Crop functional chlorophyll fluorescence inversion method, system and equipment and readable medium

Abstract

The invention provides a crop functional chlorophyll fluorescence inversion method, system and device and a readable medium, and relates to the technical field of ecological environment monitoring, and the method comprises the steps: obtaining a multi-source data cube; acquiring a crop type distribution diagram based on the multi-source data cube; constructing a sun-induced chlorophyll fluorescence SIF inversion basic model according to the crop type distribution diagram; and outputting a crop solar-induced chlorophyll fluorescence time sequence product according to the solar-induced chlorophyll fluorescence SIF inversion basic model. According to the crop functional chlorophyll fluorescence inversion method provided by the invention, the multi-source data cube is used as input, and two special processing links which are completely independent in physical mechanism and dynamic correction are constructed; high-precision and high-physical-reliability SIF time sequence product production for crops can be realized, and a reliable data basis is provided for precise monitoring of photosynthesis of an agricultural ecosystem.

Description

Technical Field

[0001] This application relates to the field of ecological environment monitoring technology, and in particular to a method, system, equipment and readable medium for crop functional chlorophyll fluorescence inversion. Background Technology

[0002] Frontiers in vegetation remote sensing focus on the temporal inversion and product generation of solar-induced chlorophyll fluorescence (SIF). SIF is a natural probe of plant photosynthetic centers; its intensity directly reflects the actual photochemical reaction status of photosystem II and is hailed as a "direct measure of vegetation photosynthesis." Compared to traditional vegetation indices, SIF has unique and mechanistic advantages in monitoring vegetation stress and quantifying primary productivity. In precision agriculture and global change ecology research, obtaining high-precision, high-spatiotemporal-resolution SIF time-series products is crucial for revealing crop growth dynamics, assessing the impact of environmental stress, and optimizing carbon cycle models. South China, as an important grain and cash crop producing region in my country, has arable land landscapes primarily composed of rice and sugarcane. These two crops belong to the C3 and C4 functional types in photosynthetic pathways, respectively. This fundamental difference leads to significant divergences in their light energy utilization efficiency, carbon fixation methods, and response strategies to environmental factors (such as light intensity, temperature, and moisture). C4 plants (sugarcane), through CO2 concentration mechanisms, typically exhibit higher photosynthetic and water use efficiency under high temperature and strong light conditions. Their response relationship between SIF signal and photosynthetically active radiation (PAR) also differs significantly from that of C3 plants (rice). Furthermore, their phenological cycles are heavily shaped by intensive human management: rice exhibits a typical early-late rice "bimodal" cycle, with the SIF signal nearly zero at harvest, forming a distinct "on / off" pattern; while sugarcane has a long, multi-year growth cycle, and its SIF time-series curve shows a single-peak, wide-valley shape. Therefore, accurately capturing and analyzing the SIF time-series dynamics of these two representative crops is crucial for accurately assessing the photosynthetic carbon sequestration capacity of regional agricultural ecosystems.

[0003] Currently, the generation of remote sensing SIF time-series products mainly relies on two mainstream technical approaches, both of which have significant limitations. At the SIF inversion level, the most similar implementation schemes are based on physical inversion algorithms of top-atmosphere radiation (such as spectral line fitting methods based on the O2-A absorption band, represented by TROPOMI and OCO-2 satellite products) or statistical / semi-empirical models based on surface reflectance (such as data-driven neural network methods). These schemes typically aim to construct a universal inversion framework applicable to all vegetation types. Their core flaw lies in the implicit assumption of "homogeneous vegetation function" in model construction, failing to distinguish and embed the intrinsic differences between C3 and C4 plants in core physiological processes such as fluorescence quantum yield and energy dissipation pathways. As a result, while the SIF values ​​retrieved by the universal model can reflect general seasonal trends, they cannot accurately characterize the fluorescence saturation characteristics of C4 crops under high light or the degree of fluorescence quenching in C3 crops under photorespiration, causing a systematic bias in the estimation of their true photosynthetic capacity. In the subsequent processing and interpretation of SIF time-series products, existing approximation schemes typically involve directly smoothing and filtering the SIF time-series data obtained from the aforementioned general model inversion (using methods such as the SG filter) to remove noise, or directly applying it to phenological analysis. While this approach is simple, its processing completely ignores the specific physical meaning inherent in the SIF signal, driven by both crop functionalities and human management. A typical dilemma is that a uniform filtering algorithm might oversmooth the sharp drop in SIF signal caused by rice harvesting, or misjudge the slow physiological decline of sugarcane as data noise, thus obscuring the crop's unique growth rhythms and management imprints. A deeper problem is that due to the model mismatch at the inversion source, any subsequent attempts to extract physiological stress information or estimate photosynthetic "light energy use efficiency" from the SIF time series become physically very weak. The inherent limitations of existing technical solutions in terms of the universality of inversion mechanisms and the indiscriminate nature of time-series analysis severely restrict the depth and accuracy of SIF remote sensing applications in heterogeneous agricultural landscapes, especially in C3 / C4 mixed cropping areas like South China.

[0004] Current remote sensing-based large-area chlorophyll fluorescence (SIF) time-series inversion generally uses uniform physical models or statistical relationships, seriously neglecting the essential differences between different crops in their core photosynthetic mechanisms and human management practices. Especially in regions like South China characterized by intensive agriculture, major crops such as rice (C3) and sugarcane (C4) exhibit significant differences in intrinsic physiological processes such as energy allocation in photosystems I and II, photoinhibition response, and water use efficiency. Simultaneously, their phenological cycles are strongly driven by human activities such as the "double-cropping rice" system and long-term sugarcane management. This dual heterogeneity, both functional and managerial, results in distinctly different characteristics in the amplitude, phase, and variation patterns of their SIF time-series signals. However, existing generalized SIF inversion models cannot effectively characterize this differentiation of intrinsic driving mechanisms. Directly applying a uniform model not only introduces systematic biases (such as underestimating the fluorescence yield of C4 crops under high light or confusing the fluorescence quenching caused by photorespiration in C3 crops), but also obscures its unique "switching" signals at key phenological nodes (such as the sudden drop in rice harvest and the slow decay of sugarcane). This results in unclear physical meaning and limited accuracy of the final SIF time series products, making it difficult to realize their due value in applications such as early diagnosis of crop stress, accurate assessment of photosynthetic productivity, and assimilation of carbon cycle models. Summary of the Invention

[0005] Therefore, it is necessary to provide a crop functional chlorophyll fluorescence inversion method to address the shortcomings of existing technologies, aiming to solve the problems of unclear physical meaning and limited accuracy of existing SIF time series products.

[0006] To solve the above problems, this application adopts the following technical solution: One of the objectives of this application is to provide a method for retrieving the fluorescence of functional chlorophyll in crops, including: Acquire multi-source data cubes; A crop type distribution map is obtained based on the multi-source data cube; A basic model for solar-induced chlorophyll fluorescence (SIF) inversion was constructed based on the crop type distribution map. The solar-induced chlorophyll fluorescence (SIF) inversion basic model is used to output the crop solar-induced chlorophyll fluorescence time series product.

[0007] In some embodiments, the step of acquiring a multi-source data cube specifically includes the following steps: Acquire multi-source remote sensing data, raw observation data of solar-induced chlorophyll fluorescence (SIF), photosynthetically active radiation (PAR), leaf area index (LAI), and vegetation cover (FVC) time-series inversion data; resample all data and project them onto a unified spatial coordinate system and spatial resolution to generate a temporally continuous and spatially matched multi-source data cube.

[0008] In some embodiments, the step of obtaining a crop type distribution map based on the multi-source data cube specifically includes the following steps: A support vector machine classifier is trained using feature data. The trained support vector machine classifier is then used to predict each pixel to obtain a preliminary crop classification map. The feature data includes the classification features of the multi-source data cube, vegetation indices, and phenological parameters. The vegetation indices include NDVI and EVI, and the phenological parameters include the start, peak, and end of the growing season. Based on the crop distribution map, phenological phenomena are extracted from the crops, and key phenological nodes are extracted using the vegetation index time series curve to generate a crop type distribution map. The phenological nodes include the early growth stage, rapid growth stage, maturity stage, and harvest stage.

[0009] In some embodiments, derivative analysis is used to process the smoothed vegetation index time series curve, and the phenological nodes are defined as inflection points based on the vegetation index time series curve.

