A method, system and procedure for quality evaluation and calibration of spaceborne lidar ocean color products

Abstract

The present invention relates to the field of marine remote sensing and satellite monitoring technology, and specifically to a method, system and program for quality evaluation and calibration of satellite-borne lidar ocean color products, especially for improving the inversion accuracy of lidar ocean water parameters. The method includes multi-source data acquisition and standardization, dynamic spatiotemporal matching window optimization, abnormal data filtering, dynamic calculation of seasonal conversion factors, multi-source cross-validation, consistency analysis, regional interpolation modeling, full-link uncertainty quantification, day and night difference correction and dynamic feedback optimization. By constructing an adaptive calibration mechanism and error control process, the inversion accuracy and stability of lidar water color products are significantly improved, providing high-credibility data support for marine ecological monitoring and carbon flux assessment.

Description

Technical Field

[0001] The present invention relates to the field of ocean remote sensing and satellite monitoring technology, and specifically to a method, system and program for evaluating and calibrating the quality of satellite-borne laser radar ocean color products, especially for improving the inversion accuracy of laser radar ocean water parameters. Background Art

[0002] Remote sensing of ocean color is a crucial tool for monitoring marine ecosystems and studying the carbon cycle. For a long time, it has relied primarily on passive optical remote sensing techniques based on the principle of sunlight reflection. However, in practical applications, passive remote sensing technology is severely limited by solar illumination conditions, weather conditions, and the diurnal cycle. In particular, it is unable to provide continuous and effective observational data at night or in cloud-covered areas, making it difficult to achieve all-day, all-weather monitoring of marine ecological parameters.

[0003] In recent years, spaceborne lidar technology, with its active remote sensing mode of emitting laser pulses and receiving scattered signals from targets, has provided new possibilities for overcoming the aforementioned limitations of passive remote sensing technology. With its advantages of high spatial resolution, day and night observation capabilities, and strong penetration into thin clouds, spaceborne lidar has gradually become an important supplementary means of ocean color observation, attracting significant attention in the marine remote sensing community both domestically and internationally.

[0004] However, most current spaceborne lidar systems are designed for atmospheric detection or terrestrial ecological monitoring, and their initial goal is not to accurately invert ocean color parameters. This leads to many challenges in data quality verification and calibration in actual marine applications. Specifically,

[0005] (1) There is a significant mismatch between the spatial scale and revisit period of lidar observations and rapidly changing marine ecological processes. Due to the small footprint of lidar laser spots, limited spatial coverage, and relatively discrete data sampling, these are not well matched to large-scale, high-frequency marine ecological processes. This makes it difficult to effectively apply traditional evaluation methods based on field measurement data and large-scale passive remote sensing data.

[0006] (2) Existing evaluation methods usually lack effective spatiotemporal matching optimization strategies, which makes the lidar data quality evaluation highly sensitive to the selection of artificially set spatiotemporal matching windows. Different window strategies are likely to produce completely different evaluation results, seriously affecting the reliability and stability of product accuracy assessment.

[0007] (3) Since ocean color signals are interfered with by complex factors such as atmospheric aerosols and bubble noise caused by wind speed, traditional evaluation methods fail to effectively identify and filter such abnormal data, resulting in large deviations in product quality evaluation results and making it difficult to effectively guide the optimization and commercial application of lidar water color products.

[0008] (4) Existing methods usually use fixed values or rough partitioning for the conversion factors involved in the inversion of ocean color parameters, which fails to effectively consider the seasonal changes and regional differences in the optical properties of marine organisms. As a result, the inversion products are insufficiently accurate and cannot provide reliable data support for marine ecological monitoring and carbon flux estimation.

[0009] Therefore, it is urgent to develop a multi-source data fusion evaluation and dynamic calibration method for the characteristics of spaceborne lidar water color products, so as to effectively improve the quality and stability of lidar water color products, solve the above-mentioned technical difficulties, and give full play to the potential advantages of lidar in the field of marine ecological environment monitoring. Summary of the Invention

[0010] To address these challenges, this paper provides a novel method for evaluating and calibrating the quality of spaceborne lidar ocean color products. This method proposes a dynamic multi-source collaborative calibration framework that combines lidar, buoy, and passive remote sensing data to optimize high-precision, all-time, and global ocean color products, providing key technical support for marine ecological research and climate change monitoring.

