A method and system for analyzing the relationship between crop growth environment and pest induction

Abstract

The present application relates to the technical field of agricultural production management, in particular to a crop growth environment and pest-induced relationship analysis method and system, comprising the following steps: collecting air temperature, air humidity, precipitation, light, soil moisture and pest records and collecting time uniformly, aligning environment and pest observation information, analyzing plant height, leaf and accumulated temperature changes to divide growth stages, aligning environment change segments before and after the stages, classifying growth stage environment change and pest occurrence sequence, and forming an environment-induced pest trigger chain record set. In the present application, the growth stage section is divided by the crop growth state change process, the growth stage switching and pest change are closely aligned in time sequence, the occurrence of pests is placed in a clear growth stage and environment change background for analysis, and the environment condition change is no longer presented in an isolated interval, thereby providing a clear and coherent analysis basis for the analysis of the relationship between crop growth environment and pest-induced relationship in agricultural production management.

Description

Technical Field

[0001] This invention relates to the field of agricultural production management technology, and in particular to a method and system for analyzing the relationship between crop growth environment and pest induction. Background Technology

[0002] The field of agricultural production management technology mainly involves management and analysis techniques throughout the entire agricultural production process. This encompasses core aspects such as collecting data on natural environmental conditions during crop cultivation, recording production factors, identifying agricultural risk factors, and analyzing production decisions based on objective data. It systematically organizes information on temperature, precipitation, light, soil physicochemical properties, and biological growth status to support the understanding and judgment of crop growth status and potential risks in agricultural production activities. Among these, the traditional method for analyzing the relationship between crop growth environment and pest induction refers to a technical solution that analyzes the correlation between environmental conditions during crop growth and pest occurrence in agricultural production management. This involves recording changes in temperature, air humidity, precipitation, soil moisture content, and the time and location of pest occurrence during the crop growth period. It compares and analyzes the frequency and types of pest occurrence under different environmental conditions, and summarizes and describes the correspondence between crop growth environment and pest induction based on long-term accumulated agricultural production records or human experience rules, for use in pest-related analysis in agricultural production management.

[0003] Traditional analyses of the relationship between crop growth environment and pest induction rely on scattered records and manual experience. The lack of unified constraints on the collection time of environmental data and pest records leads to temporal discrepancies between information from different sources, making it difficult to establish a clear correlation between environmental changes and pest occurrence. Growth stages are defined based on calendar intervals or manual judgment, resulting in a disconnect between stage boundaries and actual growth conditions. This leads to pest occurrences being roughly categorized into broad stages. Environmental conditions are often compiled using a holistic interval comparison approach, ignoring the continuous changes before and after stage transitions. These operational methods leave the analysis results at the level of empirical description, failing to reflect the temporal relationship between environmental changes and pest induction, thus hindering in-depth analysis of pest-related issues in agricultural production management. Summary of the Invention

[0004] The purpose of this invention is to overcome the shortcomings of existing technologies and to propose a method and system for analyzing the relationship between crop growth environment and pest induction.

[0005] To achieve the above objectives, the present invention adopts the following technical solution: a method for analyzing the relationship between crop growth environment and pest induction, comprising the following steps:

[0006] S1: Collect data on the collection time and status of air temperature, air humidity, precipitation, sunshine duration, soil moisture, and insect trapping in the crop planting area, as well as crop growth monitoring data; identify the communication response sequence of the collection equipment; determine the offset of the time records; and generate a unified time environment insect observation set.

[0007] S2: Based on the crop growth monitoring data and air temperature records in the unified time environment insect observation set, extract the crop phenotypic growth characteristic records and accumulated temperature records, compare the direction of change of crop growth status at adjacent observation times, and obtain the set of growth stage change segments.

[0008] S3: By comparing the changing trends of insect records near the boundary of the growth segment through the set of growth stage change segments, the close proximity of insect changes and growth stage switching in time sequence is determined, and time locations that simultaneously have the characteristics of growth state change and insect change are selected to generate a set of growth pest association anchor points.

[0009] S4: Analyze the air temperature, air humidity, precipitation, light duration and soil moisture records before and after the growth pest association anchor point set, compare the direction of change of environmental elements in the time series before and after the growth pest association anchor point, determine the continuous segments in which the environmental state has changed, and obtain the pest-induced environmental change segment set.

[0010] As a further aspect of the present invention, the unified time-environment insect pest observation set includes a time consistency identifier, an environmental element synchronous recording identifier, an insect pest observation alignment identifier, and an available observation integrity identifier; the growth stage change segment set includes a growth state transition segment, a growth characteristic synchronous change segment, a growth continuity segment identifier, and a growth stage boundary identifier; the growth pest association anchor point set includes a growth stage switching identifier, an insect pest change correspondence identifier, a growth pest time adjacency identifier, and a growth stage correspondence identifier; and the pest-induced environmental change segment set includes an environmental state transition segment, an environmental change direction segment, a pest occurrence association segment, and an environmental pest time alignment segment.

[0011] As a further aspect of the present invention, the steps for obtaining the unified time-environment insect infestation observation set are specifically as follows:

[0012] S111: Collect the collection time status of air temperature, air humidity, precipitation, sunshine duration, soil moisture and insect trapping records in the crop planting area, collect the timestamps and collection status codes of the records, filter the valid status codes, and generate a collection time index table.

[0013] S112: Based on the acquisition time index table, acquire the gateway timestamp sequence, the acquisition device timestamp sequence, and the gateway receiving time sequence, and acquire the communication sequence index and the ratio of the arrival interval of adjacent communication, compare the corresponding communication response order, calculate the synchronization offset, perform interval mapping based on the synchronization offset and the zero offset reference, and mark the offset direction to obtain the time synchronization discrimination result;

[0014] S113: Call the time synchronization discrimination result, adjust the time stamp of the acquisition device and write the correction timestamp sequence, and record and aggregate the air temperature, air humidity, precipitation, light duration, soil moisture and insect trapping based on the correction timestamp to generate a unified time environment insect observation set.

[0015] As a further aspect of the present invention, the step of obtaining the set of growth stage change segments specifically includes:

[0016] S211: Based on the unified time environment insect observation set, extract crop plant height change records, leaf area change records, and accumulated temperature change records, align the three types of records according to the observation time, and determine the positive or negative sign of the numerical difference results of adjacent observation times to generate a growth index change direction sequence set.

[0017] S212: Based on the growth index change direction sequence set, the consistency of the plant height change direction, leaf area change direction and accumulated temperature change direction at the same observation time is judged. When the three types of directions change from zero to non-zero and remain consistent in continuous observation time, the corresponding time index is recorded and the segments are connected to obtain the continuous segment information of growth state transition.

