Crop phenology monitoring method, system and equipment based on continuous accumulated temperature driving and medium

Abstract

The invention relates to a crop phenology monitoring method, system and device based on continuous accumulated temperature driving and a medium. The method comprises the following steps: acquiring optical remote sensing data of crops, and calculating to obtain annual vegetation index time series data; obtaining annual hour-level temperature monitoring data, constructing and obtaining an annual temperature model based on the annual hour-level temperature monitoring data, and calculating and obtaining annual effective accumulated temperature time sequence data based on the annual temperature model in combination with the growth base point temperature; constructing a phenology monitoring reference shape model based on the vegetation index time series data of each reference year and the effective accumulated temperature time series data of each reference year, and setting a standard phenology period; and based on the target annual vegetation index time series data and the target annual effective accumulated temperature time series data, constructing a target phenology monitoring effective accumulated temperature model, and combining a standard phenology period to generate a target annual phenology monitoring result. By adopting the method, the daily effective accumulated temperature can be calculated more accurately, and the estimation error of the phenological period is reduced.

Description

Technical Field

[0001] This application belongs to the field of phenological monitoring, and in particular relates to a method, system, equipment and medium for crop phenological monitoring based on continuous accumulated temperature. Background Technology

[0002] With the development of agricultural information technology, satellite remote sensing technology has been widely applied in the field of crop phenological monitoring, giving rise to various phenological extraction methods based on remote sensing data. These methods effectively overcome the limitations of traditional ground-based observations, such as limited coverage and low resource utilization, providing crucial support for agricultural production. Among these, the shape model method, with its ability to utilize long-term satellite and ground-based observation data to invert the growth trajectory curves of typical crop vegetation indices and avoid interference from local anomalous observations, has become an important technical direction for high-precision phenological extraction. The traditional shape model method uses calendar time as the dependent variable, assuming a direct correlation between crop growth status and growth time. It extracts phenological periods by constructing multi-year average growth trajectory curves and fitting them to the vegetation index curves of the observation area.

[0003] However, temperature is a key environmental factor affecting crop growth and development, and its interannual fluctuations can lead to significant differences in the same phenological period across different years. Traditional shape models that use calendar time as the dependent variable are insufficient to accurately reflect the driving effect of environmental factors on crop growth, limiting the accuracy of phenological extraction. Some studies have used nonlinear models to integrate environmental variables such as sunshine and temperature to optimize the fitting effect, but the computational complexity is high, which is not conducive to practical application. Other studies have used daily average temperature to calculate effective accumulated temperature instead of calendar time, but this is difficult to capture the contribution of temperature changes to crop growth, resulting in insufficient accuracy in effective accumulated temperature and phenological monitoring. Summary of the Invention

[0004] Therefore, it is necessary to provide a method, system, equipment, and medium for crop phenology monitoring based on continuous accumulated temperature that can accurately calculate effective accumulated temperature and reduce the estimation error of phenological periods, in order to address the above-mentioned technical problems.

[0005] Firstly, this application provides a crop phenology monitoring method based on continuous accumulated temperature, comprising:

[0006] Obtain optical remote sensing data of crops in the target year and optical remote sensing data of crops in each reference year, and calculate the time series data of vegetation index in the target year and the time series data of vegetation index in each reference year;

[0007] Hourly temperature monitoring data for each year is acquired. Based on the hourly temperature monitoring data for each year, a temperature model for each year is constructed. Based on the temperature model for each year and combined with the preset growth baseline temperature, the effective accumulated temperature time series data for each year is calculated. The effective accumulated temperature time series data for each year includes the effective accumulated temperature time series data for the reference year corresponding to the vegetation index time series data for the reference year and the effective accumulated temperature time series data for the target year corresponding to the vegetation index time series data for the target year.

[0008] After time-series registration of the vegetation index time-series data and the effective accumulated temperature time-series data for the reference years, a phenological monitoring reference shape model is constructed based on the vegetation index time-series data and the effective accumulated temperature time-series data for each reference year, and a standard phenological period is set based on the phenological monitoring reference shape model.

[0009] After performing time-series registration of the target year's vegetation index time-series data and the target year's effective accumulated temperature time-series data, a target phenological monitoring effective accumulated temperature model is constructed based on the target year's vegetation index time-series data and the target year's effective accumulated temperature time-series data.

[0010] Based on the phenological monitoring reference shape model and the target phenological monitoring effective accumulated temperature model, combined with the standard phenological period, the phenological monitoring results for the target year are generated.

[0011] In one embodiment, hourly-resolution temperature monitoring data for each year is acquired, a temperature model for each year is constructed based on the hourly-resolution temperature monitoring data for each year, and effective accumulated temperature time series data for each year is calculated based on the temperature model for each year and a preset growth baseline temperature, including:

[0012] Linear trend removal preprocessing and data boundary preprocessing are performed on annual hourly-level resolution temperature monitoring data to obtain frequency domain adaptive enhanced temperature monitoring data.

[0013] Based on the preset annual cycle frequency and the preset daily cycle frequency, the annual cycle spectrum and daily cycle spectrum of the temperature monitoring data are extracted from the frequency domain adaptive temperature monitoring data, respectively.

[0014] Calculate the power spectral density of the annual periodic spectrum of temperature monitoring data at each annual periodic frequency to obtain the annual periodic power spectral density corresponding to each annual periodic frequency. Also calculate the power spectral density of the daily periodic spectrum of temperature monitoring data at each daily periodic frequency to obtain the daily periodic power spectral density corresponding to each daily periodic frequency.

[0015] The annual periodic characteristic frequencies are selected from the annual periodic frequencies in descending order of annual periodic power spectral density until the ratio of the sum of the annual periodic power spectral densities of the annual periodic characteristic frequencies to the total annual power spectral density exceeds the preset annual characteristic frequency energy ratio threshold. Based on the annual periodic characteristic frequencies, the annual periodic spectrum of temperature monitoring data is optimized to obtain the annual periodic enhanced spectrum. The total annual power spectral density is the sum of the annual periodic power spectral densities.