[0010] In some embodiments, the step of constructing a basic model for solar-induced chlorophyll fluorescence (SIF) inversion based on the crop type distribution map specifically includes the following steps: The basic model for solar-induced chlorophyll fluorescence (SIF) inversion is expressed as follows:

[0011] in, The basic SIF estimate represents the fluorescence intensity at a specific wavelength. Incident light intensity, one of the prior data, can be obtained from the multi-source data cube; leaf area index (LAI) and vegetation cover (FVC) are prior data, which can be determined from the multi-source data cube. Environmental stress factors, including temperature and moisture stress, can be quantified using thermal infrared bands and precipitation data; This is the model error term, which follows a normal distribution.

[0012] In some embodiments, when the crop is rice C3, the basic model for solar-induced chlorophyll fluorescence (SIF) inversion is as follows:

[0013]

[0014] in, For maximum fluorescence efficiency, determine through ground-based experiments or literature values; recommended value: 0.01–0.05. Light saturation coefficient (unit: m²·s·μmol) -¹), reflecting the photoinhibition effect, obtained from fitting the fluorescence response curve, recommended value range: 0.001–0.005 m²·s·μmol - ¹; This is the temperature stress coefficient, calculated based on the crop's optimal temperature range. Recommended value: 0.05–0.15 ℃ - ¹; This represents the absolute value of the temperature deviation from the optimal value. This is actual data; When the crop is sugarcane C4, its basic model for solar-induced chlorophyll fluorescence (SIF) inversion is as follows:

[0015]

[0016] in, This is the fluorescence yield coefficient, with typical values ​​higher than those for rice crops; recommended value: 0.02–0.08. The water stress index; This is one of the a priori data points for actual evaporation. This is one of the prior data for potential evapotranspiration.

[0017] In some embodiments, the method further includes a step of correcting the basic model for solar-induced chlorophyll fluorescence (SIF) retrieval: The corrected basic model for solar-induced chlorophyll fluorescence (SIF) inversion is as follows:

[0018]

[0019] in, For the final SIF product; The phenological driving strategy function is used to quantify the SIF variation characteristics of crop growth stages. This is a functional driving strategy function to capture the differences in photosynthetic physiology between C3 and C4. The residual correction term is optimized using data assimilation methods. K Here is the Kalman gain matrix. For the observed values, This is an estimated value; The phenological cycle used to simulate the "on / off" pattern in rice and sugarcane is defined as a piecewise function:

[0020] Where t represents the time in days; Early stage of growth; This is the rapid growth period; It is harvest season; (Recommended value: 0.6~0.8) (Recommended value: 0.8~1.2) (Recommended value 1) (Recommended value 0~0.3) is the weighting coefficient for phenological stages, determined by fitting historical SIF data; (Recommended duration: 15-25 days) (Recommended value: 20-30 days) (Recommended value 5~15 days) is the time scale parameter, controlling the smoothness of the function, and is calibrated based on crop growth rate.

[0021] To integrate the intrinsic differences between C3 / C4 plants, including photochemical efficiency and water molecule utilization efficiency, it is expressed as:

[0022]

[0023] in, The adjustment coefficient can be set based on crop type, C3 crop. C4 crops ; The fluorescence enhancement ratio of C4 plants relative to C3 plants; where The fluorescence quantum yield is prior data and can be obtained from the literature. Recommended value: 1.1~1.3.

[0024] In some embodiments, the step of outputting crop solar-induced chlorophyll fluorescence time-series products based on the solar-induced chlorophyll fluorescence (SIF) inversion baseline model specifically includes the following steps: The solar-induced chlorophyll fluorescence (SIF) output from the basic model for SIF inversion is summarized in time series to generate a spatially continuous time series SIF dataset. At the same time, SIF values ​​of pixels are extracted separately for rice and sugarcane areas to form crop-specific SIF parameter sets.

[0025] In some embodiments, the method further includes a step of validating the crop solar-induced chlorophyll fluorescence time-series product output from the solar-induced chlorophyll fluorescence (SIF) inversion baseline model, specifically including: Spatial validation was performed by comparing the ground-based measured SIF data that were not involved in model construction and assimilation with the corresponding locations and times output by the solar-induced chlorophyll fluorescence SIF inversion basic model. The regional average SIF time series curve output by the solar-induced chlorophyll fluorescence (SIF) inversion basic model was compared with the time series curves of other independent satellite SIF products to verify the consistency of its temporal dynamics.

[0026] The second objective of this application is to provide a crop functional chlorophyll fluorescence inversion system, comprising: The data acquisition unit is used to acquire multi-source data cubes; The processing unit is used to obtain a crop type distribution map based on the multi-source data cube; The model building unit is used to construct a basic model for solar-induced chlorophyll fluorescence (SIF) inversion based on the crop type distribution map. The output unit is used to output crop solar-induced chlorophyll fluorescence time series products based on the solar-induced chlorophyll fluorescence (SIF) inversion basic model.

[0027] A third objective of this application is to provide an electronic device, comprising: One or more processors; Memory, used to store one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the method.

[0028] The fourth objective of this application is to provide a computer-readable medium on which a computer program is stored, characterized in that the computer program, when executed by a processor, implements any one of the methods described above.

[0029] The present application adopts the above technical solution, and its beneficial effects are as follows: The crop functional chlorophyll fluorescence inversion method provided in this application acquires a multi-source data cube; obtains a crop type distribution map based on the multi-source data cube; constructs a basic model for solar-induced chlorophyll fluorescence (SIF) inversion based on the crop type distribution map; and outputs a crop solar-induced chlorophyll fluorescence time series product based on the solar-induced chlorophyll fluorescence (SIF) inversion basic model. The crop functional chlorophyll fluorescence inversion method provided in this application uses a multi-source data cube as input and constructs two specialized processing links that are completely independent in terms of physical mechanism and dynamic correction. It can realize the production of high-precision and high-physical-reliability SIF time series products for crops, providing a reliable data foundation for the accurate monitoring of photosynthesis in agricultural ecosystems. Attached Figure Description

[0030] To more clearly illustrate the technical solutions of the embodiments of this application, the drawings used in the description of the embodiments of this application or the prior art will be briefly introduced below. Obviously, the 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.

[0031] Figure 1 A flowchart illustrating the steps of the crop functional chlorophyll fluorescence inversion method provided in this application embodiment.

[0032] Figure 2 This is a schematic diagram of the structure of the crop functional chlorophyll fluorescence inversion system provided in the embodiments of this application.

[0033] Figure 3 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0034] The embodiments of this application are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain this application, and should not be construed as limiting this application. In the description of this application, it should be understood that the terms "upper", "lower", "horizontal", "inner", "outer", etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings, and are only for the convenience of describing this application 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 this application. Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this application, "multiple" means two or more, unless otherwise explicitly specified.

[0035] The core flaw in existing technologies lies in their adherence to a "homogeneous mechanism assumption" paradigm: both SIF inversion modeling and time-series signal analysis employ a uniform, undifferentiated approach, ignoring the fundamental differences between C3 and C4 crops in their core photosynthetic mechanisms and phenological driving mechanisms. Specifically, the inversion model is disconnected from photosynthetic physiology: Common SIF inversion models (whether based on physical methods of atmospheric radiative transfer or statistical learning models based on surface reflectance) are built upon the inherent assumption that vegetation fluorescence responses follow the same physiological laws. This renders the models completely incapable of characterizing the fundamental differences between C3 and C4 plants in energy allocation at photochemical reaction centers, light inhibition thresholds, and responses to water stress. As a result, the SIF time-series products obtained through inversion have inherent physical deficiencies—systematically underestimating the fluorescence yield potential of C4 crops (such as sugarcane) under high light conditions, or overestimating the actual fluorescence efficiency of C3 crops (such as rice) under photorespiration, severely weakening the reliability of SIF as a quantitative relationship of "direct probe of photosynthesis." There is also a dynamic mismatch between time-series processing and phenological management: After obtaining the initial SIF inversion values, existing schemes typically employ uniform time-series filtering and reconstruction methods (such as SG filtering with fixed parameters) to attempt to smooth noise and extract trends. This method completely ignores the highly specific phenological "switching" signals shaped by strong human management. For example, for the periodic sharp drop in rice harvest and the slow, multi-year decay of sugarcane, uniform filtering either over-smooths these real physical abrupt changes or retains them as inexplicable "noise," thus obscuring the most critical start and end signals in the crop growth cycle, making the final product unusable for accurate phenological event identification and growth stage division.