[0018] S213: Based on the continuous segment information of the growth state transition, retrieve the start and end time positions of the segment, take the starting point of the segment as the growth state boundary benchmark, calculate the growth stage boundary feature value, and perform front and back split labeling according to the complete observation time axis to generate a set of growth stage change segments.

[0019] As a further aspect of the present invention, the step of obtaining the growth pest association anchor point set specifically includes:

[0020] S311: Based on the set of growth stage change segments, monitor the insect population record time series near the boundary of the growth segment, call the insect population quantity value and time index in adjacent time windows, use difference operation to obtain the change sequence of continuous time moments, and make a consistency judgment on the sign direction of the change to generate an insect population change trend sequence.

[0021] S312: Based on the insect infestation change trend sequence, call the corresponding growth segment boundary time index, perform time interval calculation for the time of insect infestation change and the time of growth stage switching, make logical judgment based on the preset adjacent judgment benchmark value, filter positions that meet the time sequence adjacent conditions, and obtain the time sequence adjacent position set.

[0022] S313: Call the set of adjacent time-series locations, retrieve the corresponding previous and subsequent growth stage identifier codes and pest change marker values, perform identifier matching and state consistency judgment, retain the locations that simultaneously meet the conditions of growth state change and pest change, and uniformly number and label them to generate a set of growth pest association anchor points.

[0023] As a further aspect of the present invention, the step of obtaining the set of environmental change fragments induced by pests specifically includes:

[0024] S411: Based on the aforementioned set of growth pest association anchor points, collect records of air temperature, air humidity, precipitation, sunshine duration and soil moisture content within the corresponding time windows before and after, perform time-series difference operation on the values ​​of the same environmental element at adjacent times, and determine the consistency of the difference results to generate a sequence of environmental element change directions.

[0025] S412: Based on the sequence of environmental element change directions, perform a merging operation on the consistent direction segments of the same environmental element on the continuous time index, determine the position boundary where the environmental state changes on the time axis, record the start and end time indices of continuous changes, and obtain a set of continuous segments of environmental state transition.

[0026] S413: Call the set of continuous environmental state transition segments, retrieve the pest occurrence record index within the corresponding time range, perform alignment comparison between the start and end time of the environmental change segment and the pest record time, retain segments with established time overlap, and generate a set of pest-induced environmental change segments.

[0027] As a further aspect of the present invention, the method further includes step S5:

[0028] S5: Using the aforementioned set of environmental change fragments induced by pests, compare the co-occurrence characteristics of environmental change fragments and the order of pest appearance under different growth stages, determine the distribution pattern of the direction of environmental change before and after the appearance of pests, classify and organize them according to crop growth stage and pest type, and establish a record set of environmentally induced pest triggering chains.

[0029] The environmentally induced pest triggering chain record set includes environmental association chains for growth stages, pest occurrence sequence association chains, environmental change direction association chains, and corresponding chain records for pests at different growth stages.

[0030] As a further aspect of the present invention, the step of obtaining the environmentally induced pest triggering chain record set specifically includes:

[0031] S511: Based on the set of environmental change fragments induced by pests, call the crop growth stage identifier and pest occurrence time index corresponding to the environmental change fragments, perform sequential comparison of the start time of environmental change and the occurrence time of pests under the same growth stage, record the temporal relationship of environmental change preceding or following the occurrence of pests, and generate a temporal co-occurrence feature sequence of environmental pests.

[0032] S512: Based on the environmental pest co-occurrence feature sequence, perform statistical merging operation on the distribution state of the environmental element change direction before and after the pest occurrence, perform frequency aggregation on the same direction relationship, and complete the state mapping on the growth stage dimension to obtain the environmental change direction distribution pattern set.

[0033] S513: Call the set of distribution patterns of environmental change direction, retrieve the corresponding pest type identifier and crop growth stage code, classify and organize the environmental change direction and the order of pest appearance, aggregate and number the records corresponding to the same stage and the same pest, and establish a record set of environmentally induced pest triggering chain.

[0034] The crop growth environment and pest induction relationship analysis system is used to execute the above-mentioned crop growth environment and pest induction relationship analysis method. The system includes:

[0035] The time consistency correction module collects the collection timestamps of air temperature, air humidity, precipitation, light duration, soil moisture content and insect trapping in the crop planting area, compares the communication response sequence between the field gateway and the collection equipment, estimates and corrects the time offset, and forms a unified time environment insect observation set.

[0036] The growth state segmentation module, based on the unified time environment insect observation set, obtains the sequence of crop plant height change, leaf area change and accumulated temperature change, compares the change direction of adjacent observation times, determines the continuous segment of growth state transitioning from stable to changing, and generates a set of growth stage change segments.

[0037] The insect-related anchoring module uses the set of growth stage change segments to compare the changing trend of the number of insects trapped before and after the segment boundary, calculates and judges the temporal proximity relationship between insect changes and growth stage switching, filters out the locations where growth state changes and insect fluctuations occur simultaneously, and generates a set of growth pest-related anchor points.

[0038] The environmental transfer identification module, based on the set of growth pest-related anchor points, obtains the corresponding sequence of air temperature, air humidity, precipitation, light duration and soil moisture content before and after, compares and judges the continuous segments of the direction of change of environmental elements, aligns the environmental change segments with the time of pest occurrence, and generates a set of environmental change segments induced by pests.

[0039] The triggering chain construction module utilizes the set of environmental change fragments induced by pests to compare the co-occurrence distribution of the direction of environmental change and the order of pest occurrence under differentiated growth stages, completes the discrimination and classification of environmental change patterns and pest types, and constructs a record set of environmentally induced pest triggering chains.

[0040] Compared with the prior art, the advantages and positive effects of the present invention are as follows:

[0041] In this invention, by organizing environmental and pest observation records around a unified time marker, multi-source information is organized on the same timeline. The growth stage is divided into segments based on the changes in crop growth status, and the transitions between growth stages and changes in pests are closely aligned in chronological order. Furthermore, the continuous change direction of environmental elements is traced around the stage boundaries. The occurrence of pests is analyzed within a clear growth stage and environmental change context. This prevents changes in environmental conditions from being presented as isolated intervals, but rather as stage-related fragments combined with the pest timeline, forming a traceable trigger chain description. This provides a clear and coherent analytical foundation for analyzing the relationship between crop growth environment and pest induction in agricultural production management. Attached Figure Description

[0042] Figure 1 This is a schematic diagram of the workflow of the present invention;

[0043] Figure 2 This is a flowchart illustrating the process of obtaining a unified time-environment insect observation set in this invention.

[0044] Figure 3 This is a flowchart illustrating the process of obtaining the set of growth stage change segments in this invention.

[0045] Figure 4 This is a flowchart illustrating the process of obtaining the set of growth pest-related anchor points in this invention.

[0046] Figure 5 This is a flowchart illustrating the process of obtaining the set of environmental change fragments induced by pests in this invention.