[0016] Intra-day periodic characteristic frequencies are selected from each intra-day periodic frequency in descending order of intra-day periodic power spectral density, until the ratio of the sum of the intra-day periodic power spectral densities of each intra-day periodic characteristic frequency to the total power spectral density of the same year exceeds a preset threshold for the energy ratio of characteristic frequencies within the year. Based on the intra-day periodic characteristic frequencies, the intra-day periodic spectrum of temperature monitoring data is optimized to obtain the intra-day periodic enhanced spectrum. The total intra-day power spectral density is the sum of the power spectral densities of all intra-day periods.

[0017] An annual temperature model is constructed by performing inverse frequency domain transformation on the intra-annual and intra-day periodic enhancement spectra.

[0018] Based on the annual temperature model and combined with the preset growth baseline temperature, the annual effective accumulated temperature data at each effective accumulated temperature sampling time is calculated, and the annual effective accumulated temperature time series data is constructed based on the annual effective accumulated temperature data at each effective accumulated temperature sampling time.

[0019] In one embodiment, the expression for the annual temperature model is: ; ; In the formula, For annual temperature models, This represents the total number of annual periodic characteristic frequencies. This represents the total number of intraday periodic characteristic frequencies. For the serial number The amplitude of the annual periodic characteristic frequency. For the serial number The annual periodic characteristic frequency, For the serial number The phase of the annual periodic characteristic frequency. For the serial number The amplitude of the intraday periodic characteristic frequency. For the serial number Intraday cyclical characteristic frequency, For the serial number The phase of the intraday periodic characteristic frequency. For the serial number Effective accumulated temperature sampling time, For the serial number The annual effective accumulated temperature time series data corresponding to the effective accumulated temperature sampling time. The effective accumulated temperature begins at the starting time. For annual temperature models, This is the growth base point temperature.

[0020] In one embodiment, the target year's phenological monitoring results include a target phenological monitoring effective accumulated temperature calibration model and a target year's phenological period. Based on the phenological monitoring reference shape model and the target phenological monitoring effective accumulated temperature model, combined with standard phenological periods, the target year's phenological monitoring results are generated, including:

[0021] Based on the objective of minimizing the fitting error, the matching error minimization parameters are calculated by matching the reference shape model for phenological monitoring and the effective accumulated temperature model for target phenological monitoring.

[0022] The effective accumulated temperature model for target phenological monitoring is calibrated based on the parameters that minimize matching error, thus generating a calibration model for the effective accumulated temperature model for target phenological monitoring.

[0023] The phenological period of the target year is calculated based on the matching error minimization parameter and the standard phenological period;

[0024] Based on the target phenological monitoring effective accumulated temperature calibration model and the target annual phenological period, the target annual phenological monitoring results are generated.

[0025] In one embodiment, the matching error minimization parameters include horizontal scaling parameters, vertical scaling parameters, and translation parameters. The expressions for the fitting error minimization target and the target year's phenological monitoring results are as follows: ; ; In the formula, To minimize the fitting error, , and These are the horizontal scaling parameters, vertical scaling parameters, and translation parameters, respectively. The total number of effective accumulated temperature pairing points. The serial number of the effective accumulated temperature model for target phenological monitoring is The effective accumulated temperature value at the paired point. The serial number of the effective accumulated temperature model for target phenological monitoring is The vegetation index of the effective accumulated temperature paired with the point, As a reference shape model for phenological monitoring, For the target year's phenological period, This is the standard phenological period.

[0026] In one embodiment, the expression for the vegetation index in the reference annual vegetation index time series data and the target annual vegetation index time series data is: ; In the formula, The vegetation index, This is near-infrared band data. This is red band data.

[0027] In one embodiment, after time-series registration of the vegetation index time-series data and the effective accumulated temperature time-series data for the reference year, a phenological monitoring reference shape model is constructed based on the vegetation index time-series data and the effective accumulated temperature time-series data for each reference year. Standard phenological periods are then set based on the phenological monitoring reference shape model, including:

[0028] After time-series registration of the vegetation index time-series data and the effective accumulated temperature time-series data of the reference year, the effective accumulated temperature model for phenological monitoring of each reference year is constructed with the vegetation index time-series data of each reference year as the dependent variable and the effective accumulated temperature time-series data of each reference year corresponding to the vegetation index time-series data of each reference year as the independent variable.

[0029] The effective accumulated temperature model for phenological monitoring in each reference year was averaged to construct a reference shape model for phenological monitoring.

[0030] Standard phenological periods were identified in the phenological monitoring reference shape model.

[0031] Secondly, this application also provides a crop phenology monitoring system based on continuous accumulated temperature, comprising:

[0032] The remote sensing monitoring data processing module is used to acquire remote sensing data of crop optical monitoring in the target year and crop optical remote sensing data in each reference year, and to calculate the time series data of vegetation index in the target year and the time series data of vegetation index in each reference year.

[0033] The effective accumulated temperature data generation module is used to acquire hourly resolution temperature monitoring data for each year, construct temperature models for each year based on the hourly resolution temperature monitoring data for each year, and calculate the effective accumulated temperature time series data for each year based on the temperature models for each year and the preset growth baseline temperature. Among them, the annual effective accumulated temperature time series data includes the reference annual effective accumulated temperature time series data corresponding to the reference annual vegetation index time series data and the target annual effective accumulated temperature time series data corresponding to the target annual vegetation index time series data.

[0034] The reference shape model construction module is used to perform time-series registration of the vegetation index time-series data and the effective accumulated temperature time-series data of the reference year, and then construct a phenological monitoring reference shape model based on the vegetation index time-series data and the effective accumulated temperature time-series data of each reference year, and set the standard phenological period based on the phenological monitoring reference shape model.

[0035] The target monitoring model construction module is used to perform time-series registration of the target year's vegetation index time-series data and the target year's effective accumulated temperature time-series data, and then construct the target phenological monitoring effective accumulated temperature model based on the target year's vegetation index time-series data and the target year's effective accumulated temperature time-series data.