[0036] To address the fundamental limitations imposed by the aforementioned "homogeneous mechanism assumption," this invention aims to establish a "full-link differentiated" SIF product production system, encompassing both mechanism inversion and dynamic correction. The core objective of this invention is not only post-hoc temporal correction but, more importantly, to construct a differentiated mechanism model highly adapted to crop photosynthetic functionalities at the inversion source. Specifically, for rice (C3) and sugarcane (C4), dedicated inversion links are established, embedding their unique light response curves (e.g., the near-linear relationship between C3's non-rectangular hyperbola and C4's) and key stress factors (e.g., C3's sensitivity to photorespiration and C4's sensitivity to water stress), ensuring the physiological accuracy and type-specificity of initial SIF values ​​from a physical foundation. Furthermore, a further objective of this invention is to design highly coupled differentiated dynamic correction strategies for these two types of SIF sequences, which have clear physical meanings but vastly different temporal morphologies. This is achieved by introducing a phenological-driven strategy function. With functional driving strategy function This invention employs physical-guided real-time and dynamic correction to the baseline inversion values. The former aims to accurately characterize and enhance the unique phenological patterns of rice's "double-peak, double-switch" and sugarcane's "single-peak, long-period" phenological patterns; the latter is used to quantitatively correct the intrinsic fluorescence efficiency advantage of C4 plants relative to C3 plants. Ultimately, by achieving an innovative coupling of "differentiated mechanism inversion" and "differentiated dynamic correction," this invention produces highly reliable SIF time-series products that not only accurately reflect the photosynthetic physiological state of crops but also clearly analyze their unique growth and management rhythms. This provides a comprehensive, specialized technical solution from raw data to advanced products for precision agricultural management, early warning of stress, and regional carbon sequestration capacity assessment, completely solving the core problem of "incompatibility" of general models in heterogeneous agricultural landscape applications.

[0037] 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.

[0038] Example 1 Please see Figure 1 This invention provides a method for inverting the fluorescence of functional chlorophyll in crops, including the following steps S10 to S40. The specific implementation schemes of each step are described in detail below.

[0039] Step S10: Obtain the multi-source data cube.

[0040] In this embodiment, the step of acquiring the multi-source data cube specifically includes the following steps: Acquire multi-source remote sensing data, raw observation data of solar-induced chlorophyll fluorescence (SIF), photosynthetically active radiation (PAR), leaf area index (LAI), and vegetation cover (FVC) time-series inversion data; resample all data and project them onto a unified spatial coordinate system and spatial resolution to generate a temporally continuous and spatially matched multi-source data cube.

[0041] Specifically, multi-source remote sensing data: Sentinel-2 MSI and Landsat-8 OLI Level-1C or Level-2A products for specified areas and time periods are downloaded from the European Space Agency (ESA) and the United States Geological Survey (USGS) for radiometric calibration and atmospheric correction.

[0042] Specifically, raw SIF observation data: SIF observation data from OCO-2 / 3, TROPOMI, or GOSAT satellite products (such as SIF@740nm or SIF@757nm), or ground stations (such as FluxNet), are acquired as training, validation, and assimilation data.

[0043] Specifically, PAR data: Photosynthetically active radiation data at daily or monthly scales are obtained from MODIS (MOD / MYD16) or ECMWF ERA5 reanalysis datasets.

[0044] Specifically, LAI / FVC time series data: using physical models such as PROSAIL or machine learning methods such as random forest, high spatiotemporal resolution LAI and FVC time series products are generated by inverting reflectance data based on Sentinel-2 / Landsat-8.

[0045] Furthermore, environmental stress data may also be included: temperature data obtained from MODIS surface temperature products, and precipitation data obtained from TRMM, GPM, or reanalysis data, for calculating temperature stress and moisture stress indices.

[0046] Furthermore, all data is resampled and projected onto a unified spatial coordinate system (such as WGS 84 UTM) and spatial resolution (such as 10 meters or 30 meters). This generates temporally continuous and spatially matched multi-source data cubes to fill in data gaps caused by cloud contamination (e.g., using time series interpolation).

[0047] Step S20: Obtain a crop type distribution map based on the multi-source data cube.

[0048] In this embodiment, the step of obtaining a crop type distribution map based on the multi-source data cube specifically includes the following steps: A support vector machine (SVM) classifier is trained using feature data. The trained SVM classifier then predicts each pixel to obtain a preliminary crop classification map. The feature data includes the classification features of the multi-source data cube, vegetation indices, and phenological parameters. The vegetation indices include NDVI and EVI, and the phenological parameters include the start, peak, and end of the growing season. Phenological data is extracted from the crop distribution map. Key phenological nodes are extracted using the vegetation index time-series curves to generate a crop type distribution map. The phenological nodes include the early growth stage, rapid growth stage, maturity stage, and harvest stage.

[0049] In this embodiment, the time-series curves of vegetation indices such as NDVI and EVI for each pixel during the growing season are calculated from the preprocessed multi-source data cube. Temporal decomposition (e.g., double logistic function fitting) or derivative methods are used to extract phenological parameters from the NDVI time-series curves, such as the start date, peak date, end date, and growth rate of the growing season. Multispectral reflectance (multiple bands), vegetation index time series, extracted phenological parameters, and possible texture features are combined to form a high-dimensional feature vector. Using C3 (rice) / C4 (sugarcane) training samples obtained from field sampling or historical map patches, a support vector machine (SVM) classifier is trained to classify the study area pixel by pixel, generating a C3 / C4 crop type distribution map.

[0050] In this embodiment, phenological data are extracted from the crop distribution map obtained in the previous step for rice and sugarcane regions, respectively. The smoothed NDVI time-series curves are processed using derivative analysis: the first derivative (rate of change) is used to identify the start and end points of growth / decline; the second derivative (acceleration of change) is used to identify inflection points (such as the transition from accelerated growth to decelerated growth). Based on the C3 / C4 phenological node division rules given in the scheme, the specific dates for the initial growth stage, rapid growth stage, maturity stage, and harvest stage are determined for each pixel, generating a "crop phenological calendar".

[0051] Rice phenological node division rules: (1) In the early stage of rice growth: the first derivative of NDVI changes from negative or close to zero to a continuous positive value, and the second derivative shows a positive maximum value (the point of acceleration of change). (2) During the rapid growth period of rice: the first derivative of NDVI remains high and positive, and a maximum value may occur during this period; (3) Rice maturity period: NDVI reaches its annual maximum or plateau period; (4) Rice harvesting period: NDVI drops sharply in a short period of time.

[0052] Sugarcane phenological node division rules: (1) In the early stage of sugarcane growth: the first derivative of NDVI changes from a low value to a continuous positive value; (2) During the rapid growth period of sugarcane: the first derivative of NDVI remains at a moderately positive value for a long time, and multiple small peaks may appear instead of a single sharp peak; (3) Sugarcane maturity period: NDVI reaches its annual maximum or plateau period; (4) Sugarcane harvesting period: NDVI drops sharply in a short period of time.

[0053] Step S30: Construct a basic model for solar-induced chlorophyll fluorescence (SIF) inversion based on the crop type distribution map.

[0054] In this embodiment, the basic model for constructing the solar-induced chlorophyll fluorescence (SIF) inversion based on the crop type distribution map is expressed as follows:

[0055] in, Basic SIF estimates (unit: mW·m - ²·sr - ¹·nm - ¹), representing the fluorescence intensity at a specific wavelength (760 nm); (Photosynthetically active radiation, unit: μmol·m) - ²·s - ¹) Incident light intensity is one of the prior data, which can be obtained from remote sensing data or meteorological data; Leaf area index (LAI) and vegetation cover (FVC) are prior data, which can be determined from remote sensing data or ground measurements. These are environmental stress factors, including temperature and moisture stress, which can be quantified using thermal infrared bands and precipitation data. This is the model error term, assumed to follow a normal distribution.

