment value using the following formula:

[0107] ;

[0108] In the formula, This represents the sclerotinia disease risk assessment value corresponding to the i-th spatial unit, where i represents the spatial unit number; 0.35 represents the weight corresponding to the normalized infection exposure index. This represents the normalized infection exposure index corresponding to the i-th spatial unit, with the symbol ~ indicating the normalization result and E representing the infection exposure level; the second 0.35 represents the weight corresponding to the normalized fitness retention index. This represents the normalized invasiveness retention index corresponding to the i-th spatial unit, where I represents the degree of invasiveness retention; 0.30 represents the weight corresponding to the normalized disease anomaly index. Let A represent the normalized disease anomaly index corresponding to the i-th spatial unit, and let A represent the degree of anomaly corresponding to the disease anomaly response; the sum of the three weights is 1.

[0109] Will Spatial cells smaller than 0.35 are classified as low-risk. Spatial cells with a value not less than 0.35 and less than 0.65 are classified as medium-risk. Spatial units with a value of not less than 0.65 are identified as high-risk. The risk assessment value, risk level and corresponding spatial location of the sclerotinia disease risk are correlated to generate spatial distribution information of sclerotinia disease risk, and the sclerotinia disease risk warning result of rapeseed is obtained.

[0110] Example 7, see Figure 1 and Figure 2 Based on the above embodiments, the rapeseed sclerotinia stem rot risk early warning system provided by the present invention includes a remote sensing status construction module, a petal infection exposure module, a canopy adaptability maintenance module, a disease anomaly identification module, and a risk early warning module.

[0111] The remote sensing status construction module is used to form a basic remote sensing characterization of the rapeseed planting area under different observation time phases. Through the rapeseed remote sensing status construction, rapeseed remote sensing time series status data is obtained, and the rapeseed remote sensing time series status data is sent to the petal infection exposure module.

[0112] The petal infection and exposure module is used to identify the petal shedding process related to the exposure of the sclerotinia rot infection vector. Through petal infection and exposure characterization, petal infection and exposure data are obtained, and the rapeseed remote sensing time-series status data and petal infection and exposure data are sent to the canopy adaptability maintenance module.

[0113] The canopy adaptability maintenance module is used to determine the ability of the canopy to maintain the moist conditions related to sclerotinia infection after the canopy has formed a moist state. Through the canopy adaptability maintenance assessment, canopy adaptability maintenance data is obtained, and the rapeseed remote sensing time-series status data, petal infection exposure data and canopy adaptability maintenance data are sent to the disease anomaly identification module.

[0114] The disease anomaly identification module is used to identify abnormal remote sensing responses that are relative to the normal growth changes of rapeseed and are time-related to the infection process. Through disease anomaly response identification, disease anomaly response data is obtained, and the petal infection exposure data, canopy adaptability maintenance data and disease anomaly response data are sent to the risk warning module.

[0115] The risk warning module is used to comprehensively determine the risk of sclerotinia disease occurrence in different spatial units, obtain the risk warning result of rapeseed sclerotinia disease through sclerotinia disease risk warning, and output the risk warning result of rapeseed sclerotinia disease.

[0116] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0117] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention.

[0118] The present invention and its embodiments have been described above. This description is not restrictive, and the accompanying drawings are only one embodiment of the present invention; the actual structure is not limited thereto. In conclusion, if those skilled in the art are inspired by this description and design similar structures and embodiments without departing from the spirit of the invention, such designs should fall within the protection scope of the present invention.

Claims

1. A method for early warning of rapeseed sclerotinia stem rot risk based on remote sensing characterization data, characterized in that: The method includes the following steps: Step S1: Constructing the remote sensing status of rapeseed to obtain the remote sensing time series status data of rapeseed; Step S2: Petal infection exposure characterization. Based on the rapeseed remote sensing time-series data, extract the continuous change characteristics of flower quantity in each spatial unit, determine the flowering period transition position where the flower quantity changes from increasing to decreasing, perform time-series analysis on the petal reduction process after the flowering period transition, and combine the canopy's spatial carrying capacity for fallen petals to determine the degree of exposure conditions for fallen petals to form infection carriers within the canopy, thus obtaining petal infection exposure data. Step S3: Canopy invasion resistance maintenance assessment. Based on the rapeseed remote sensing time-series state data, identify the changes in thermal response and spectral moisture characterization before and after the formation of canopy wet state, establish the response trajectory of each spatial unit from wet state to pre-wet reference state, and determine the canopy wet condition maintenance capability based on the response recovery time, response deviation decay rate and the degree of consistency of recovery direction, and obtain canopy invasion resistance maintenance data. Step S4: Identify abnormal disease responses and obtain abnormal disease response data; Step S5: Sclerotinia stem rot risk warning. Generate spatial distribution information of sclerotinia stem rot risk and obtain sclerotinia stem rot risk warning results for rapeseed.