[0047] Figure 6 This is a flowchart illustrating the process of obtaining the record set of environmentally induced pest infestation triggering chains in this invention. Detailed Implementation

[0048] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0049] In the description of this invention, it should be understood that the terms "length," "width," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," and "outer," etc., indicating orientation or positional relationships, are based on the orientation or positional relationships shown in the accompanying drawings and are only for the convenience of describing the 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, and therefore should not be construed as a limitation of the invention. Furthermore, in the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0050] Please see Figure 1 This invention provides a technical solution, a method for analyzing the relationship between crop growth environment and pest induction, comprising the following steps:

[0051] S1: Collect data on the collection time status and crop growth monitoring of air temperature, air humidity, precipitation, sunshine duration, soil moisture content, and insect trapping in the crop planting area; identify the communication response sequence of the collection equipment; determine the offset of the time records; adjust the time stamp of the collection equipment to be consistent with the gateway time; synchronize and organize the available observation records; and generate a unified time environment insect observation set.

[0052] S2: Based on the crop growth monitoring data and air temperature records from the unified time environmental insect observation set, extract the crop phenotypic growth characteristic records and accumulated temperature records, compare the direction of change of crop growth status at adjacent observation times, determine the continuous segments where the growth status shifts from stable to changing, screen the time positions where growth characteristic changes occur simultaneously, and divide the crop growth before and after segments to obtain the set of growth stage change segments.

[0053] S3: By comparing the changing trends of insect records near the boundaries of growth segments through the set of growth stage change segments, determine the close proximity of insect changes and growth stage transitions in time sequence, filter time locations that simultaneously possess the characteristics of growth state transition and insect change, and mark the corresponding preceding and following growth stage identifiers to generate a set of growth pest association anchor points.

[0054] S4: Analyze the air temperature, air humidity, precipitation, light duration and soil moisture records before and after the growth pest association anchor point set, compare the direction of change of environmental elements in the time series before and after the growth pest association anchor point, determine the continuous segments of environmental state transfer, and align the environmental change segments with the corresponding pest occurrence records in time to obtain the pest-induced environmental change segment set.

[0055] S5: Using the set of environmental change fragments induced by pests, compare the co-occurrence characteristics of environmental change fragments and the order of pest appearance under different growth stages, determine the distribution pattern of the direction of environmental change before and after the appearance of pests, classify and organize them according to crop growth stage and pest type, and establish a record set of environmentally induced pest triggering chains.

[0056] The unified time-based environmental pest observation set includes time consistency markers, synchronous environmental element recording markers, pest observation alignment markers, and available observation integrity markers. The growth stage change segment set includes growth state transition segments, synchronous growth characteristic change segments, growth continuity segment markers, and growth stage boundary markers. The growth pest association anchor point set includes growth stage switching markers, pest change corresponding markers, growth pest time adjacency markers, and growth stage correspondence markers. The pest-induced environmental change segment set includes environmental state transition segments, environmental change direction segments, pest occurrence association segments, and environmental pest time alignment segments. The environmental-induced pest triggering chain record set includes growth stage environmental association chains, pest occurrence time sequence association chains, environmental change direction association chains, and growth stage pest corresponding chain records.

[0057] Please see Figure 2 The specific steps for obtaining a unified time-environment insect observation set are as follows:

[0058] S111: Collect the collection time status of air temperature, air humidity, precipitation, sunshine duration, soil moisture and insect trapping records in the crop planting area, collect the timestamps and collection status codes of the records, filter the valid status codes, and generate a collection time index table.

[0059] A Modbus-RTU polling command is sent to distributed sensor nodes deployed in crop planting areas (such as winter wheat planting demonstration areas). The sensor nodes respond to the command and upload raw data packets containing air temperature, air humidity, precipitation records, light duration records, soil moisture records, and insect trapping records. The decoder in the central control unit performs hexadecimal parsing on the raw data packets, extracting the numerical parts of each environmental parameter, as well as the acquisition record timestamps (Unix timestamp format, accurate to milliseconds) and acquisition status codes (8-bit binary codes) transmitted along with the data packets. The status code "00001010" is defined as the standard mask for normal sensor operation and valid data, while status codes "11110000" and higher are defined as the range of equipment failure or data anomalies. The processor processes each extracted acquisition status code and standard mask. Bitwise AND operations and numerical comparisons are performed. When the acquisition status code equals "00001010", the data is determined to be a valid item. When the acquisition status code does not equal the value, it is determined to be an invalid item and is directly removed. After filtering, the processor extracts the acquisition record timestamp from the valid data items as the primary key, and writes the corresponding environmental parameter values ​​such as air temperature and humidity and the device ID as attribute values ​​into a pre-built database table to generate an acquisition time index table, as shown in Table 1. Table 1 lists the raw data and filtering results collected by the monitoring points in the winter wheat planting area from 14:00 to 14:40 on May 20, 2024. The status code is represented in hexadecimal, and 0x0A corresponds to binary 00001010. For the data in sequence 2 in Table 1, since the status code is 0xFF (device offline), the processor performs a removal operation and generates an acquisition time index table.

[0060] Table 1: Status of Raw Data Collection at Monitoring Points

[0061] ;

[0062] See Table 1 for the filtering logic of the raw collected data. Only records with a status code of 0x0A are retained for subsequent processing.

[0063] S112: Based on the acquisition time index table, acquire the gateway timestamp sequence, the acquisition device timestamp sequence, and the gateway receiving time sequence, and acquire the communication sequence index and the ratio of the arrival interval of adjacent communication, compare the corresponding communication response order, calculate the synchronization offset, perform interval mapping based on the synchronization offset and the zero offset reference, mark the offset direction, and obtain the time synchronization discrimination result;