[0056] Furthermore, when the crop is rice, the rice (C3) inversion model is constructed as follows: Rice photosynthesis is significantly affected by photorespiration, and the relationship between SIF and PAR exhibits a non-linear saturation. A light suppression term is added to the model:

[0057]

[0058] in, Maximum fluorescence efficiency (unit: mW·m) - ²·sr - ¹·nm - ¹ / (μmol·m - ²·s - ¹)), determined through ground experiments or literature values, recommended value: 0.01–0.05; Light saturation coefficient (unit: m²·s·μmol) - ¹), reflecting the photoinhibition effect, obtained from fitting the fluorescence response curve, recommended value range: 0.001–0.005 m²·s·μmol - ¹; This is the temperature stress coefficient, calculated based on the crop's optimal temperature range. Recommended value: 0.05–0.15 ℃ - ¹; This represents the absolute value of the temperature deviation from the optimal value. This is actual data.

[0059] Furthermore, when the crop is sugarcane, the sugarcane (C4) inversion model is constructed as follows: Sugarcane has high photosynthetic efficiency and a high light saturation point. Its photosynthetic efficiency (SIF) and light saturation point (PAR) are approximately linearly related, and it is more sensitive to water stress. The model uses a combination of linear terms and a water factor.

[0060]

[0061] in, Fluorescence yield coefficient (unit: mW·m) - ²·sr - ¹·nm - ¹ / (μmol·m - ²·s - ¹), with typical values ​​higher than those for C3 crops; recommended values: 0.02–0.08. The water stress index; This is one of the a priori data points for actual evaporation. This is one of the prior data for potential evapotranspiration.

[0062] It is understandable that this embodiment achieves a paradigm shift from "apparent inversion" to "mechanism embedding," fundamentally improving the physiological reliability and quantitative accuracy of SIF products. Existing technologies treat SIF as a universal physical quantity for inversion, completely ignoring the essential differences in the core mechanisms of photosynthesis between C3 and C4 plants. This invention pioneers a differentiated inversion model that highly aligns with the physiological pathways of crop photosynthetic functions, constructing dedicated nonlinear models for the photorespiration and photoinhibition characteristics of C3 crops (rice) and the high light efficiency and water response characteristics of C4 crops (sugarcane). This solves the systematic bias problem caused by the singular physiological mechanism assumptions of general models from the source of inversion, enabling the initial SIF estimate to truly reflect the photosynthetic physiological state of different crops. This lays a solid physiological foundation for transforming SIF into a reliable photosynthetic probe, overcoming the fundamental defect of "fuzzy source mechanism."

[0063] Furthermore, it also includes a step of correcting the basic model for solar-induced chlorophyll fluorescence (SIF) inversion.

[0064] In this embodiment, to improve the physical meaning and accuracy of the model, the present invention introduces a strategy function, which adjusts the model's parameters during runtime. Real-time correction is performed. The strategy function dynamically adjusts the SIF estimate based on crop phenological management and functional type differences. The corrected SIF output is:

[0065]

[0066] in, For the final SIF product; The phenological driving strategy function is used to quantify the SIF variation characteristics of crop growth stages. This is a functional driving strategy function to capture the differences in photosynthetic physiology between C3 and C4. The residual correction term is optimized using data assimilation methods. K Here is the Kalman gain matrix. For the observed values, This is an estimated value.

[0067] Used to simulate the "on-off" phenological cycle of rice and sugarcane. Defined as a piecewise function:

[0068] Where t is time (day); Early stage of growth; This is the rapid growth period; It is harvest season; (Recommended value: 0.6~0.8) (Recommended value: 0.8~1.2) (Recommended value 1) (Recommended value 0~0.3) is the weighting coefficient for phenological stages, determined by fitting historical SIF data; (Recommended duration: 15-25 days) (Recommended value: 20-30 days) (Recommended value 5~15 days) is the time scale parameter, controlling the smoothness of the function, and is calibrated based on crop growth rate.

[0069] This is used to integrate the intrinsic differences between C3 and C4 plants, including photochemical efficiency and water molecule utilization efficiency. Expressed as:

[0070]

[0071] in, The adjustment coefficient can be set based on crop type, C3 crop. C4 crops ; The fluorescence enhancement ratio of C4 plants relative to C3 plants; where The fluorescence quantum yield is prior data and can be obtained from the literature. Recommended value: 1.1~1.3.

[0072] It is understandable that this embodiment constructs a specialized processing link with a closed loop of "mechanism inversion - dynamic correction," achieving accurate extraction and enhancement of crop-specific phenological and management signals. Existing "unified inversion + unified filtering" schemes, when processing highly heterogeneous agricultural landscapes, can obscure physical mechanisms and blur management signals. This invention integrates the concept of differentiation throughout the entire link from inversion to product generation: first, an initial SIF sequence with clear physiological significance is generated through a specialized model; then, a dynamic correction strategy is "tailor-made" for it, introducing a dual strategy function driven by phenology and function. This closed-loop design, which is coordinated before and after and guided by physical meaning, not only corrects random errors but also actively enhances the true phenological signals of rice's "double-peak, double-switch" and sugarcane's "single-peak, long-period" patterns. Simultaneously, it quantifies the intrinsic physiological advantages of C4 plants, thereby maximizing the preservation of effective photosynthetic information and systematically suppressing non-physiological noise in complex farmland environments.

[0073] It should be noted that this embodiment uses a phenological-driven strategy function. The construction and application of [the technology], especially its piecewise function form as a quantification of the unique phenological "switch" mode of crops, and its use through phenological stage weighting coefficients ([…]). ) and time scale parameters ( Methods for precise control; functional driving strategy functions The construction and application of [the technology], especially its adjustment coefficient. Compared with fluorescence enhancement ratio To quantify the mechanism of intrinsic physiological dominance in C4 plants; to utilize data assimilation techniques (such as ensemble Kalman filtering) to analyze the residual terms. The purpose of the optimized calibration steps is to ensure that the final output SIF time series product has both high numerical accuracy and clear crop physiological and phenological management implications.

[0074] Step S40: Output crop solar-induced chlorophyll fluorescence time series products based on the solar-induced chlorophyll fluorescence (SIF) inversion basic model.

[0075] In this embodiment, the step of outputting the crop solar-induced chlorophyll fluorescence time series product based on the solar-induced chlorophyll fluorescence (SIF) inversion basic model specifically includes the following steps: The solar-induced chlorophyll fluorescence (SIF) output from the basic model for SIF inversion is summarized in time series to generate a spatially continuous time series SIF dataset. At the same time, SIF values ​​of pixels are extracted separately for rice and sugarcane areas to form crop-specific SIF parameter sets.

[0076] Specifically, in this embodiment, the SIF_final output from the above steps is summarized according to time series (e.g., monthly) to generate a spatially continuous GeoTIFF raster file. At the same time, SIF values ​​of pixels are extracted separately for rice and sugarcane areas to form crop-specific SIF parameter sets (time series curves and statistical features).

[0077] Furthermore, it also includes a step of validating the crop solar-induced chlorophyll fluorescence time-series product output from the solar-induced chlorophyll fluorescence (SIF) inversion basic model, specifically including: Spatial validation was performed by comparing the corresponding locations and times output by the solar-induced chlorophyll fluorescence SIF inversion basic model with ground-based measured SIF data that were not involved in model construction and assimilation. The regional average SIF time series curve output by the solar-induced chlorophyll fluorescence SIF inversion basic model was then compared with the time series curves of other independent satellite SIF products to verify the consistency of its temporal dynamics.

[0078] Specifically, this embodiment uses ground-based measured SIF data (such as point data measured by a portable fluorometer) that were not involved in model construction and assimilation, and compares them with the corresponding location and time SIF_final output by the model to calculate indicators such as the coefficient of determination R² and the root mean square error RMSE.

[0079] Specifically, in this embodiment, the regional average SIF time series curve generated by the model is compared with the time series curves of other independent satellite SIF products (such as TROPOMI SIF, but usually with lower resolution) to verify the consistency of their temporal dynamics.

[0080] It is understood that the crop functional chlorophyll fluorescence inversion method provided in this embodiment takes multi-source remote sensing data (including surface reflectance, photosynthetically active radiation PAR and prior parameters LAI and FVC) as input, and constructs two dedicated processing links that are completely independent in terms of physical mechanism and dynamic correction. This enables the production of high-precision and high-physical-reliability SIF time-series products for rice (C3) and sugarcane (C4) respectively, providing a reliable data foundation for the accurate monitoring of photosynthesis in agricultural ecosystems.