[0036] The phenological monitoring results generation module is used to generate the phenological monitoring results for the target year based on the phenological monitoring reference shape model and the target phenological monitoring effective accumulated temperature model, combined with the standard phenological period.

[0037] Thirdly, this application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the method as described in any of the first aspects of this application.

[0038] Fourthly, this application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method described in any of the first aspects of this application.

[0039] The aforementioned crop phenological monitoring methods, systems, equipment, and media based on continuous accumulated temperature can accurately characterize the temperature-driven mechanism changes in crop growth by constructing an annual temperature model and calculating effective accumulated temperature time-series data. By establishing a phenological analysis model based on continuous accumulated temperature, it can precisely depict the dynamic process of crop physiological development. By using trigonometric functions to fit daily temperature dynamic changes, it can overcome the limitations of relying on daily average temperature to calculate effective accumulated temperature, reconstruct the complete trajectory of daily temperature changes, and capture the cumulative effect of temperature fluctuations on crop growth. This allows the quantification of effective accumulated temperature to better align with the physiological response patterns of crop growth. Furthermore, by constructing a multi-year collaborative phenological monitoring model system, it can significantly improve the accuracy and reliability of crop phenological stage identification. By establishing a phenological monitoring reference shape model, it can effectively eliminate the interference of abnormal fluctuations in single-year data on monitoring results, enabling standardized comparative analysis between data from different years. This improves the robustness of monitoring results, achieves precise capture of phenological changes, and provides stable and reliable full-cycle monitoring for crops, thus assisting in the adjustment of agricultural production management strategies. Attached Figure Description

[0040] To more clearly illustrate the technical solutions in the embodiments or related technologies of this application, the accompanying drawings used in the description of the embodiments or related technologies will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0041] Figure 1 A flowchart illustrating a crop phenology monitoring method based on continuous accumulated temperature driven by one embodiment of this application;

[0042] Figure 2 A flowchart illustrating an effective accumulated temperature calculation algorithm provided in one embodiment of this application;

[0043] Figure 3 This is a schematic diagram of a crop phenology monitoring system based on continuous accumulated temperature driven by an embodiment of this application. Detailed Implementation

[0044] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0045] In one exemplary embodiment of this application, such as Figure 1 As shown, a crop phenology monitoring method based on continuous accumulated temperature is provided. This embodiment illustrates the application of this method to a phenology monitoring terminal. It is understood that this method can also be applied to a phenology monitoring server, and further to a phenology monitoring system including both a phenology monitoring terminal and a phenology monitoring server, and is implemented through the interaction between the two. In this embodiment, the method includes the following steps:

[0046] Step 101: Obtain the target year crop optical monitoring remote sensing data and the reference year crop optical remote sensing data, and calculate the target year vegetation index time series data and the reference year vegetation index time series data.

[0047] Specifically, the phenological monitoring terminal can acquire optical remote sensing data of crops for the target year and optical remote sensing data of crops for multiple reference years through a satellite remote sensing data receiving module or a satellite remote sensing database. The terminal can perform preprocessing operations such as cloud masking, atmospheric correction, and geometric correction on the acquired optical remote sensing data of crops for the target year and reference years to remove interference factors. Based on the band data of each light wave in the optical remote sensing data of crops for the target year and reference years, the terminal can calculate the vegetation index time series data of the target year and the vegetation index time series data of the reference years, and combine this with the corresponding time series moments to construct the target year vegetation index time series data and the vegetation index time series data of the reference years.

[0048] Optionally, the phenological monitoring terminal can calculate the vegetation index time series data of the target year and the vegetation index in the vegetation index time series data of each reference year based on the near-infrared band data and red band data of the target year crop optical remote sensing data and the crop optical remote sensing data of each reference year.

[0049] For example, the target year's crop optical monitoring remote sensing data and the crop optical remote sensing data for multiple reference years can be remote sensing data obtained based on MOD09GQ (Collection 6, title: h28v06, 250m) and MOD09GA (Collection 6, title: h28v06, 500m). The phenological monitoring terminal can generate a cloud mask based on MOD09GA to remove MOD09GQ images of the study area that are covered by clouds. The phenological monitoring terminal can generate a mask file using the quality assessment bands in MOD09GA, and then resample it to the same resolution as MOD09GQ to filter cloud-covered MOD09GQ products for the study area. The phenological monitoring terminal can calculate the vegetation index (NDVI) value based on band 1 (red band) and band 2 (near-red band) in MOD09GQ. In addition, the phenological monitoring terminal can obtain ground-observed phenological and temperature data from local meteorological units to verify the accuracy of phenological estimation.

[0050] Step 102: Obtain hourly resolution temperature monitoring data for each year, construct temperature models for each year based on the hourly resolution temperature monitoring data for each year, and calculate the effective accumulated temperature time series data for each year based on the temperature models for each year and the preset growth baseline temperature.

[0051] Optionally, the annual effective accumulated temperature time series data may include the reference annual effective accumulated temperature time series data corresponding to the reference annual vegetation index time series data and the target annual effective accumulated temperature time series data corresponding to the target annual vegetation index time series data.

[0052] Optionally, the phenological monitoring terminal can construct an annual temperature model based on trigonometric functions.

[0053] Optionally, the phenological monitoring terminal can perform frequency domain analysis preprocessing on annual hourly resolution temperature monitoring data to obtain frequency domain adaptive enhanced temperature monitoring data. Frequency domain analysis preprocessing may include, but is not limited to, linear trend removal preprocessing and data boundary preprocessing. Based on preset annual and daily periodic frequencies, the phenological monitoring terminal can extract the annual periodic spectrum and daily periodic spectrum of the temperature monitoring data from the frequency domain adaptive enhanced temperature monitoring data. The phenological monitoring terminal can calculate the power spectral density values ​​of the annual periodic spectrum and the daily periodic spectrum of the temperature monitoring data at their respective annual periodic frequencies. The phenological monitoring terminal can filter the annual periodic characteristic frequencies in descending order of annual periodic power spectral density to obtain the annual period enhanced spectrum. The phenological monitoring terminal can select the daily periodic characteristic frequencies in descending order of daily periodic power spectral density to obtain the daily period enhanced spectrum. The phenological monitoring terminal can perform inverse frequency domain transformation on the annual and daily period enhanced spectra and construct an annual temperature model based on time-domain trigonometric functions. The phenological monitoring terminal can obtain the annual effective accumulated temperature data at each effective accumulated temperature sampling time by integrating the annual temperature model and the crop growth baseline temperature, and then construct a continuous and complete annual effective accumulated temperature time series data.