[0064] Access the data table named "Time_Index_Table" to record detailed communication interaction metadata of devices entering the network within a specific time window. Locate and extract the communication records of device ID "Sensor_A1" between 09:00:00 and 09:00:05 on October 15, 2023. Parse the database fields line by line to construct sequence data and collect the gateway timestamp sequence. The sequence record shows the local time of the gateway when it sends a synchronization request; the specific value is then extracted. Simultaneously collect device timestamp sequences This refers to the local clock record of the terminal device when it receives an instruction or sends data; the value is... And collect the time sequence received by the gateway. That is, the moment when the gateway physical layer actually receives the terminal response signal, the value is... The Pandas library is used to align the indices of the three sequences and extract the communication order index. and using the formula Calculate the ratio of arrival intervals between adjacent communications, in order to For example, let's substitute the numerical calculation. Compare the corresponding communication response order and calculate the ratio. Compare with the preset linear transmission reference value 1, if This confirms that the communication sequence was not out of order. The synchronization offset is then calculated using the formula... Perform point-by-point calculations, substituting the first set of data to obtain... Substituting the second set of data, we get Based on the synchronization offset and the zero offset reference, a range mapping is performed and the offset direction is marked. The zero offset reference is then set. And set the synchronization tolerance threshold. The threshold The setting is based on network jitter variance. Three times that, assuming statistics from the first 100 communications. Then set Construct the decision interval The "lagging zone" For "synchronization zone", The "advanced zone" will be calculated as follows: Mapped to The interval is determined to be "synchronous". Mapped to The interval is marked with a "negative offset" (hysteresis) in the offset direction. Mapped to The interval is marked with a "positive offset" (leading) direction. The judgment marks of the summarized sampling points are used to obtain the time synchronization discrimination result. The value is... The Pandas library is used to align the indices of the three sequences and extract the communication order index. and using the formula Calculate the ratio of arrival intervals between adjacent communications, in order to For example, let's substitute the numerical calculation. Compare the corresponding communication response order and calculate the ratio. Compare with the preset linear transmission reference value 1, if This confirms that the communication sequence was not out of order. The synchronization offset is then calculated using the formula... Perform point-by-point calculations, substituting the first set of data to obtain... Substituting the second set of data, we get Based on the synchronization offset and the zero offset reference, a range mapping is performed and the offset direction is marked. The zero offset reference is then set. And set the synchronization tolerance threshold. The threshold The setting is based on network jitter variance. Three times that, assuming statistics from the first 100 communications. Then set Construct the decision interval The "lagging zone" For "synchronization zone", The "advanced zone" will be calculated as follows: Mapped to The interval is determined to be "synchronous". Mapped to The interval is marked with a "negative offset" (hysteresis) in the offset direction. Mapped to The interval is marked with a "positive offset" (leading) offset direction. The judgment marks of the sampling points are summarized to obtain the time synchronization judgment result.

[0065] S113: Call the time synchronization discrimination result, adjust the time stamp of the acquisition device and write the correction timestamp sequence, and record and aggregate the air temperature, air humidity, precipitation, light duration, soil moisture and insect trapping based on the correction timestamp to generate a unified time environmental insect observation set.

[0066] When the determination result is "asynchronous", the time correction procedure is initiated, and the average difference between the gateway timestamp and the acquisition device timestamp obtained in the aforementioned calculation process is extracted as the correction factor (the average difference is approximately...). The processor iterates through all data items in the acquisition time index table, adds a correction factor to the acquisition device timestamp value, and generates a correction timestamp sequence. For example, the original device timestamp of 2945ms is corrected to 2998.6ms (rounded to 2999ms). If the judgment result is "synchronization", the original acquisition device timestamp is directly reused as the correction timestamp sequence. After the timestamp correction is completed, the multi-source data fusion operation is performed, and the time aggregation window is set to... ms means that, taking each corrected time point as the center, we search for valid data uploaded by sensors (air temperature and humidity sensors, light sensors, soil sensors, and insect monitoring lamps) within the time window, and use linear interpolation to complete the data from the previous and next time points to generate a unified time environment insect observation set.

[0067] Please see Figure 3 The specific steps for obtaining the set of growth stage change segments are as follows:

[0068] S211: Based on the unified time environment insect observation set, extract the crop plant height change record, leaf area change record and accumulated temperature change record, align the three types of records according to the observation time, and determine the positive and negative signs of the numerical difference results of adjacent observation times to generate a growth index change direction sequence set.

[0069] To meet the growth monitoring needs of winter wheat from the jointing stage to the heading stage, plant height (in cm) and leaf area (in cm) values ​​at fixed daily observation times (e.g., 14:00) were extracted from the database. The processor constructs a three-dimensional array matrix using the daily effective accumulated temperature (unit: °C) and the accumulated temperature value for the day, respectively, to store the above three parameters under the time series. The processor performs a difference operation to obtain the current time value. The value minus the previous moment The numerical value is used to obtain the corresponding gradient value of the change, and the zero drift threshold of the plant height change is set as... cm, the zero-drift threshold for leaf area change is cm2, the zero-drift threshold of accumulated temperature change is ℃ (the threshold is set based on twice the sensor measurement accuracy error). The processor traverses the difference results. When the difference in plant height is greater than 0.1, the direction of plant height change is marked as "1" (positive growth), less than -0.1 is marked as "-1" (reverse decline), and between -0.1 and 0.1 is marked as "0" (stagnation). Similarly, the leaf area and accumulated temperature data are processed into ternary symbols to generate a set of growth index change direction sequences, as shown in Table 2. Table 2 lists the original data and processed direction sequences of winter wheat monitoring points for 5 consecutive observation days. The data on April 12th was judged to have a change direction of 0 because the accumulated temperature change did not exceed the threshold of 1.0℃. On April 13th, all three indicators showed significant positive changes and were marked as [1, 1, 1]. This completes the conversion of continuous values ​​into discrete state vectors.

[0070] Table 2: Raw Data and Trends of Crop Growth Indicators

[0071] ;

[0072] See Table 2, which shows the process of converting the original measurement data into a discrete direction sequence after differential thresholding. Serial numbers 4 and 5 show the continuous and consistent growth state.

[0073] S212: Based on the sequence set of growth index change directions, the consistency of the plant height change direction, leaf area change direction and accumulated temperature change direction at the same observation time is judged. When the three types of directions change from zero to non-zero and remain consistent in continuous observation time, the corresponding time index is recorded and the segments are connected to obtain the continuous segment information of growth state transition.

[0074] To perform a consistency scan of the multidimensional data, the processor sets logical judgment rules: only when at the same time... When the values ​​of the changes in plant height, leaf area, and accumulated temperature are all equal and not zero (i.e., the sequence vector is [1, 1, 1] or [-1, -1, -1]), the time point is determined as an "effective growth active point." The processor traverses the time axis in the forward direction to find the jump point in the growth state, i.e., the time point. The state vector contains at least one 0 (e.g., [1, 0, 0] or [0, 0, 0]), while at time... Transforming into a completely uniform nonzero state (e.g., [1, 1, 1]) will change the time step. Marked as the starting index of the segment Continue searching backwards until a certain moment is detected. If the state vector reappears at 0 or its direction is inconsistent, the time will be... Marked as segment termination index The processor connects the successfully paired start index and end index to form a closed growth state transition continuous segment, and repeats this operation until the entire observation period is traversed to obtain the growth state transition continuous segment information.