[0081] Specifically, in the differentiated mechanism inversion layer, the general SIF inversion model is completely abandoned, and dedicated nonlinear models highly matched with the intrinsic characteristics of the photosynthetic physiological pathways of C3 and C4 plants are constructed respectively. For rice (C3), which has a low light saturation point and is significantly affected by photorespiration, a non-rectangular hyperbolic model including a light inhibition term is used for inversion to accurately characterize the saturation decline of its fluorescence yield under strong light. For sugarcane (C4), which has high photosynthetic efficiency and a high light saturation point, a near-linear model sensitive to water stress is used for inversion to enhance the model's characterization of its sustained high-efficiency photosynthetic capacity under high light conditions. This step incorporates the intrinsic physiological differences of crops from the source of inversion, ensuring the physical accuracy of the initial SIF estimates. After obtaining the type-specific initial SIF time series, this invention enters the differentiated dynamic correction layer, which designs and couples physically meaningful strategy functions for real-time correction, targeting the unique time series dynamics shaped by both human management and physiological characteristics.

[0082] Through the innovative coupling of the two core components, "differentiated mechanism inversion" and "differentiated dynamic correction," this invention ultimately outputs a highly reliable SIF time-series product that clearly characterizes the differences in photosynthetic function between C3 and C4 crops and accurately reflects their unique phenological and management rhythms. This method not only realizes a paradigm shift in SIF remote sensing from "general estimation" to "specialized analysis," but also provides a full-chain, specialized technical solution—from physical basis to advanced products—for photosynthetic dynamic monitoring, early stress diagnosis, and carbon sink assessment of heterogeneous agricultural landscapes by deeply integrating crop physiological mechanisms with remote sensing inversion.

[0083] Furthermore, the crop functional chlorophyll fluorescence inversion method provided in this application greatly enhances the practical application value and decision support capability of SIF time-series products in precision agricultural management and ecosystem photosynthetic monitoring. Because the SIF time-series products produced by this invention possess both high physiological accuracy and distinct phenological specificity, their value in practical applications far exceeds that of existing general-purpose products. For example, the rice SIF time-series processed by this invention can clearly indicate key phenological nodes of tillering, heading, maturity, and harvesting, and can diagnose photosynthetic stress caused by high temperature or drought at an early stage; the sugarcane SIF time-series can accurately reflect its long vegetative growth stage and dynamic changes in water use efficiency. This enables SIF data to directly serve crop yield prediction, irrigation optimization decisions, and accurate assessment of C3 / C4 crop photosynthetic carbon sinks at the regional scale, providing unprecedented high-quality, highly interpretable data support for smart agriculture and global change ecology research.

[0084] In summary, this invention achieves a technological leap from "general data products" to "specialized physiological products" by deeply integrating crop photosynthetic physiological mechanisms into the entire process of SIF remote sensing inversion and product generation. Its core advantages are reflected in the deepening of the product's physiological connotation, the leap in information fidelity, and the fundamental enhancement of its practical value in agricultural and ecological applications.

[0085] Example 2 Please see Figure 2 This application provides a crop functional chlorophyll fluorescence inversion system, comprising: Data acquisition unit 10 is used to acquire multi-source data cubes.

[0086] In this embodiment, the step of acquiring the multi-source data cube specifically includes the following steps: Acquire multi-source remote sensing data, raw observation data of solar-induced chlorophyll fluorescence (SIF), photosynthetically active radiation (PAR), leaf area index (LAI), and vegetation cover (FVC) time-series inversion data; resample all data and project them onto a unified spatial coordinate system and spatial resolution to generate a temporally continuous and spatially matched multi-source data cube.

[0087] Specifically, multi-source remote sensing data: Sentinel-2 MSI and Landsat-8 OLI Level-1C or Level-2A products for specified areas and time periods are downloaded from the European Space Agency (ESA) and the United States Geological Survey (USGS) for radiometric calibration and atmospheric correction.

[0088] Specifically, raw SIF observation data: SIF observation data from OCO-2 / 3, TROPOMI, or GOSAT satellite products (such as SIF@740nm or SIF@757nm), or ground stations (such as FluxNet), are acquired as training, validation, and assimilation data.

[0089] Specifically, PAR data: Photosynthetically active radiation data at daily or monthly scales are obtained from MODIS (MOD / MYD16) or ECMWF ERA5 reanalysis datasets.

[0090] Specifically, LAI / FVC time series data: using physical models such as PROSAIL or machine learning methods such as random forest, high spatiotemporal resolution LAI and FVC time series products are generated by inverting reflectance data based on Sentinel-2 / Landsat-8.

[0091] Furthermore, environmental stress data may also be included: temperature data obtained from MODIS surface temperature products, and precipitation data obtained from TRMM, GPM, or reanalysis data, for calculating temperature stress and moisture stress indices.

[0092] Furthermore, all data is resampled and projected onto a unified spatial coordinate system (such as WGS 84 UTM) and spatial resolution (such as 10 meters or 30 meters). This generates temporally continuous and spatially matched multi-source data cubes to fill in data gaps caused by cloud contamination (e.g., using time series interpolation).

[0093] The processing unit is used to obtain a crop type distribution map based on the multi-source data cube.

[0094] In this embodiment, the step of obtaining a crop type distribution map based on the multi-source data cube specifically includes the following steps: A support vector machine (SVM) classifier is trained using feature data. The trained SVM classifier then predicts each pixel to obtain a preliminary crop classification map. The feature data includes the classification features of the multi-source data cube, vegetation indices, and phenological parameters. The vegetation indices include NDVI and EVI, and the phenological parameters include the start, peak, and end of the growing season. Phenological data is extracted from the crop distribution map. Key phenological nodes are extracted using the vegetation index time-series curves to generate a crop type distribution map. The phenological nodes include the early growth stage, rapid growth stage, maturity stage, and harvest stage.

[0095] In this embodiment, the time-series curves of vegetation indices such as NDVI and EVI for each pixel during the growing season are calculated from the preprocessed multi-source data cube. Temporal decomposition (e.g., double logistic function fitting) or derivative methods are used to extract phenological parameters from the NDVI time-series curves, such as the start date, peak date, end date, and growth rate of the growing season. Multispectral reflectance (multiple bands), vegetation index time series, extracted phenological parameters, and possible texture features are combined to form a high-dimensional feature vector. Using C3 (rice) / C4 (sugarcane) training samples obtained from field sampling or historical map patches, a support vector machine (SVM) classifier is trained to classify the study area pixel by pixel, generating a C3 / C4 crop type distribution map.

[0096] In this embodiment, phenological data are extracted from the crop distribution map obtained in the previous step for rice and sugarcane regions, respectively. The smoothed NDVI time-series curves are processed using derivative analysis: the first derivative (rate of change) is used to identify the start and end points of growth / decline; the second derivative (acceleration of change) is used to identify inflection points (such as the transition from accelerated growth to decelerated growth). Based on the C3 / C4 phenological node division rules given in the scheme, the specific dates for the initial growth stage, rapid growth stage, maturity stage, and harvest stage are determined for each pixel, generating a "crop phenological calendar".

[0097] Rice phenological node division rules: (1) In the early stage of rice growth: the first derivative of NDVI changes from negative or close to zero to a continuous positive value, and the second derivative shows a positive maximum value (the point of acceleration of change). (2) During the rapid growth period of rice: the first derivative of NDVI remains high and positive, and a maximum value may occur during this period; (3) Rice maturity period: NDVI reaches its annual maximum or plateau period; (4) Rice harvesting period: NDVI drops sharply in a short period of time.

[0098] Sugarcane phenological node division rules: (1) In the early stage of sugarcane growth: the first derivative of NDVI changes from a low value to a continuous positive value; (2) During the rapid growth period of sugarcane: the first derivative of NDVI remains at a moderately positive value for a long time, and multiple small peaks may appear instead of a single sharp peak; (3) Sugarcane maturity period: NDVI reaches its annual maximum or plateau period; (4) Sugarcane harvesting period: NDVI drops sharply in a short period of time.

[0099] The model building unit is used to construct a basic model for solar-induced chlorophyll fluorescence (SIF) inversion based on the crop type distribution map.

[0100] In this embodiment, the basic model for constructing the solar-induced chlorophyll fluorescence (SIF) inversion based on the crop type distribution map is expressed as follows:

[0101] in, Basic SIF estimates (unit: mW·m - ²·sr - ¹·nm - ¹), representing the fluorescence intensity at a specific wavelength (760 nm); (Photosynthetically active radiation, unit: μmol·m) - ²·s - ¹) Incident light intensity is one of the prior data, which can be obtained from remote sensing data or meteorological data; Leaf area index (LAI) and vegetation cover (FVC) are prior data, which can be determined from remote sensing data or ground measurements. These are environmental stress factors, including temperature and moisture stress, which can be quantified using thermal infrared bands and precipitation data. This is the model error term, assumed to follow a normal distribution.