2. The method for early warning of rapeseed sclerotinia stem rot risk based on remote sensing characterization data according to claim 1, characterized in that: In step S1, the rapeseed remote sensing status construction involves acquiring multi-temporal remote sensing images of the target rapeseed planting area, performing radiometric correction, geometric correction, spatial registration, and rapeseed planting area extraction on the multi-temporal remote sensing images, and dividing the rapeseed planting area into spatial units based on a set spatial scale. The flowering period characterization, canopy structure characterization, spectral moisture characterization, and thermal response characterization corresponding to each spatial unit are extracted and correlated according to the observation time phase to obtain rapeseed remote sensing time-series status data.

3. The method for early warning of rapeseed sclerotinia stem rot risk based on remote sensing characterization data according to claim 2, characterized in that: In step S2, the petal infection exposure characterization includes the following steps: construction of flower quantity time-series response, identification of flowering period transition location, extraction of petal abscission response, assessment of canopy bearing capacity, and determination of infection exposure; The construction of the flower quantity time-series response involves extracting flowering period features corresponding to the spectral response of rapeseed petals based on the rapeseed remote sensing time-series state data, determining the petal coverage degree within a spatial unit based on the corresponding flowering period features at each observation time, and arranging the petal coverage degree under continuous observation time phases in chronological order to obtain the flower quantity time-series response data.

4. The method for early warning of rapeseed sclerotinia stem rot risk based on remote sensing characterization data according to claim 3, characterized in that: In step S2, the flower period transition location identification involves calculating the direction of flower quantity change between adjacent observation time phases based on the flower quantity time-series response data, identifying candidate time phases where the flower quantity changes from an increasing state to a decreasing state, and determining the corresponding candidate time phase as the flower period transition location when the petal coverage of two consecutive observation intervals after the candidate time phase both show a decreasing change, thereby obtaining the flower period transition data. The petal shedding response extraction is performed by determining the petal shedding analysis interval corresponding to each spatial unit based on the flowering period conversion data, determining the petal reduction amount based on the difference in petal coverage between adjacent observation phases within the analysis interval, and determining the petal shedding response degree based on the petal reduction amount and its rate of change within the continuous observation interval, thereby obtaining petal shedding response data. The assessment of canopy carrying capacity is based on the extraction of canopy coverage and canopy porosity characteristics within the petal shedding analysis interval from the rapeseed remote sensing time-series data. The canopy surface distribution that can be used to receive fallen petals is determined based on the canopy coverage, and the spatial permeability of fallen petals passing through the canopy is determined based on the canopy porosity characteristics. The relative degree of receiving and retention of fallen petals inside the canopy is determined by combining the two, thus obtaining canopy carrying capacity data. The infection exposure is determined by correlating the petal shedding response data and canopy bearing capacity data with the canopy bearing capacity. When the petal shedding process and canopy bearing conditions coexist within the same petal shedding analysis interval, the infection exposure level of the corresponding spatial unit is determined based on the petal shedding response level and canopy bearing capacity, thus obtaining petal infection exposure data.

5. The method for early warning of rapeseed sclerotinia stem rot risk based on remote sensing characterization data according to claim 4, characterized in that: In step S3, the canopy adaptation and maintenance assessment includes the following steps: canopy wetting event identification, wetting reference state determination, canopy recovery trajectory construction, adaptation and maintenance feature extraction, and adaptation and maintenance capability determination; The canopy wetting event identification is based on the rapeseed remote sensing time-series state data to obtain thermal response characterization and spectral moisture characterization under continuous observation time phases. When the canopy temperature decreases and the spectral moisture increases occur in the same observation time phase or adjacent observation time phases, the corresponding change process is determined as a canopy wetting event, and canopy wetting event data is obtained. The wet reference state is determined by determining the thermal response characterization and spectral moisture characterization of multiple consecutive observation phases before the occurrence of the wet event based on the canopy wet event data. The reference state before wetting and the corresponding reference fluctuation range are determined based on the response change range of the multiple consecutive observation phases. The first observation phase after the response of the wet event reaches the extreme value is determined as the starting phase of the recovery analysis, and the wet reference state data is obtained.