[0075] S213: Based on the information of continuous segments of growth state transition, retrieve the start and end time positions of the segments, and use the starting point of the segment as the boundary benchmark of the growth state, using the formula;

[0076] ;

[0077] Calculate the growth stage boundary feature values, split and label the growth stage before and after according to the complete observation time axis, and generate a set of growth stage change segments;

[0078] in, Represents the characteristic value that marks the boundary between growth stages. Representative numbering rules The start time value of the corresponding segment. Representative numbering rules The corresponding end time value for the segment. The arithmetic mean of the start times for all segments. The arithmetic mean of the termination times of all segments. Represents the normalized segment-time modulation coefficient, Represents the number of segments;

[0079] The formula's operational logic lies in the following: the first term is calculated by summing the lengths of the segments themselves. "Deviation of the segment from the overall timeline" and and utilize Weighting is applied to amplify the weight of critical growth periods that are long in duration and occur at times deviating from the average position. The denominator... It serves to normalize energy and eliminate dimensional differences caused by varying numbers of segments. (Second term) This simply calculates the average duration of the segment. The purpose of subtracting the two (difference) is to remove the background of the normal growth duration and highlight the discrete and uneven characteristics of the crop growth rhythm in time distribution.

[0080] Set up a practical example: Assume that during the monitoring period, [the following were identified] Key sections:

[0081] Section 1: Starting Point (Day 100), End The duration is 10 days;

[0082] Section 2: Start ,termination The monitoring period is 15 days, and the total monitoring period is set to 100 days.

[0083] The calculation process is as follows:

[0084] Calculate the average: , ;

[0085] Calculate coefficients : , ;

[0086] Calculate the first numerator cumulative term:

[0087] Section 1: ;

[0088] Section 2: ;

[0089] Sum of numerators: ;

[0090] Calculate the denominator of the first term: ;

[0091] Calculate the result of the first term: ;

[0092] Calculate the second term (average duration): ;

[0093] Substitute into the formula to calculate:

[0094] ;

[0095] The calculated eigenvalues Compared with the baseline interval in the standard winter wheat growth model The comparison results are within the range, indicating that the current stage growth rhythm of the crop conforms to the growth pattern of the superior variety. If the value exceeds the range, it means that the crop growth stage distribution is too scattered or concentrated, and agricultural intervention suggestions need to be triggered. The processor divides and marks the complete time axis according to the calculated start and end time points (such as day 100 and day 150) to generate a set of growth stage change segments.

[0096] Please see Figure 4 The specific steps for obtaining the set of anchor points associated with pests and diseases are as follows:

[0097] S311: Based on the set of growth stage change segments, monitor the insect population record time series near the boundary of the growth segment, call the insect population quantity value and time index in adjacent time windows, use difference operation to obtain the change sequence of continuous time moments, and make a consistency judgment on the sign direction of the change to generate an insect population change trend sequence.

[0098] The processor reads the boundary time points of each marked growth segment. For example, it extracts April 15th as the boundary time for winter wheat to transition from the jointing stage to the booting stage. It sets the radius of the time retrieval window, referencing the average lag days between pest outbreaks and climate change in the data. A practical example is set as follows: Monitoring data of the same crop in the same region over the past three years is selected, and the average time difference between environmental factor mutation points and pest outbreak points is calculated to be 3 days, 4 days, and 3.5 days, respectively. The average of 3.5 is rounded up to 4 days. The processor constructs a time retrieval window based on the radius, from April 11th to April 19th. Within this time range, it extracts daily pest trapping records from a unified time-based environmental pest observation set, obtaining the corresponding pest population count and time index. The processor performs a first-order difference operation on the extracted consecutive pest population counts. For example, the count on April 13th is 145, and on April 12th it is 1. With 32 insects, the difference value is 13. To filter out background noise, a baseline value for judging insect population fluctuations is set. The calculation process for the baseline value is as follows: select the daily change in insect population over 10 consecutive days during the inactive period, calculate the standard deviation, and set the baseline value to be twice the standard deviation. Assuming the standard deviation of the change during the inactive period is 3, the baseline value is 6. The processor compares the daily difference results with the baseline value. When the difference value is greater than 6, it is judged as a surge; when the difference value is less than -6, it is judged as a sharp decrease; when the absolute value of the difference value is less than or equal to 6, it is judged as stable. For example, the change value of 13 on April 13 is greater than the threshold of 6, so it is marked as a surge. The change value of 7 on April 14 is greater than 6, so it is also marked as a surge. Only subsequences with the same non-zero directional marking for 2 consecutive days or more are retained. The starting time point is taken as the moment of sudden change in insect population trend, and the continuous direction is taken as the trend attribute to generate an insect population change trend sequence.

[0099] S312: Based on the insect infestation change trend sequence, call the corresponding growth segment boundary time index, perform time interval calculation for the time of insect infestation change and the time of growth stage switching, perform logical judgment based on the preset adjacent judgment benchmark value, filter positions that meet the time sequence adjacent conditions, and obtain the time sequence adjacent position set.

[0100] The occurrence time of each insect infestation mutation event is extracted. For example, the identified infestation surge began on April 13th, which is the 103rd day. Simultaneously, the nearest neighbor growth stage switching time recorded in the growth stage change segment is retrieved, such as April 15th, which is the 105th day. The processor performs time interval calculation, which yields a time interval of 2 days. Based on crop phenology principles, a neighboring threshold is set. The threshold setting process is as follows: The average duration of the current growth stage (e.g., 15 days) and the crop's sensitivity coefficient to insect pests (with a value range of 0) are obtained. For example, the critical growth period of winter wheat is set to 0.25. The baseline value is calculated by multiplying the average duration by the sensitivity coefficient and rounding up, i.e., 15 multiplied by 0.25 equals 3.75, which is rounded up to 4 days. The rationale for setting this value is that the physiological changes of the covered crop cause changes in the microenvironment, which in turn induce pest aggregation. The processor will make a logical judgment between the calculated time interval and the adjacent judgment baseline value. If the time interval is less than or equal to the baseline value, it is determined that the pest change and the growth stage switch are strongly correlated in time, and the set of adjacent positions in time is obtained.

[0101] S313: Call the time-series adjacent location set, retrieve the corresponding previous and subsequent growth stage identifier codes and insect pest change marker values, perform identifier matching and state consistency judgment, retain the locations that simultaneously meet the conditions of growth state change and insect pest change, and uniformly number and label them to generate a growth pest association anchor point set.

[0102] The processor retrieves the corresponding attribute information from the database to obtain the identification codes of the preceding and following growth stages at the time of growth stage transition, such as from the jointing stage to the booting stage. A stage transition code is generated by combining these codes. At the same time, the processor obtains the pest change marker value corresponding to the time of sudden change in pest situation, such as a surge or a sharp decrease. The processor constructs an association logic verification matrix, which is pre-set with legal association patterns defined by the agronomic expert knowledge base. For example, a surge in pest situation should accompany the transition from the jointing stage to the booting stage (due to the increase in the plant's vegetative body and leaf area index, which is conducive to aphid reproduction), while a sharp decrease in pest situation should not occur without external intervention. The extracted combinations are matched with the verification matrix. If the match is successful, the position is determined to meet biological consistency. The processor assigns a unified number to the position, using a combination of crop ID, growth stage code, pest code, and timestamp to generate a set of growth-pest association anchor points.