[0102] Furthermore, when the crop is rice, the rice (C3) inversion model is constructed as follows: Rice photosynthesis is significantly affected by photorespiration, and the relationship between SIF and PAR exhibits a non-linear saturation. A light suppression term is added to the model:

[0103]

[0104] in, Maximum fluorescence efficiency (unit: mW·m) - ²·sr - ¹·nm - ¹ / (μmol·m - ²·s - ¹)), determined through ground experiments or literature values, recommended value: 0.01–0.05; Light saturation coefficient (unit: m²·s·μmol) - ¹), reflecting the photoinhibition effect, obtained from fitting the fluorescence response curve, recommended value range: 0.001–0.005 m²·s·μmol - ¹; This is the temperature stress coefficient, calculated based on the crop's optimal temperature range. Recommended value: 0.05–0.15 ℃ - ¹; This represents the absolute value of the temperature deviation from the optimal value. This is actual data.

[0105] Furthermore, when the crop is sugarcane, the sugarcane (C4) inversion model is constructed as follows: Sugarcane has high photosynthetic efficiency and a high light saturation point. Its photosynthetic efficiency (SIF) and light saturation point (PAR) are approximately linearly related, and it is more sensitive to water stress. The model uses a combination of linear terms and a water factor.

[0106]

[0107] in, Fluorescence yield coefficient (unit: mW·m) - ²·sr - ¹·nm - ¹ / (μmol·m - ²·s - ¹), with typical values ​​higher than those for C3 crops; recommended values: 0.02–0.08. The water stress index; This is one of the a priori data points for actual evaporation. This is one of the prior data for potential evapotranspiration.

[0108] It is understandable that this embodiment achieves a paradigm shift from "apparent inversion" to "mechanism embedding," fundamentally improving the physiological reliability and quantitative accuracy of SIF products. Existing technologies treat SIF as a universal physical quantity for inversion, completely ignoring the essential differences in the core mechanisms of photosynthesis between C3 and C4 plants. This invention pioneers a differentiated inversion model that highly aligns with the physiological pathways of crop photosynthetic functions, constructing dedicated nonlinear models for the photorespiration and photoinhibition characteristics of C3 crops (rice) and the high light efficiency and water response characteristics of C4 crops (sugarcane). This solves the systematic bias problem caused by the singular physiological mechanism assumptions of general models from the source of inversion, enabling the initial SIF estimate to truly reflect the photosynthetic physiological state of different crops. This lays a solid physiological foundation for transforming SIF into a reliable photosynthetic probe, overcoming the fundamental defect of "fuzzy source mechanism."

[0109] Furthermore, it also includes a calibration module, which calibrates the basic model for solar-induced chlorophyll fluorescence (SIF) inversion.

[0110] In this embodiment, to improve the physical meaning and accuracy of the model, the present invention introduces a strategy function, which adjusts the model's parameters during runtime. Real-time correction is performed. The strategy function dynamically adjusts the SIF estimate based on crop phenological management and functional type differences. The corrected SIF output is:

[0111]

[0112] in, For the final SIF product; The phenological driving strategy function is used to quantify the SIF variation characteristics of crop growth stages. This is a functional driving strategy function to capture the differences in photosynthetic physiology between C3 and C4. The residual correction term is optimized using data assimilation methods. K Here is the Kalman gain matrix. For the observed values, This is an estimated value.

[0113] Used to simulate the "on-off" phenological cycle of rice and sugarcane. Defined as a piecewise function:

[0114] Where t is time (day); Early stage of growth; This is the rapid growth period; It is harvest season; (Recommended value: 0.6~0.8) (Recommended value: 0.8~1.2) (Recommended value 1) (Recommended value 0~0.3) is the weighting coefficient for phenological stages, determined by fitting historical SIF data; (Recommended duration: 15-25 days) (Recommended value: 20-30 days) (Recommended value 5~15 days) is the time scale parameter, controlling the smoothness of the function, and is calibrated based on crop growth rate.

[0115] This is used to integrate the intrinsic differences between C3 and C4 plants, including photochemical efficiency and water molecule utilization efficiency. Expressed as:

[0116]

[0117] in, The adjustment coefficient can be set based on crop type, C3 crop. C4 crops ; The fluorescence enhancement ratio of C4 plants relative to C3 plants; where The fluorescence quantum yield is prior data and can be obtained from the literature. Recommended value: 1.1~1.3.

[0118] It is understandable that this embodiment constructs a specialized processing link with a closed loop of "mechanism inversion - dynamic correction," achieving accurate extraction and enhancement of crop-specific phenological and management signals. Existing "unified inversion + unified filtering" schemes, when processing highly heterogeneous agricultural landscapes, can obscure physical mechanisms and blur management signals. This invention integrates the concept of differentiation throughout the entire link from inversion to product generation: first, an initial SIF sequence with clear physiological significance is generated through a specialized model; then, a dynamic correction strategy is "tailor-made" for it, introducing a dual strategy function driven by phenology and function. This closed-loop design, which is coordinated before and after and guided by physical meaning, not only corrects random errors but also actively enhances the true phenological signals of rice's "double-peak, double-switch" and sugarcane's "single-peak, long-period" patterns. Simultaneously, it quantifies the intrinsic physiological advantages of C4 plants, thereby maximizing the preservation of effective photosynthetic information and systematically suppressing non-physiological noise in complex farmland environments.

[0119] It should be noted that this embodiment uses a phenological-driven strategy function. The construction and application of [the technology], especially its piecewise function form as a quantification of the unique phenological "switch" mode of crops, and its use through phenological stage weighting coefficients ([…]). ) and time scale parameters ( Methods for precise control; functional driving strategy functions The construction and application of [the technology], especially its adjustment coefficient. Compared with fluorescence enhancement ratio To quantify the mechanism of intrinsic physiological dominance in C4 plants; to utilize data assimilation techniques (such as ensemble Kalman filtering) to analyze the residual terms. The purpose of the optimized calibration steps is to ensure that the final output SIF time series product has both high numerical accuracy and clear crop physiological and phenological management implications.

[0120] The output unit is used to output crop solar-induced chlorophyll fluorescence time series products based on the solar-induced chlorophyll fluorescence (SIF) inversion basic model.

[0121] In this embodiment, the step of outputting the crop solar-induced chlorophyll fluorescence time series product based on the solar-induced chlorophyll fluorescence (SIF) inversion basic model specifically includes the following steps: The solar-induced chlorophyll fluorescence (SIF) output from the basic model for SIF inversion is summarized in time series to generate a spatially continuous time series SIF dataset. At the same time, SIF values ​​of pixels are extracted separately for rice and sugarcane areas to form crop-specific SIF parameter sets.

[0122] Specifically, in this embodiment, the SIF_final output from the above steps is summarized according to time series (e.g., monthly) to generate a spatially continuous GeoTIFF raster file. At the same time, SIF values ​​of pixels are extracted separately for rice and sugarcane areas to form crop-specific SIF parameter sets (time series curves and statistical features).

[0123] Furthermore, it also includes a verification module for verifying the crop solar-induced chlorophyll fluorescence time-series product output by the solar-induced chlorophyll fluorescence (SIF) inversion basic model, specifically including: Spatial validation was performed by comparing the corresponding locations and times output by the solar-induced chlorophyll fluorescence SIF inversion basic model with ground-based measured SIF data that were not involved in model construction and assimilation. The regional average SIF time series curve output by the solar-induced chlorophyll fluorescence SIF inversion basic model was then compared with the time series curves of other independent satellite SIF products to verify the consistency of its temporal dynamics.

[0124] Specifically, this embodiment uses ground-based measured SIF data (such as point data measured by a portable fluorometer) that were not involved in model construction and assimilation, and compares them with the corresponding location and time SIF_final output by the model to calculate indicators such as the coefficient of determination R² and the root mean square error RMSE.

[0125] Specifically, in this embodiment, the regional average SIF time series curve generated by the model is compared with the time series curves of other independent satellite SIF products (such as TROPOMI SIF, but usually with lower resolution) to verify the consistency of their temporal dynamics.