6. The method for early warning of rapeseed sclerotinia stem rot risk based on remote sensing characterization data according to claim 5, characterized in that: In step S3, the canopy recovery trajectory is constructed by taking the initial phase of the recovery analysis as the starting point, extracting the thermal response characterization and spectral moisture characterization of the corresponding spatial units in the subsequent observation phases, calculating the degree of response deviation of each observation phase relative to the pre-wetting reference state, and correlating them in chronological order to obtain the canopy recovery trajectory data. The invasion-resistance retention feature extraction, based on the canopy recovery trajectory data, determines the time required for the thermal response characterization and spectral moisture characterization to recover to the reference fluctuation range, and determines the continuous retention feature of the wet state according to the response deviation decay rate and the consistency of the recovery direction of the two types of responses during the recovery process, thereby obtaining invasion-resistance retention feature data. The ability to maintain the canopy's resistance to invasion is determined by determining the level of moisture retention for each spatial unit based on the resistance to invasion characteristic data. The time interval corresponding to moisture retention is matched with the petal infection exposure time interval determined in step S2. When the two time intervals overlap, the degree of resistance to invasion for the corresponding spatial unit is determined based on the duration of the overlap and the level of moisture retention, thus obtaining the canopy resistance to invasion data.

7. The method for early warning of rapeseed sclerotinia stem rot risk based on remote sensing characterization data according to claim 6, characterized in that: In step S4, the disease abnormal response identification involves selecting reference spatial units whose phenological stage differences fall within a set matching range based on the rapeseed remote sensing time-series state data. A reference response interval for the corresponding phenological stage is established based on the spectral moisture characterization, canopy structure characterization, and thermal response characterization of the reference spatial units. The remote sensing response of the target spatial unit is compared with the reference response interval. When the remote sensing response of the target spatial unit exceeds the corresponding reference response interval for two consecutive observation phases, the corresponding response is identified as a candidate disease abnormal response. Furthermore, based on the petal infection exposure data and canopy adaptability maintenance data, candidate disease abnormal responses that occur after infection exposure and are time-related to the canopy moisture maintenance process are screened to obtain disease abnormal response data. In step S5, the sclerotinia rot risk warning is used to comprehensively determine the sclerotinia rot occurrence risk in different spatial units within the rapeseed planting area. Specifically, it involves determining infection exposure indicators, canopy suitability maintenance indicators, and disease abnormality indicators based on the petal infection exposure data, canopy suitability maintenance data, and disease abnormality response data, respectively. After normalizing each indicator, they are combined according to a set weight to obtain a sclerotinia rot risk assessment value. The risk level of each spatial unit is determined based on the comparison between the sclerotinia rot risk assessment value and the risk level threshold, generating sclerotinia rot risk spatial distribution information, and obtaining the rapeseed sclerotinia rot risk warning result.

8. A rapeseed sclerotinia stem rot risk early warning system based on remote sensing characterization data, used to implement the rapeseed sclerotinia stem rot risk early warning method based on remote sensing characterization data as described in any one of claims 1-7, characterized in that: It includes a remote sensing status construction module, a petal infection and exposure module, a canopy adaptability maintenance module, a disease anomaly identification module, and a risk warning module.

9. The rapeseed sclerotinia stem rot risk early warning system based on remote sensing characterization data according to claim 8, characterized in that: The remote sensing status construction module is used to form a basic remote sensing characterization of the rapeseed planting area under different observation time phases. Through the rapeseed remote sensing status construction, rapeseed remote sensing time series status data is obtained, and the rapeseed remote sensing time series status data is sent to the petal infection exposure module. The petal infection and exposure module is used to identify the petal shedding process related to the exposure of the sclerotinia rot infection vector. Through petal infection and exposure characterization, petal infection and exposure data are obtained, and the rapeseed remote sensing time-series status data and petal infection and exposure data are sent to the canopy adaptability maintenance module. The canopy adaptability maintenance module is used to determine the ability of the canopy to maintain the moist conditions related to sclerotinia infection after the canopy has formed a moist state. Through the canopy adaptability maintenance assessment, canopy adaptability maintenance data is obtained, and the rapeseed remote sensing time-series status data, petal infection exposure data and canopy adaptability maintenance data are sent to the disease anomaly identification module. The disease anomaly identification module is used to identify abnormal remote sensing responses that are relative to the normal growth changes of rapeseed and are time-related to the infection process. Through disease anomaly response identification, disease anomaly response data is obtained, and the petal infection exposure data, canopy adaptability maintenance data and disease anomaly response data are sent to the risk warning module. The risk warning module is used to comprehensively determine the risk of sclerotinia disease occurrence in different spatial units, obtain the risk warning result of rapeseed sclerotinia disease through sclerotinia disease risk warning, and output the risk warning result of rapeseed sclerotinia disease.

Images