[0103] Please see Figure 5 The specific steps for obtaining the set of environmental change fragments induced by pests are as follows:

[0104] S411: Based on the growth and pest association anchor point set, collect air temperature, air humidity, precipitation, sunshine duration and soil moisture records within the corresponding time window before and after, perform time series difference operation on the values ​​of the same environmental element at adjacent time points, and make consistency judgment on the direction of the difference results to generate a sequence of environmental element change direction.

[0105] Analyze the center timestamp corresponding to each anchor point (For example, April 13, 2024, which is the 103rd cumulative day), based on the effective time window of the impact of environmental factors on insect behavior, the scope of backtracking sampling is set before and after. The range was set with reference to the average lag period of insect population response to the environment. Day, construct an environmental data acquisition window During this time period, raw values ​​of air temperature, air humidity, precipitation, sunshine duration, and soil moisture content are retrieved to generate a five-dimensional environmental parameter matrix. The processor performs a first-order temporal difference operation on each column of the matrix, calculated using the following formula: ,in The environmental element value at the current moment. The values ​​are from the previous moment. To eliminate the influence of sensor measurement noise and natural microclimate fluctuations, change judgment thresholds are set for each environmental element. The threshold is set based on three times the sensor's measurement accuracy error. For example, if the accuracy of an air temperature sensor is... Then set a temperature change threshold. Air humidity accuracy is Set humidity threshold The processor iterates through the difference result sequence and calculates the... With the corresponding threshold Perform absolute value comparison and sign determination when When the direction of the change is "+1" (significant increase), when When, it is marked as "-1" (significant decrease), when When the time is set to "0" (steady-state fluctuation), as shown in Table 3, Table 3 lists the temperature data difference process centered on April 13. The difference value on April 12 is 0.8, which is greater than the threshold of 0.3, and is marked as +1. The difference value on April 13 is 1.5, and is marked as +1. The difference value on April 11 is 0.2, which does not exceed the threshold, and is marked as 0. This generates a sequence of environmental element change directions.

[0106] Table 3: Time Series Difference Determination Table for Environmental Elements

[0107] ;

[0108] See Table 3, which shows that after time-series differential and threshold filtering, the air temperature data showed a continuous upward trend from April 12 to April 14.

[0109] S412: Based on the sequence of environmental element change directions, perform a merging operation on the segments with consistent directions of the same environmental element on the continuous time index, determine the location boundary where the environmental state shifts on the time axis, record the start and end time indices of continuous changes, and obtain a set of continuous segments of environmental state transition.

[0110] For each independent environmental dimension (such as an air temperature sequence), a same-direction merging algorithm is executed. The processor initializes a status register, scans from the first valid time point of the sequence, and reads the change direction marker at the current moment. ,like If not zero, record the current time. Start time index for provisional fragment It then locks the current direction state (e.g., "+1") and continues reading the next moment. The mark Perform a consistency logic check; if and If the values ​​are equal, the segment remains open and the scan continues. and If the values ​​are not equal (become 0 or reversed), the current continuous change process is considered terminated, and the time is recorded. For segment end time index The processor calculates the duration of a segment. (Unit: days) To filter out occasional short-term fluctuations, a minimum continuous length baseline value is set. The baseline value is set based on the typical cycle of weather changes, such as a frontal passage or warming process lasting at least 2 days, therefore the baseline is set accordingly. , calculate and When comparing, When the event is confirmed to be a valid environmental state transition event, its attributes (such as "air temperature - continuous temperature rise") and start and end times are recorded. Write it into a list to obtain a set of continuous fragments of environment state transitions.

[0111] S413: Call the continuous fragment set of environmental state transition, retrieve the pest occurrence record index within the corresponding time range, perform alignment comparison between the start and end time of the environmental change fragment and the pest record time, retain the fragments with the established time overlap relationship, and generate a fragment set of pest-induced environmental change fragments.

[0112] Retrieve the corresponding pest occurrence record index within the time range to obtain the specific time points of pest outbreaks or significant increases. (For example, a specific date like April 13th), the processor performs a time-domain alignment comparison to construct a time overlap logic judgment formula: ,in To allow for an advance induction period, used to cover the cumulative effect time of environmental changes preceding pest occurrence, a threshold is set. Today, let's verify using specific data: For the aforementioned "rapid warming continuum" [April 12th, April 14th], the start time is April 12th, and the end time is April 14th, which coincides with the pest infestation period. The date is April 13th, logically determined as follows: If both True and True, it determines that there is a temporal inclusion and induction relationship between the warming fragment and the occurrence of pests, retains the fragment, and marks it as a "pest-associated warming event". Conversely, if there is a precipitation fragment that occurs from April 16 to April 17, although it is within the collection window, its start time on the 16th is later than the occurrence time of the pests on the 13th. The logical judgment is False, and it is judged as an irrelevant environmental fluctuation after the occurrence of pests and is removed. A set of fragments of environmental changes induced by pests is generated.

[0113] Please see Figure 6 The specific steps for obtaining the environmentally induced pest triggering chain record set are as follows:

[0114] S511: Based on the set of environmental change fragments induced by pests, the crop growth stage identifier and pest occurrence time index corresponding to the environmental change fragment are called. The order of comparison between the start time of environmental change and the occurrence time of pests under the same growth stage is performed to record the temporal relationship between environmental change and pest occurrence, and generate a temporal co-occurrence feature sequence of environmental pests.

[0115] Extract the core attributes contained in each record, including the start time of the environmental change segment. Crop growth stage identifiers (such as "S04-jointing stage") and corresponding pest occurrence time indexes The processor constructs a timing logic comparator, which performs a numerical judgment on the time priority of each record under the same growth stage identifier and calculates the timing difference. Set the discrimination rule: if This indicates that the change in environmental factors (such as increased temperature and decreased humidity) begins earlier than the outbreak of pests, classifying it as a "pre-induction type" relationship, and assigning the time sequence label "LEAD"; if This indicates that environmental changes and pest infestations occurred almost simultaneously or with a slight lag, classifying it as a "companionate / lagging response type" relationship. The time sequence is assigned the label "SYNC / LAG". Taking the data in Table 3 as an example, the environmental warming began on April 12th. The insect infestation surge occurred on April 13th. ), calculated Therefore, the warming event is marked as "LEAD", which means that environmental change precedes the occurrence of pests, implying the probability that environmental factors are the trigger. The processor traverses the fragments in the set and generates a sequence of co-occurrence characteristics of environmental pests.