[0126] The crop functional chlorophyll fluorescence inversion system provided in this embodiment takes multi-source remote sensing data (including surface reflectance, photosynthetically active radiation PAR, and prior parameters LAI and FVC) as input. By constructing two dedicated processing links that are completely independent in terms of physical mechanism and dynamic correction, it can realize the production of high-precision and high-physical-reliability SIF time-series products for rice (C3) and sugarcane (C4), respectively, providing a reliable data foundation for the accurate monitoring of photosynthesis in agricultural ecosystems.

[0127] Specifically, in the differentiated mechanism inversion layer, the general SIF inversion model is completely abandoned, and dedicated nonlinear models highly matched with the intrinsic characteristics of the photosynthetic physiological pathways of C3 and C4 plants are constructed respectively. For rice (C3), which has a low light saturation point and is significantly affected by photorespiration, a non-rectangular hyperbolic model including a light inhibition term is used for inversion to accurately characterize the saturation decline of its fluorescence yield under strong light. For sugarcane (C4), which has high photosynthetic efficiency and a high light saturation point, a near-linear model sensitive to water stress is used for inversion to enhance the model's characterization of its sustained high-efficiency photosynthetic capacity under high light conditions. This step incorporates the intrinsic physiological differences of crops from the source of inversion, ensuring the physical accuracy of the initial SIF estimates. After obtaining the type-specific initial SIF time series, this invention enters the differentiated dynamic correction layer, which designs and couples physically meaningful strategy functions for real-time correction, targeting the unique time series dynamics shaped by both human management and physiological characteristics.

[0128] Through the innovative coupling of the two core components, "differentiated mechanism inversion" and "differentiated dynamic correction," this invention ultimately outputs a highly reliable SIF time-series product that clearly characterizes the differences in photosynthetic function between C3 and C4 crops and accurately reflects their unique phenological and management rhythms. This method not only realizes a paradigm shift in SIF remote sensing from "general estimation" to "specialized analysis," but also provides a full-chain, specialized technical solution—from physical basis to advanced products—for photosynthetic dynamic monitoring, early stress diagnosis, and carbon sink assessment of heterogeneous agricultural landscapes by deeply integrating crop physiological mechanisms with remote sensing inversion.

[0129] Furthermore, the crop functional chlorophyll fluorescence inversion system provided in this application greatly enhances the practical application value and decision support capabilities of SIF time-series products in precision agricultural management and ecosystem photosynthetic monitoring. Because the SIF time-series products produced by this invention possess both high physiological accuracy and distinct phenological specificity, their value in practical applications far exceeds that of existing general-purpose products. For example, the rice SIF time-series processed by this invention can clearly indicate key phenological nodes of tillering, heading, maturity, and harvesting, and can diagnose photosynthetic stress caused by high temperature or drought at an early stage; the sugarcane SIF time-series can accurately reflect its long vegetative growth stage and the dynamic changes in water use efficiency. This enables SIF data to directly serve crop yield prediction, irrigation optimization decisions, and accurate assessment of C3 / C4 crop photosynthetic carbon sinks at the regional scale, providing unprecedented high-quality, highly interpretable data support for smart agriculture and global change ecology research.

[0130] In summary, this invention achieves a technological leap from "general data products" to "specialized physiological products" by deeply integrating crop photosynthetic physiological mechanisms into the entire process of SIF remote sensing inversion and product generation. Its core advantages are reflected in the deepening of the product's physiological connotation, the leap in information fidelity, and the fundamental enhancement of its practical value in agricultural and ecological applications.

[0131] Based on the same inventive concept, embodiments of the present invention also provide an electronic device. Figure 3 This is a structural block diagram of an electronic device provided in an embodiment of the present invention. Figure 3 As shown, an embodiment of the present invention provides an electronic device including: one or more processors 101, a memory 102, and one or more I / O interfaces 103. The memory 102 stores one or more programs, which, when executed by the one or more processors, cause the one or more processors to implement any of the crop functional chlorophyll fluorescence inversion methods described in the above embodiments; the one or more I / O interfaces 103 are connected between the processor and the memory, configured to enable information interaction between the processor and the memory.

[0132] The processor 101 is a device with data processing capabilities, including but not limited to a central processing unit (CPU); the memory 102 is a device with data storage capabilities, including but not limited to random access memory (RAM, more specifically SDRAM, DDR, etc.), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), and flash memory (FLASH); the I / O interface (read / write interface) 103 is connected between the processor 101 and the memory 102, and can realize information interaction between the processor 101 and the memory 102, including but not limited to a data bus (Bus).

[0133] In some embodiments, the processor 101, memory 102, and I / O interface 103 are interconnected via bus 104, and thus connected to other components of the computing device.

[0134] In some embodiments, the one or more processors 101 include a field-programmable gate array.

[0135] This invention also provides a computer-readable medium. The computer-readable medium stores a computer program, which, when executed by a processor, implements the steps in any of the crop functional chlorophyll fluorescence inversion methods described in the above embodiments. The computer-readable storage medium can be volatile or non-volatile.

[0136] This invention also provides a computer program product, including computer-readable code, or a non-volatile computer-readable storage medium carrying computer-readable code. When the computer-readable code is run in the processor of an electronic device, the processor in the electronic device executes the above-described crop functional chlorophyll fluorescence inversion method.

[0137] Those skilled in the art will understand that all or some of the steps, systems, and apparatuses disclosed above, and their functional modules / units, can be implemented as software, firmware, hardware, or suitable combinations thereof. In hardware implementations, the division between functional modules / units mentioned above does not necessarily correspond to the division of physical components; for example, a physical component may have multiple functions, or a function or step may be performed collaboratively by several physical components. Some or all physical components may be implemented as software executed by a processor, such as a central processing unit, digital signal processor, or microprocessor, or as hardware, or as an integrated circuit, such as an application-specific integrated circuit (ASIC). Such software can be distributed on a computer-readable storage medium, which may include computer storage media (or non-transitory media) and communication media (or transient media).

[0138] As is well known to those skilled in the art, the term computer storage medium includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information such as computer-readable program instructions, data structures, program modules or other data. Computer storage media include, but are not limited to, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM), static random access memory (SRAM), flash memory or other memory technologies, and portable compact disc read-only memory (CD). ROM, digital multifunction disc (DVD) or other optical disc storage, magnetic cartridges, magnetic tapes, disk storage or other magnetic storage devices, or any other medium that can be used to store desired information and can be accessed by a computer. Furthermore, it is well known to those skilled in the art that communication media typically contain computer-readable program instructions, data structures, program modules, or other data in modulated data signals such as carrier waves or other transmission mechanisms, and may include any information delivery medium.

[0139] The computer-readable program instructions described herein can be downloaded from computer-readable storage media to various computing / processing devices, or downloaded via a network, such as the Internet, local area network, wide area network, and / or wireless network, to an external computer or external storage device. The network may include copper transmission cables, fiber optic transmission, wireless transmission, routers, firewalls, switches, gateway computers, and / or edge servers. A network adapter card or network interface in each computing / processing device receives the computer-readable program instructions from the network and forwards them to the computer-readable storage media in the respective computing / processing device.

[0140] The computer program instructions used to perform the operations of this invention may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, state setting data, or source code or object code written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Smalltalk, C++, etc., and conventional procedural programming languages ​​such as the "C" language or similar programming languages. The computer-readable program instructions may be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or may be connected to an external computer (e.g., via the Internet using an Internet service provider). In some embodiments, electronic circuitry, such as programmable logic circuitry, field-programmable gate arrays (FPGAs), or programmable logic arrays (PLAs), is personalized by utilizing state information from the computer-readable program instructions. This electronic circuitry can execute the computer-readable program instructions to implement various aspects of the invention.

[0141] The computer program product described herein can be implemented specifically through hardware, software, or a combination thereof. In one alternative embodiment, the computer program product is specifically embodied in a computer storage medium; in another alternative embodiment, the computer program product is specifically embodied in a software product, such as a software development kit (SDK), etc.

[0142] Various aspects of the present invention are described herein with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It should be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer-readable program instructions.

[0143] These computer-readable program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing apparatus to produce a machine such that, when executed by the processor of the computer or other programmable data processing apparatus, they create means for implementing the functions / actions specified in one or more blocks of the flowchart and / or block diagram. These computer-readable program instructions can also be stored in a computer-readable storage medium that causes a computer, programmable data processing apparatus, and / or other device to operate in a particular manner; thus, the computer-readable medium storing the instructions comprises an article of manufacture that includes instructions for implementing aspects of the functions / actions specified in one or more blocks of the flowchart and / or block diagram.