[0054] Step 103: After performing time-series registration of the vegetation index time-series data and the effective accumulated temperature time-series data for the reference year, a phenological monitoring reference shape model is constructed based on the vegetation index time-series data and the effective accumulated temperature time-series data for each reference year, and a standard phenological period is set based on the phenological monitoring reference shape model.

[0055] Specifically, the phenological monitoring terminal can perform time-series registration of vegetation index time-series data and corresponding effective accumulated temperature time-series data for each reference year. The terminal can then use the registered vegetation index time-series data as the dependent variable and the corresponding effective accumulated temperature time-series data as the independent variable to construct an effective accumulated temperature model for each reference year. The terminal can then average this model to construct a phenological monitoring reference shape model. Within this model, the terminal can identify characteristic nodes corresponding to key crop growth stages and set standard phenological periods based on these nodes.

[0056] Step 104: After performing time-series registration of the target year's vegetation index time-series data and the target year's effective accumulated temperature time-series data, a target phenological monitoring effective accumulated temperature model is constructed based on the target year's vegetation index time-series data and the target year's effective accumulated temperature time-series data.

[0057] Specifically, the phenological monitoring terminal can perform time-series registration of the target year's vegetation index time-series data and the target year's effective accumulated temperature time-series data. The phenological monitoring terminal can then use the time-series registered target year's vegetation index time-series data as the dependent variable and the time-series registered target year's effective accumulated temperature time-series data as the independent variable to construct a target phenological monitoring effective accumulated temperature model.

[0058] Step 105: Based on the phenological monitoring reference shape model and the target phenological monitoring effective accumulated temperature model, and combined with the standard phenological period, generate the phenological monitoring results for the target year.

[0059] Indicatively, plant phenology refers to a series of seasonal changes in growth and development that are adapted to the climate, such as budding, branching, leaf unfolding, flowering, fruiting, leaf fall, and dormancy. The dynamic period of vegetation that is adapted to this is called the phenological period.

[0060] Optionally, the phenological monitoring terminal can match a phenological monitoring reference shape model and a target phenological monitoring effective accumulated temperature model based on a preset fitting error minimization target, and calculate the matching error minimization parameter that minimizes the fitting error between the two through algorithmic iteration. Based on the obtained matching error minimization parameter, the phenological monitoring terminal can calibrate and optimize the target phenological monitoring effective accumulated temperature model to generate a target phenological monitoring effective accumulated temperature calibration model. Based on the matching error minimization parameter and a preset standard phenological period, the phenological monitoring terminal can calculate the phenological period of the target year. Finally, the phenological monitoring terminal can integrate the target phenological monitoring effective accumulated temperature calibration model and the target year phenological period to generate the phenological monitoring results for the target year.

[0061] The aforementioned crop phenological monitoring method based on continuous accumulated temperature integrates crop optical monitoring remote sensing data and hourly resolution temperature monitoring data to achieve comprehensive capture and precise fusion of vegetation growth information and temperature-driven factors, thereby improving the accuracy and adaptability of the phenological monitoring model. By using annual hourly resolution temperature monitoring data, accurate quantification of effective accumulated temperature can be achieved, avoiding quantification deviations caused by intraday temperature differences. This ensures that effective accumulated temperature accurately reflects the contribution of temperature to crop growth, providing shape models with independent variables that better align with the physiological laws of crop growth, thus improving the accuracy of the shape models. Furthermore, by analyzing time-series vegetation index data and effective accumulated temperature data from multiple reference years... By performing time-series registration, a reference shape model for phenological monitoring can be constructed, and standard phenological periods can be set. This enables accurate analysis of typical crop growth trajectories and temperature response patterns, and establishes a reference benchmark for phenological monitoring. This enhances the model's adaptability and stability to environmental changes in different years, and improves the accuracy of phenological monitoring for the target year. By constructing an effective accumulated temperature model for phenological monitoring, combined with the reference shape model and standard phenological periods, the phenological periods of crops in the target year can be extracted efficiently. This reduces the interference of local abnormal observations and interannual environmental fluctuations on phenological monitoring, and improves the accuracy and reliability of phenological period extraction. In turn, this provides technical support for key agricultural activities such as agricultural production planning, timing of pest and disease control, and crop yield estimation.

[0062] In an optional embodiment of this application, please refer to Figure 1 and Figure 2 Step S102 involves acquiring hourly-level resolution temperature monitoring data for each year, constructing annual temperature models based on the annual hourly-level resolution temperature monitoring data, and calculating the effective accumulated temperature time series data for each year based on the annual temperature models and the preset growth baseline temperature. This may include:

[0063] Step S201: Perform linear trend removal preprocessing and data boundary preprocessing on the annual hourly resolution temperature monitoring data to obtain frequency domain adaptive enhanced temperature monitoring data.

[0064] Optionally, the phenological monitoring terminal can perform linear trend removal preprocessing on the acquired annual hourly-resolution temperature monitoring data to eliminate the interference of multi-year cycle temperature change trends on the analysis of intra-year temperature fluctuations.

[0065] Optionally, the phenological monitoring terminal can simultaneously perform data boundary preprocessing on the temperature data to ensure the continuity and integrity of the temperature data.

[0066] Step S202: Based on the preset annual cycle frequency and the preset daily cycle frequency, extract the annual cycle spectrum and daily cycle spectrum of the temperature monitoring data from the frequency domain adaptive enhanced temperature monitoring data.

[0067] Optionally, the expressions for the intra-year cycle frequency and the intra-day cycle frequency can be: ; ; In the formula, For the first Annual cycle frequency, The annual base period is used as the reference period. For the first Daily cycle frequency, This is the daily baseline period.