[0116] S512: Based on the co-occurrence characteristic sequence of environmental pests, perform statistical merging operation on the distribution state of the direction of change of environmental elements before and after the appearance of pests, aggregate the frequency of the same direction relationship, and complete the state mapping in the growth stage dimension to obtain the distribution pattern set of the direction of environmental change.

[0117] Initialize a multidimensional frequency statistics matrix. The matrix dimensions include crop growth stage (e.g., jointing stage, booting stage), environmental factor type (e.g., temperature, humidity), direction of change (+1, -1), and time series marker (LEAD, SYNC / LAG). The processor performs a traversal and merge operation, mapping each record in the sequence to a corresponding cell in the matrix and accumulating the count values ​​within each cell. For example, statistics show that during the "jointing stage," there are 15 records related to "air temperature" with a direction of "+1" (warming) and a time series of "LEAD," while during the same stage, there are 12 records related to "air humidity" with a direction of "-1" (dehumidification) and a time series of "LEAD." Simultaneously, the relative frequency of each combination is calculated, and a significance screening threshold is set. The threshold is set based on 5% of the total sample size. Assuming the total number of records is 200, then... The processor treats combinations with statistical frequencies below a threshold as random noise and filters them out, retaining only high-frequency co-occurrence patterns. It completes state mapping in the growth stage dimension, and outputs the selected high-frequency combinations in a structured manner to obtain a set of distribution patterns of environmental change directions.

[0118] S513: Call the distribution pattern set of environmental change direction, retrieve the corresponding pest type identifier and crop growth stage code, classify and organize the environmental change direction and pest occurrence order, aggregate and number the records of the same stage and the same pest, and establish a record set of environmentally induced pest triggering chain.

[0119] Using crop growth stage codes (e.g., S04) and pest type identifiers (e.g., P-APHID-aphid) as primary keys, a chained data structure is created for retrieval and clustering, aiming to reconstruct "environmental change". Crop / Microenvironment Response The processor logically connects the environmental change patterns of the same stage and the same pest outbreak to form a causal chain. For example, in stage S04 targeting aphids, if "temperature +1" and "humidity -1" are identified as high-frequency lead-inducing factors, then the two are aggregated to construct a composite trigger chain: "[S04] Temperature rises..." Humidity decrease [P-APHID] outbreak, and assign a unique index number to the trigger chain (such as CHAIN-S04-APHID-001), while recording the number of samples supported by the chain and the confidence score (calculated based on statistical frequency). If "rainfall +1" is found to be the main antecedent factor in another growth stage S05 (heading stage), another independent trigger chain is generated, and an environmentally induced pest trigger chain record set is established.

[0120] The system for analyzing the relationship between crop growth environment and pest induction is used to perform the above-mentioned methods for analyzing the relationship between crop growth environment and pest induction. The system includes:

[0121] The time consistency correction module collects the collection timestamps of air temperature, air humidity, precipitation, light duration, soil moisture content and insect trapping in the crop planting area, compares the communication response sequence between the field gateway and the collection equipment, estimates and corrects the time offset, and forms a unified time environment insect observation set.

[0122] The growth state segmentation module, based on a unified time-environment insect observation set, acquires the sequences of crop plant height changes, leaf area changes, and accumulated temperature changes, compares the direction of change at adjacent observation times, determines the continuous segments where the growth state transitions from stable to changing, and generates a set of growth stage change segments.

[0123] The insect-related anchoring module uses a set of growth stage change segments to compare the changing trend of the number of insects trapped before and after the segment boundary, calculates and judges the temporal proximity between insect changes and growth stage switching, filters out locations where growth state changes and insect fluctuations occur simultaneously, and generates a set of growth-insect-related anchor points.

[0124] The environmental transfer identification module, based on the growth pest association anchor point set, obtains the corresponding before and after air temperature, air humidity, precipitation, light duration and soil moisture sequence, compares and judges the continuous segments of the direction of change of environmental elements, aligns the environmental change segments with the time of pest occurrence, and generates a set of environmental change segments induced by pests.

[0125] The trigger chain construction module utilizes a set of environmental change fragments induced by pests to compare the co-occurrence distribution of the direction of environmental change and the order of pest occurrence under differentiated growth stages, thereby completing the identification and classification of environmental change patterns and pest types, and constructing a record set of environmentally induced pest trigger chains.

[0126] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention in any other way. Any person skilled in the art may make changes or modifications to the above-disclosed technical content to create equivalent embodiments that can be applied to other fields. However, any simple modifications, equivalent changes, and modifications made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the protection scope of the present invention.

Claims

1. A method for analyzing the relationship between crop growth environment and pest induction, characterized in that, Includes the following steps: S1: Collect data on the collection time and status of air temperature, air humidity, precipitation, sunshine duration, soil moisture, and insect trapping in the crop planting area, as well as crop growth monitoring data; identify the communication response sequence of the collection equipment; determine the offset of the time records; and generate a unified time environment insect observation set. S2: Based on the unified time environment insect observation set, extract the crop plant height change record, leaf area change record and accumulated temperature change record, compare the change direction of crop growth status at adjacent observation times, and obtain the growth stage change segment set; S3: By comparing the changing trends of insect records near the boundary of the growth segment through the set of growth stage change segments, the close proximity of insect changes and growth stage switching in time sequence is determined, and time locations that simultaneously have the characteristics of growth state change and insect change are selected to generate a set of growth pest association anchor points. S4: Analyze the air temperature, air humidity, precipitation, light duration and soil moisture records before and after the growth pest association anchor point set, compare the direction of change of environmental elements in the time series before and after the growth pest association anchor point, determine the continuous segments in which the environmental state has changed, and obtain the set of environmental change segments induced by pests. S5: Using the aforementioned set of environmental change fragments induced by pests, compare the co-occurrence characteristics of environmental change fragments and the order of pest appearance under different growth stages, determine the distribution pattern of the direction of environmental change before and after the appearance of pests, classify and organize them according to crop growth stage and pest type, and establish a record set of environmentally induced pest triggering chains. The environmental-induced pest triggering chain record set includes the growth stage environmental association chain, the pest occurrence time association chain, the environmental change direction association chain, and the corresponding chain record of pests in the growth stage. The specific steps for obtaining the environmentally induced pest triggering chain record set are as follows: S511: Based on the set of environmental change fragments induced by pests, call the crop growth stage identifier and pest occurrence time index corresponding to the environmental change fragments, perform sequential comparison of the start time of environmental change and the occurrence time of pests under the same growth stage, record the temporal relationship of environmental change preceding or following the occurrence of pests, and generate a temporal co-occurrence feature sequence of environmental pests. S512: Based on the environmental pest co-occurrence feature sequence, perform statistical merging operation on the distribution state of the environmental element change direction before and after the pest occurrence, perform frequency aggregation on the same direction relationship, and complete the state mapping on the growth stage dimension to obtain the environmental change direction distribution pattern set. S513: Call the set of distribution patterns of environmental change direction, retrieve the corresponding pest type identifier and crop growth stage code, classify and organize the environmental change direction and the order of pest appearance, aggregate and number the records corresponding to the same stage and the same pest, and establish a record set of environmentally induced pest triggering chain.