[0144] Computer-readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other device to cause a series of operational steps to be performed on the computer, other programmable data processing apparatus, or other device to produce a computer-implemented process, thereby causing the instructions executed on the computer, other programmable data processing apparatus, or other device to perform the functions / actions specified in one or more boxes of a flowchart and / or block diagram.

[0145] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present invention. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of an instruction, which contains one or more executable instructions for implementing a specified logical function. In some alternative implementations, the functions marked in the blocks may occur in a different order than those shown in the drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, may be implemented using a dedicated hardware-based system that performs the specified function or action, or using a combination of dedicated hardware and computer instructions.

[0146] Example embodiments have been disclosed herein, and while specific terminology has been used, it is for illustrative purposes only and should be construed as such, and is not intended to be limiting. In some instances, it will be apparent to those skilled in the art that features, characteristics, and / or elements described in conjunction with particular embodiments may be used alone, or in combination with features, characteristics, and / or elements described in conjunction with other embodiments, unless otherwise expressly indicated. Therefore, those skilled in the art will understand that various changes in form and detail may be made without departing from the scope of the invention as set forth in the appended claims.

[0147] It is understood that the technical features of the above embodiments can be combined arbitrarily. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0148] The above are merely preferred embodiments of this application, and only specifically describe the technical principles of this application. These descriptions are only for explaining the principles of this application and should not be construed as limiting the scope of protection of this application in any way. Based on this explanation, any modifications, equivalent substitutions, and improvements made within the spirit and principles of this application, as well as other specific embodiments of this application that can be conceived by those skilled in the art without creative effort, should be included within the scope of protection of this application.

Claims

1. A crop functional-type chlorophyll fluorescence inversion method, characterized in that, The method comprises the following steps: obtaining a multi-source data cube; obtaining a crop type distribution map based on the multi-source data cube; constructing a solar-induced chlorophyll fluorescence SIF inversion base model according to the crop type distribution map; outputting a crop solar-induced chlorophyll fluorescence time series product according to the solar-induced chlorophyll fluorescence SIF inversion base model.

2. The crop functional type leaf chlorophyll fluorescence inversion method of claim 1, wherein, In the step of obtaining a multi-source data cube, the following steps are specifically included: obtaining multi-source remote sensing data, solar-induced chlorophyll fluorescence SIF original observation data, photosynthetically active radiation PAR, leaf area index LAI and vegetation coverage FVC time series inversion data; resampling and projecting all data to a unified spatial coordinate system and spatial resolution to generate a time-continuous and spatially-matched multi-source data cube.

3. The crop functional type leaf chlorophyll fluorescence inversion method of claim 1, wherein, In the step of obtaining a crop type distribution map based on the multi-source data cube, the following steps are specifically included: training a support vector machine classifier using feature data, and predicting each pixel using the trained support vector machine classifier to obtain a preliminary crop classification map, wherein the feature data includes classification features of the multi-source data cube, vegetation indices and phenological parameters, the vegetation indices include NDVI and EVI, and the phenological parameters include a growth season start period, a peak period and an end period; extracting crop phenology according to the crop distribution map, and applying a vegetation index time series curve to extract key phenological nodes to generate a crop type distribution map, wherein the phenological nodes include a growth initial period, a rapid growth period, a mature period and a harvesting period.

4. The crop functional type leaf chlorophyll fluorescence inversion method of claim 3, wherein, processing the smoothed vegetation index time series curve using a derivative analysis method, and defining the phenological nodes as inflection points based on the vegetation index time series curve.

5. The crop functional type leaf chlorophyll fluorescence inversion method of claim 4, wherein, In the step of constructing a solar-induced chlorophyll fluorescence SIF inversion base model according to the crop type distribution map, the following steps are specifically included: The solar-induced chlorophyll fluorescence SIF inversion base model is expressed as: ； wherein, is the base SIF estimate, representing the fluorescence intensity in a specific waveband; is the incoming light intensity, obtained from the multi-source datacube; leaf area index LAI and fraction of vegetation cover FVC are prior data, determined from the multi-source datacube; are environmental stress factors, including temperature, water stress, quantified by thermal infrared waveband and precipitation data; is the model error term, which follows a normal distribution.

6. The crop functional type leaf chlorophyll fluorescence inversion method of claim 5, wherein, When the crop is rice C3, the solar-induced chlorophyll fluorescence SIF inversion base model is: ； ； wherein, is the maximum fluorescence efficiency, determined by ground experiments or literature values; is the light saturation coefficient, reflecting the light inhibition effect, obtained from the fitting of the fluorescence light response curve; is the temperature stress coefficient, calculated based on the optimal temperature range of crops; is the absolute value of the deviation of temperature from the optimal value; is the actual data; When the crop is sugarcane C4, the solar-induced chlorophyll fluorescence SIF inversion base model is: ； ； wherein, is the fluorescence yield coefficient; is the water stress index; is the actual evapotranspiration; is the potential evapotranspiration.

7. The crop functional type leaf chlorophyll fluorescence inversion method of claim 6, wherein, The method further includes a step of correcting the solar-induced chlorophyll fluorescence SIF inversion base model: The corrected solar-induced chlorophyll fluorescence SIF inversion base model is: ； ； wherein, is the final SIF product, is a phenology-driven strategy function quantifying the SIF variation characteristics of crop growth stages; is a functional-driven strategy function capturing the C3 / C4 photosynthetic physiological differences; is a residual correction term optimized by a data assimilation method, K is a Kalman gain matrix, is an observation value, is an estimation value; A "switching" mode phenology cycle for simulating rice and sugar cane is defined as a piecewise function: where t is time in days; is the initial growth stage; is the fast growth stage; is the harvest stage; , , , is the phenology stage weight coefficient, determined by fitting historical SIF data; , , is the time scale parameter, controlling the function smoothness, calibrated based on crop growth rate; For integrating the C3 / C4 plant intrinsic differences, including photochemical efficiency and water use efficiency, expressed as: wherein, is a tuning factor, set based on crop type, C3 crop , C4 crop ; is the fluorescence enhancement ratio of C4 relative to C3 plants; wherein is the fluorescence quantum yield.

8. The crop functional type leaf chlorophyll fluorescence inversion method of claim 7, wherein, In the step of outputting a crop solar-induced chlorophyll fluorescence time series product according to the solar-induced chlorophyll fluorescence SIF inversion base model, the following steps are specifically included: The solar-induced chlorophyll fluorescence SIF inversion base model outputs crop solar-induced chlorophyll fluorescence, which is summarized in time sequence to generate a spatially-continuous time sequence SIF data set; SIF values of pixels are extracted according to rice areas and sugarcane areas respectively to form a crop-specific SIF parameter set.

9. The crop functional type leaf chlorophyll fluorescence inversion method of claim 8, wherein, The method further includes a step of verifying the solar-induced chlorophyll fluorescence SIF inversion base model output crop solar-induced chlorophyll fluorescence time series product, specifically including: spatial verification is performed by comparing ground measured SIF data not involved in model construction and assimilation with corresponding positions and times output by the solar-induced chlorophyll fluorescence SIF inversion base model. The regional average SIF time series curve output by the sun-induced chlorophyll fluorescence SIF inversion basic model is compared with the time series curve of other independent satellite SIF products in the same period to verify the consistency of the time dynamics.

10. A crop functional-type chlorophyll fluorescence inversion system, characterized in that, The crop functional type sun-induced chlorophyll fluorescence inversion system comprises: a data acquisition unit configured to acquire a multi-source data cube; a processing unit configured to obtain a crop type distribution map based on the multi-source data cube; a model construction unit configured to construct a sun-induced chlorophyll fluorescence SIF inversion basic model according to the crop type distribution map; an output unit configured to output a crop sun-induced chlorophyll fluorescence time series product according to the sun-induced chlorophyll fluorescence SIF inversion basic model.

11. An electronic device, comprising: comprise: one or more processors; a memory for storing one or more programs; when the one or more programs are executed by the one or more processors, the one or more processors implement the method of any one of claims 1 to 9.

12. A computer readable medium having stored thereon a computer program, characterized in that The computer program is executed by the processor to implement the method of any one of claims 1 to 9. The computer program is executed by the processor to implement the method of any one of claims 1 to 9.

Images