[0068] Optional, annual base period It can operate for 365 days, and the phenological monitoring terminal can also be set according to the growth cycle of different plants.

[0069] For example, taking the extraction of early rice phenological stages as an example, the phenological monitoring terminal can extract the annual baseline period based on the growth cycle of early rice. Set to 85 days.

[0070] Optional, daily base period It can be for 1 day.

[0071] Optional, and It can be a positive integer.

[0072] For example, The value can be less than or equal to the annual base period. (At this time, the annual base period) Positive integers (in days), if necessary. It can also be a value greater than the annual base period. (At this time, the annual base period) Positive integers (in days) are allowed, and no restrictions are imposed here.

[0073] Optional, The value can be less than or equal to the daily reference period. (At this time, the daily base period) Positive integers (in hours), if necessary. It can also be a value greater than the daily base period. (At this time, the daily base period) Positive integers (in hours) are allowed, and no restrictions are imposed here.

[0074] Step S203: Calculate the power spectral density value of the annual periodic spectrum of the temperature monitoring data at each annual periodic frequency to obtain the annual periodic power spectral density corresponding to each annual periodic frequency, and calculate the power spectral density value of the daily periodic spectrum of the temperature monitoring data at each daily periodic frequency to obtain the daily periodic power spectral density corresponding to each daily periodic frequency.

[0075] To illustrate, power spectral density (PSD) is an important concept in signal processing and frequency domain analysis, describing the power distribution of a signal in the frequency domain. PSD represents the power distribution of a signal across its frequency components; it can be expressed as the power density of a signal at a specific frequency, i.e., the power per unit frequency.

[0076] Step S204: Select the annual periodic characteristic frequency from each annual periodic frequency in descending order of annual periodic power spectral density until the ratio of the sum of the annual periodic power spectral densities of each annual periodic characteristic frequency to the total annual power spectral density exceeds the preset annual characteristic frequency energy ratio threshold. Optimize the annual periodic spectrum of the temperature monitoring data based on the annual periodic characteristic frequency to obtain the annual periodic enhanced spectrum. The total annual power spectral density is the sum of the power spectral densities of all annual periods.

[0077] Step S205: Select intra-day periodic characteristic frequencies from each intra-day periodic frequency in descending order of intra-day periodic power spectral density, until the ratio of the sum of the intra-day periodic power spectral densities of each intra-day periodic characteristic frequency to the total power spectral density of the same year exceeds the preset annual characteristic frequency energy ratio threshold. Optimize the intra-day periodic spectrum of temperature monitoring data based on the intra-day periodic characteristic frequencies to obtain the intra-day periodic enhanced spectrum. The total intra-day power spectral density is the sum of the power spectral densities of all intra-day periods.

[0078] Step S206: Perform frequency domain inverse transform on the intra-annual periodic enhancement spectrum and the intra-day periodic enhancement spectrum to construct the annual temperature model.

[0079] Step S207: Based on the annual temperature model and combined with the preset growth baseline temperature, calculate the annual effective accumulated temperature data at each effective accumulated temperature sampling time, and construct the annual effective accumulated temperature time series data based on the annual effective accumulated temperature data at the effective accumulated temperature sampling time.

[0080] In the aforementioned crop phenology monitoring method driven by continuous accumulated temperature, linear trend removal and data boundary preprocessing of annual hourly-resolution temperature monitoring data can improve the frequency domain adaptability of temperature data. Furthermore, by extracting the annual and daily periodic spectra of temperature monitoring data and calculating the corresponding power spectral density, core periodic characteristic frequencies are selected according to the power spectral density ratio threshold to optimize the spectrum. The annual temperature model constructed through frequency domain inverse transformation can more accurately restore the periodic variation characteristics of temperature. By combining the annual effective accumulated temperature time series data with the preset growth baseline temperature, the cumulative contribution of temperature fluctuations to crop growth can be captured more accurately, providing high-precision effective accumulated temperature data support for crop phenology monitoring.

[0081] In an optional embodiment of this application, the expression for the annual temperature model can be: ; ; In the formula, For annual temperature models, This represents the total number of annual periodic characteristic frequencies. This represents the total number of intraday periodic characteristic frequencies. For the serial number The amplitude of the annual periodic characteristic frequency. For the serial number The annual periodic characteristic frequency, For the serial number The phase of the annual periodic characteristic frequency. For the serial number The amplitude of the intraday periodic characteristic frequency. For the serial number Intraday cyclical characteristic frequency, For the serial number The phase of the intraday periodic characteristic frequency. For the serial number Effective accumulated temperature sampling time, For the serial number The annual effective accumulated temperature time series data corresponding to the effective accumulated temperature sampling time. The effective accumulated temperature begins at the starting time. For annual temperature models, This is the growth base point temperature.

[0082] In an optional embodiment of this application, the target year's phenological monitoring results may include a target phenological monitoring effective accumulated temperature calibration model and a target year's phenological period. Based on the phenological monitoring reference shape model and the target phenological monitoring effective accumulated temperature model, combined with a standard phenological period, the target year's phenological monitoring results are generated, which may include:

[0083] Specifically, the phenological monitoring terminal can calculate the matching error minimization parameters by matching the phenological monitoring reference shape model and the target phenological monitoring effective accumulated temperature model based on the goal of minimizing the fitting error.

[0084] Specifically, the phenological monitoring terminal can calibrate the effective accumulated temperature model for target phenological monitoring based on the parameters for minimizing matching error, and generate a calibration model for the effective accumulated temperature of target phenological monitoring.

[0085] Specifically, the phenological monitoring terminal can calculate the phenological period of the target year based on the matching error minimization parameter and the standard phenological period.

[0086] Specifically, the phenological monitoring terminal can generate the phenological monitoring results for the target year based on the effective accumulated temperature calibration model for the target phenological monitoring and the phenological period of the target year.