2. The method for analyzing the relationship between crop growth environment and pest induction according to claim 1, characterized in that, The unified time-environment insect pest observation set includes time consistency identifiers, environmental element synchronous recording identifiers, insect pest observation alignment identifiers, and available observation integrity identifiers. The growth stage change segment set includes growth state transition segments, growth characteristic synchronous change segments, growth continuity segment identifiers, and growth stage boundary identifiers. The growth pest association anchor point set includes growth stage switching identifiers, insect pest change corresponding identifiers, growth pest time adjacency identifiers, and growth stage correspondence identifiers. The pest-induced environmental change segment set includes environmental state transition segments, environmental change direction segments, pest occurrence association segments, and environmental pest time alignment segments.

3. The method for analyzing the relationship between crop growth environment and pest induction according to claim 1, characterized in that, The specific steps for obtaining the unified time-environment insect infestation observation set are as follows: S111: Collect the collection time status of air temperature, air humidity, precipitation, sunshine duration, soil moisture and insect trapping records in the crop planting area, collect the timestamps and collection status codes of the records, filter the valid status codes, and generate a collection time index table. S112: Based on the acquisition time index table, acquire the gateway timestamp sequence, the acquisition device timestamp sequence, and the gateway receiving time sequence, and acquire the communication sequence index and the ratio of the arrival interval of adjacent communication, compare the corresponding communication response order, calculate the synchronization offset, perform interval mapping based on the synchronization offset and the zero offset reference, and mark the offset direction to obtain the time synchronization discrimination result; S113: Call the time synchronization discrimination result, adjust the time stamp of the acquisition device and write the correction timestamp sequence, and record and aggregate the air temperature, air humidity, precipitation, light duration, soil moisture and insect trapping based on the correction timestamp to generate a unified time environment insect observation set.

4. The method for analyzing the relationship between crop growth environment and pest induction according to claim 3, characterized in that, The specific steps for obtaining the set of growth stage change segments are as follows: S211: Based on the unified time environment insect observation set, extract crop plant height change records, leaf area change records, and accumulated temperature change records, align the three types of records according to the observation time, and determine the positive or negative sign of the numerical difference results of adjacent observation times to generate a growth index change direction sequence set. S212: Based on the growth index change direction sequence set, the consistency of the plant height change direction, leaf area change direction and accumulated temperature change direction at the same observation time is judged. When the three types of directions change from zero to non-zero and remain consistent in continuous observation time, the corresponding time index is recorded and the segments are connected to obtain the continuous segment information of growth state transition. S213: Based on the continuous segment information of the growth state transition, retrieve the start and end time positions of the segment, take the starting point of the segment as the growth state boundary benchmark, calculate the growth stage boundary feature value, and perform front and back split labeling according to the complete observation time axis to generate a set of growth stage change segments.

5. The method for analyzing the relationship between crop growth environment and pest induction according to claim 4, characterized in that, The specific steps for obtaining the set of growth pest association anchor points are as follows: S311: Based on the set of growth stage change segments, monitor the insect population record time series near the boundary of the growth segment, call the insect population quantity value and time index in adjacent time windows, use difference operation to obtain the change sequence of continuous time moments, and make a consistency judgment on the sign direction of the change to generate an insect population change trend sequence. S312: Based on the insect infestation change trend sequence, call the corresponding growth segment boundary time index, perform time interval calculation for the time of insect infestation change and the time of growth stage switching, make logical judgment based on the preset adjacent judgment benchmark value, filter positions that meet the time sequence adjacent conditions, and obtain the time sequence adjacent position set. S313: Call the set of adjacent time-series locations, retrieve the corresponding previous and subsequent growth stage identifier codes and pest change marker values, perform identifier matching and state consistency judgment, retain the locations that simultaneously meet the conditions of growth state change and pest change, and uniformly number and label them to generate a set of growth pest association anchor points.

6. The method for analyzing the relationship between crop growth environment and pest induction according to claim 5, characterized in that, The specific steps for obtaining the set of environmental change fragments induced by pests are as follows: S411: Based on the aforementioned set of growth pest association anchor points, collect records of air temperature, air humidity, precipitation, sunshine duration and soil moisture content within the corresponding time windows before and after, perform time-series difference operation on the values ​​of the same environmental element at adjacent times, and determine the consistency of the difference results to generate a sequence of environmental element change directions. S412: Based on the sequence of environmental element change directions, perform a merging operation on the consistent direction segments of the same environmental element on the continuous time index, determine the position boundary where the environmental state changes on the time axis, record the start and end time indices of continuous changes, and obtain a set of continuous segments of environmental state transition. S413: Call the set of continuous environmental state transition segments, retrieve the pest occurrence record index within the corresponding time range, perform alignment comparison between the start and end time of the environmental change segment and the pest record time, retain segments with established time overlap, and generate a set of pest-induced environmental change segments.

7. A system for analyzing the relationship between crop growth environment and pest induction, characterized in that, The system is used to implement the method for analyzing the relationship between crop growth environment and pest induction as described in any one of claims 1-6, and the system includes: The time consistency correction module collects the collection timestamps of air temperature, air humidity, precipitation, light duration, soil moisture content and insect trapping in the crop planting area, compares the communication response sequence between the field gateway and the collection equipment, estimates and corrects the time offset, and forms a unified time environment insect observation set. The growth state segmentation module, based on the unified time environment insect observation set, obtains the sequence of crop plant height change, leaf area change and accumulated temperature change, compares the change direction of adjacent observation times, determines the continuous segment of growth state transitioning from stable to changing, and generates a set of growth stage change segments. The insect-related anchoring module uses the set of growth stage change segments to compare the changing trend of the number of insects trapped before and after the segment boundary, calculates and judges the temporal proximity relationship between insect changes and growth stage switching, filters out the locations where growth state changes and insect fluctuations occur simultaneously, and generates a set of growth pest-related anchor points. The environmental transfer identification module, based on the set of growth pest-related anchor points, obtains the corresponding sequence of air temperature, air humidity, precipitation, light duration and soil moisture content before and after, compares and judges the continuous segments of the direction of change of environmental elements, aligns the environmental change segments with the time of pest occurrence, and generates a set of environmental change segments induced by pests. The triggering chain construction module utilizes the set of environmental change fragments induced by pests to compare the co-occurrence distribution of the direction of environmental change and the order of pest occurrence under differentiated growth stages, completes the discrimination and classification of environmental change patterns and pest types, and constructs a record set of environmentally induced pest triggering chains.

Images