[0087] In an optional embodiment of this application, the matching error minimization parameters may include horizontal scaling parameters, vertical scaling parameters, and translation parameters. The expressions for the fitting error minimization target and the target year's phenological monitoring results can be: ; ; In the formula, To minimize the fitting error, , and These are the horizontal scaling parameters, vertical scaling parameters, and translation parameters, respectively. The total number of effective accumulated temperature pairing points. The serial number of the effective accumulated temperature model for target phenological monitoring is The effective accumulated temperature value at the paired point. The serial number of the effective accumulated temperature model for target phenological monitoring is The vegetation index of the effective accumulated temperature paired with the point, As a reference shape model for phenological monitoring, For the target year's phenological period, This is the standard phenological period.

[0088] In an optional embodiment of this application, the expression for the vegetation index in the reference annual vegetation index time series data and the target annual vegetation index time series data can be: ; In the formula, The vegetation index, It can be used for near-infrared band data from optical remote sensing data of crops in the target year and optical remote sensing data of crops in each reference year. It can provide remote sensing data of crop optical monitoring for the target year and red band data of crop optical remote sensing data for each reference year.

[0089] In an optional embodiment of this application, after time-series registration of the vegetation index time-series data and the effective accumulated temperature time-series data for the reference year, a phenological monitoring reference shape model is constructed based on the vegetation index time-series data and the effective accumulated temperature time-series data for each reference year, and a standard phenological period is set based on the phenological monitoring reference shape model, which may include:

[0090] Specifically, the phenological monitoring terminal can perform time-series registration of the vegetation index time-series data and the effective accumulated temperature time-series data of the reference year, and then construct the effective accumulated temperature model for each reference year by using the vegetation index time-series data of each reference year as the dependent variable and the effective accumulated temperature time-series data of each reference year corresponding to the vegetation index time-series data of each reference year as the independent variable.

[0091] Specifically, the phenological monitoring terminal can average the effective accumulated temperature model of phenological monitoring for each reference year to construct a phenological monitoring reference shape model.

[0092] Specifically, the phenological monitoring terminal can identify the standard phenological period in the phenological monitoring reference shape model.

[0093] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0094] Based on the same inventive concept, this application also provides a crop phenology monitoring system based on continuous accumulated temperature (CELT) for implementing the above-mentioned crop phenology monitoring method based on CELT. The solution provided by this system is similar to the solution described in the above method. Therefore, the specific limitations of one or more embodiments of the crop phenology monitoring system based on CELT provided below can be found in the limitations of the crop phenology monitoring method based on CELT described above, and will not be repeated here.

[0095] In one exemplary embodiment, such as Figure 3 As shown, a crop phenology monitoring system 300 based on continuous accumulated temperature is provided, comprising:

[0096] The remote sensing monitoring data processing module 301 can be used to acquire remote sensing data of crop optical monitoring in the target year and crop optical remote sensing data in each reference year, and calculate the time series data of vegetation index in the target year and vegetation index in each reference year.

[0097] The effective accumulated temperature data generation module 302 can be used to acquire hourly resolution temperature monitoring data for each year, construct annual temperature models based on the annual hourly resolution temperature monitoring data, and calculate the annual effective accumulated temperature time series data based on the annual temperature models and a preset growth baseline temperature. The annual effective accumulated temperature time series data includes reference annual effective accumulated temperature time series data corresponding to the reference annual vegetation index time series data and target annual effective accumulated temperature time series data corresponding to the target annual vegetation index time series data.

[0098] The reference shape model construction module 303 can be used to perform time-series registration of the vegetation index time-series data and the effective accumulated temperature time-series data of the reference year, and then construct a phenological monitoring reference shape model based on the vegetation index time-series data and the effective accumulated temperature time-series data of each reference year, and set a standard phenological period based on the phenological monitoring reference shape model.

[0099] The target monitoring model construction module 304 can be used to perform time-series registration of the target year vegetation index time-series data and the target year effective accumulated temperature time-series data, and then construct the target phenological monitoring effective accumulated temperature model based on the target year vegetation index time-series data and the target year effective accumulated temperature time-series data.

[0100] The phenological monitoring result generation module 305 can be used to generate phenological monitoring results for a target year based on the phenological monitoring reference shape model and the target phenological monitoring effective accumulated temperature model, combined with the standard phenological period.

[0101] In an optional embodiment of this application, the effective accumulated temperature data generation module 302 can also be used for:

[0102] Linear trend removal preprocessing and data boundary preprocessing are performed on annual hourly resolution temperature monitoring data to obtain frequency domain adaptive enhanced temperature monitoring data.

[0103] Based on the preset annual cycle frequency and the preset daily cycle frequency, the annual cycle spectrum and daily cycle spectrum of the temperature monitoring data are extracted from the frequency domain adaptive temperature monitoring data, respectively.

[0104] Calculate the power spectral density of the annual periodic spectrum of temperature monitoring data at each annual periodic frequency to obtain the annual periodic power spectral density corresponding to each annual periodic frequency. Also calculate the power spectral density of the daily periodic spectrum of temperature monitoring data at each daily periodic frequency to obtain the daily periodic power spectral density corresponding to each daily periodic frequency.

[0105] The annual periodic characteristic frequencies are selected from the annual periodic frequencies in descending order of annual periodic power spectral density until the ratio of the sum of the annual periodic characteristic frequencies and the total annual power spectral density exceeds the preset annual characteristic frequency energy ratio threshold. Based on the annual periodic characteristic frequencies, the annual periodic spectrum of temperature monitoring data is optimized to obtain the annual periodic enhanced spectrum. The total annual power spectral density is the sum of the annual periodic power spectral densities.

[0106] Intra-day characteristic frequencies are selected from all intra-day periodic frequencies in descending order of intra-day periodic power spectral density, until the sum of the intra-day periodic power spectral densities of all intra-day periodic characteristic frequencies exceeds the ratio of the total power spectral density of the same year to a preset threshold for the energy proportion of characteristic frequencies within the year. Based on the intra-day periodic characteristic frequencies, the intra-day periodic spectrum of temperature monitoring data is optimized to obtain the intra-day periodic enhanced spectrum. The total intra-day power spectral density is the sum of the power spectral densities of all intra-day periods.

[0107] An annual temperature model is constructed by performing inverse frequency domain transformation on the intra-annual and intra-day periodic enhancement spectra.

[0108] Based on the annual temperature model and combined with the preset growth baseline temperature, the annual effective accumulated temperature data at each effective accumulated temperature sampling time is calculated, and the annual effective accumulated temperature time series data is constructed based on the annual effective accumulated temperature data at each effective accumulated temperature sampling time.

[0109] In an optional embodiment of this application, the phenological monitoring result generation module 305 may also be used for:

[0110] Based on the objective of minimizing the fitting error, the matching error minimization parameters are calculated by matching the reference shape model for phenological monitoring and the effective accumulated temperature model for target phenological monitoring.

[0111] The effective accumulated temperature model for target phenological monitoring is calibrated based on the parameters that minimize matching error, thus generating a calibration model for the effective accumulated temperature model for target phenological monitoring.

[0112] The phenological period of the target year is calculated based on the matching error minimization parameter and the standard phenological period.

[0113] Based on the target phenological monitoring effective accumulated temperature calibration model and the target annual phenological period, the target annual phenological monitoring results are generated.

[0114] In an optional embodiment of this application, the reference shape model construction module 303 can also be used for:

[0115] After time-series registration of the vegetation index time-series data and the effective accumulated temperature time-series data for the reference years, the effective accumulated temperature model for phenological monitoring for each reference year is constructed with the vegetation index time-series data for each reference year as the dependent variable and the effective accumulated temperature time-series data for each reference year corresponding to the vegetation index time-series data for each reference year as the independent variable.

[0116] The effective accumulated temperature models for phenological monitoring in each reference year were averaged to construct a reference shape model for phenological monitoring.

[0117] Standard phenological periods were identified in the phenological monitoring reference shape model.

[0118] In one embodiment, a computer device is provided, including a memory and a processor, the memory storing a computer program, the processor executing the computer program to implement the steps of a crop phenology monitoring method based on continuous accumulated temperature as described above.

[0119] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the steps in the above method embodiments.

[0120] For the device embodiments, since they basically correspond to the method embodiments, the relevant parts can be referred to in the description of the method embodiments. The device embodiments described above are merely illustrative. The components described as separate parts may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this disclosure according to actual needs. Those skilled in the art can understand and implement this without creative effort.

[0121] The above-described embodiments are merely illustrative of several implementation methods of the embodiments of this application, and their descriptions are relatively specific and detailed. However, they should not be construed as limiting the scope of the patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the embodiments of this application, and these modifications and improvements all fall within the protection scope of the embodiments of this application.

Claims

1. A crop phenology monitoring method based on continuous accumulated temperature driving, characterized in that, The method comprises: obtaining target annual crop optical monitoring remote sensing data and each reference annual crop optical monitoring remote sensing data, and calculating target annual vegetation index time series data and each reference annual vegetation index time series data; obtaining each annual hour-level resolution temperature monitoring data, constructing each annual temperature model based on each annual hour-level resolution temperature monitoring data, and calculating each annual effective accumulated temperature time series data based on each annual temperature model and a preset growth base point temperature; wherein the annual effective accumulated temperature time series data comprises reference annual effective accumulated temperature time series data corresponding to the reference annual vegetation index time series data and target annual effective accumulated temperature time series data corresponding to the target annual vegetation index time series data; after time series registration of the reference annual vegetation index time series data and the reference annual effective accumulated temperature time series data, constructing a phenological monitoring reference shape model based on each reference annual vegetation index time series data and each reference annual effective accumulated temperature time series data, and setting a standard phenological period based on the phenological monitoring reference shape model; after time series registration of the target annual vegetation index time series data and the target annual effective accumulated temperature time series data, constructing a target phenological monitoring effective accumulated temperature model based on the target annual vegetation index time series data and the target annual effective accumulated temperature time series data; generating a target annual phenological monitoring result based on the phenological monitoring reference shape model and the target phenological monitoring effective accumulated temperature model and in combination with the standard phenological period.

2. The method of claim 1, wherein, The obtaining each annual hour-level resolution temperature monitoring data, constructing each annual temperature model based on each annual hour-level resolution temperature monitoring data, and calculating each annual effective accumulated temperature time series data based on each annual temperature model and a preset growth base point temperature comprises: performing linear trend removal preprocessing and data boundary preprocessing on the annual hour-level resolution temperature monitoring data to obtain frequency domain adaptive enhanced temperature monitoring data; extracting temperature monitoring data annual periodic frequency spectrum and temperature monitoring data daily periodic frequency spectrum from the frequency domain adaptive enhanced temperature monitoring data based on preset annual periodic frequencies and preset daily periodic frequencies; calculating power spectral density values of the temperature monitoring data annual periodic frequency spectrum on each annual periodic frequency to obtain annual periodic power spectral densities corresponding to each annual periodic frequency, and calculating the power spectral density values of the temperature monitoring data daily periodic frequency spectrum on each daily periodic frequency to obtain daily periodic power spectral densities corresponding to each daily periodic frequency; selecting annual periodic characteristic frequencies from each annual periodic frequency in descending order of the annual periodic power spectral densities, until a ratio of a sum of the annual periodic power spectral densities of each annual periodic characteristic frequency to an annual total power spectral density exceeds a preset annual characteristic frequency energy proportion threshold, and optimizing the temperature monitoring data annual periodic frequency spectrum based on the annual periodic characteristic frequencies to obtain annual periodic enhanced frequency spectrum, wherein the annual total power spectral density is a sum of all the annual periodic power spectral densities. selecting, in the order from large to small of the intraday cycle power spectral densities, an intraday cycle characteristic frequency from each of the intraday cycle frequencies until a ratio of a sum of the intraday cycle power spectral densities of each of the intraday cycle characteristic frequencies to an intraday total power spectral density exceeds a preset intrayear characteristic frequency energy proportion threshold, and optimizing, based on the intraday cycle characteristic frequency, the intraday cycle frequency spectrum of the temperature monitoring data to obtain an intraday cycle enhanced frequency spectrum, the intraday total power spectral density being a sum of all the intraday cycle power spectral densities; performing frequency domain inverse transformation on the intrayear cycle enhanced frequency spectrum and the intraday cycle enhanced frequency spectrum to construct the annual temperature model; based on the annual temperature model, combining a preset growth base point temperature, and calculating annual effective accumulated temperature data of each effective accumulated temperature sampling time to construct annual effective accumulated temperature time series data according to the annual effective accumulated temperature data of the effective accumulated temperature sampling time.

3. The method of claim 2, wherein, The expression of the annual temperature model is: ； ； In the formula, is the annual temperature model, is the total number of intra-annual periodic characteristic frequencies of the intra-annual periodic characteristic frequencies, is the total number of intra-daily periodic characteristic frequencies of the intra-daily periodic characteristic frequencies, is the intra-annual periodic characteristic frequency amplitude of the intra-annual periodic characteristic frequency with the serial number is the intra-annual periodic characteristic frequency of the intra-annual periodic characteristic frequency with the serial number is the intra-annual periodic characteristic frequency phase of the intra-annual periodic characteristic frequency with the serial number is the intra-daily periodic characteristic frequency amplitude of the intra-daily periodic characteristic frequency with the serial number is the intra-daily periodic characteristic frequency of the intra-daily periodic characteristic frequency with the serial number is the intra-daily periodic characteristic frequency phase of the intra-daily periodic characteristic frequency with the serial number is the effective accumulated temperature sampling time with the serial number is the annual effective accumulated temperature time series data corresponding to the effective accumulated temperature sampling time with the serial number is the effective accumulated temperature accumulation starting time, is the annual temperature model, is the growth base temperature.​​​​​​​​ 4. The method of claim 1, wherein, The target annual phenology monitoring result includes a target phenology monitoring effective accumulated temperature calibration model and a target annual phenological phase, and the target annual phenology monitoring result is generated based on the phenology monitoring reference shape model and the target phenology monitoring effective accumulated temperature model in combination with the standard phenological phase, including: matching the phenology monitoring reference shape model and the target phenology monitoring effective accumulated temperature model based on a fitting error minimization target to calculate a matching error minimization parameter; calibrating the target phenology monitoring effective accumulated temperature model based on the matching error minimization parameter to generate the target phenology monitoring effective accumulated temperature calibration model; calculating the target annual phenological phase according to the matching error minimization parameter and the standard phenological phase; generating the target annual phenology monitoring result according to the target phenology monitoring effective accumulated temperature calibration model and the target annual phenological phase.

5. The method of claim 4, wherein, The matching error minimization parameter includes a horizontal scaling parameter, a vertical scaling parameter and a translation parameter, and the expression of the fitting error minimization target and the target annual phenology monitoring result is: ； ； wherein, is the fitting error minimization objective, , and are the horizontal scaling parameter, the vertical scaling parameter and the translation parameter, respectively, is the total number of effective accumulated temperature pairing points, is the effective accumulated temperature value of the effective accumulated temperature pairing point with the sequence number of in the target phenology monitoring effective accumulated temperature model, is the vegetation index of the effective accumulated temperature pairing point with the sequence number of in the target phenology monitoring effective accumulated temperature model, is the phenology monitoring reference shape model, is the target annual phenophase, is the standard phenophase.

6. The method of claim 5, wherein, The expression of the vegetation index in the reference annual vegetation index time series data and the target annual vegetation index time series data is: ； In the formula, is the vegetation index, is near-infrared band data, is red band data.

7. The method according to any one of claims 1 to 6, characterized in that, After the reference annual vegetation index time series data and the reference annual effective accumulated temperature time series data are time series registered, a phenology monitoring reference shape model is constructed based on each of the reference annual vegetation index time series data and each of the reference annual effective accumulated temperature time series data, and a standard phenological phase is set based on the phenology monitoring reference shape model, including: After the reference annual vegetation index time series data and the reference annual effective accumulated temperature time series data are time series registered, each reference annual phenology monitoring effective accumulated temperature model is constructed by taking each reference annual vegetation index time series data as a dependent variable and taking each reference annual effective accumulated temperature time series data corresponding to each reference annual vegetation index time series data as an independent variable; each reference annual phenology monitoring effective accumulated temperature model is averaged to construct the phenology monitoring reference shape model; the standard phenological phase is identified in the phenology monitoring reference shape model.

8. A crop phenology monitoring system based on continuous accumulated temperature driving, characterized in that, The system comprises: a remote sensing monitoring data processing module, configured to acquire target annual crop optical monitoring remote sensing data and each reference annual crop optical remote sensing data, and calculate target annual vegetation index time series data and each reference annual vegetation index time series data; an effective accumulated temperature data generation module, configured to acquire each annual hour-level resolution temperature monitoring data, construct each annual temperature model based on the annual hour-level resolution temperature monitoring data, and calculate each annual effective accumulated temperature time series data based on the annual temperature model and in combination with a preset growth base temperature; wherein the annual effective accumulated temperature time series data comprises reference annual effective accumulated temperature time series data corresponding to the reference annual vegetation index time series data and target annual effective accumulated temperature time series data corresponding to the target annual vegetation index time series data; a reference shape model construction module, configured to, after time series registration of the reference annual vegetation index time series data and reference annual effective accumulated temperature time series data, construct a phenological monitoring reference shape model based on each reference annual vegetation index time series data and each reference annual effective accumulated temperature time series data, and set a standard phenological period based on the phenological monitoring reference shape model; a target monitoring model construction module, configured to, after time series registration of the target annual vegetation index time series data and the target annual effective accumulated temperature time series data, construct a target phenological monitoring effective accumulated temperature model based on the target annual vegetation index time series data and the target annual effective accumulated temperature time series data; a phenological monitoring result generation module, configured to generate a target annual phenological monitoring result based on the phenological monitoring reference shape model and the target phenological monitoring effective accumulated temperature model in combination with the standard phenological period. 9.A computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the computer device is configured to perform the method according to any one of claims 1-8. The processor executes the computer program to implement the method in any one of claims 1 to 7.

10. A computer-readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to implement the method in any one of claims 1 to 7.

Images