[0011] In order to achieve the above-mentioned purpose, the present invention adopts the following technical solutions:

[0012] A method for evaluating and calibrating the quality of ocean color products of a spaceborne laser radar, comprising the following steps:

[0013] Step S1: Multi-source data collection and standardization:

[0014] Integrate spaceborne lidar water color products, field measurement data, passive remote sensing water color products, and auxiliary data, and perform temporal alignment, spatial registration, and unit standardization of data from different sources;

[0015] Step S2: Dynamic spatiotemporal matching window optimization:

[0016] Based on the spatiotemporal heterogeneity of buoys and lidar, a multi-scale matching window is constructed, and the optimal matching strategy is automatically selected through comprehensive scoring;

[0017] Step S3: Abnormal data filtering and preprocessing:

[0018] Remove outliers and filter atmospheric and bubble noise based on aerosol optical depth (AOD) and wind speed thresholds;

[0019] Step S4: Dynamic calculation of seasonal conversion factor:

[0020] Based on the measured particle backscatter coefficient bbp and the 180° scattering coefficient βp(π) obtained by lidar inversion, the regional specific conversion factor χp(π) is calculated by season and a global seasonal lookup table is generated.

[0021] Step S5: Multi-source cross-validation and consistency analysis:

[0022] Cross-compare lidar water color products, passive water color products, and measured data, and use the weighted average method to generate a consistency report to identify systematic deviations and mark low-confidence areas;

[0023] Step S6: Construction of spatiotemporal interpolation model of regional conversion factors:

[0024] Based on the differences in ocean dynamic zoning and bio-optical properties, the global ocean is divided into several sea areas, and the conversion factor χp(π) is predicted seasonally and spatiotemporally using an interpolation algorithm.

[0025] Step S7: quantify uncertainty in the entire link:

[0026] Through Monte Carlo simulation, we quantify the errors in data acquisition, spatiotemporal matching, band conversion, and inversion algorithms, and generate confidence interval maps of lidar water color products.

[0027] Step S8: day and night difference correction:

[0028] The day and night differences of the conversion factor χp(π) are calculated and corrected for the day and night observation data respectively;

[0029] Step S9 Dynamic Feedback and Product Optimization:

[0030] The calibrated conversion factor χp(π) and error parameters are fed back to the lidar inversion process in real time to update the matching logic and parameter priorities, and output a final water color product quality report containing R², RMSE, and confidence interval indicators.

[0031] Preferably, in step S1, the buoy profile data is aligned with the satellite transit time using a cubic spline interpolation method to address the time asynchrony problem among different observation methods, such as satellite-borne lidar water color products, passive remote sensing water color products, and field measured data.

[0032] The spatial registration process uses WGS84 coordinate system conversion and bilinear interpolation technology to positionally couple the kilometer-level grid data of MODIS with the sub-hundred-meter light spots of the lidar, ensuring that the spatial offset between the buoy's measured point and the satellite observation trajectory is controlled within a hundred-meter error.

[0033] Unit standardization focuses on resolving the differences in physical quantities between sensors. Based on the radiation transfer model, the buoy backscatter coefficient is unified to the lidar wavelength, and auxiliary parameters including wind speed, aerosol optical depth AOD and sea surface temperature are dimensionlessly processed to form a multi-source data set that can be directly compared, providing a consistent benchmark data input for subsequent dynamic matching and calibration.

[0034] Preferably, the comprehensive score calculation formula of the multi-scale matching window in step S2 is:

[0035] Overall score = 0.4 × R 2 + 0.3×(1-RMSE) + 0.3×(1-MAPE)

[0036] Among them, R 2 is the coefficient of determination, RMSE is the root mean square error, and MAPE is the mean absolute percentage error; the matching window with the highest comprehensive score is selected as the optimal strategy.

[0037] Preferably, in step S3, observation samples with AOD>0.3 and wind speed≥9m / s are eliminated or marked to prevent inversion anomalies caused by atmospheric interference and sea bubble noise.

[0038] Preferably, the calculation formula of the conversion factor χp(π) in step S4 is:

[0039] χp(π)=bbp / (2πβp(π)),

[0040] Regional statistics are also conducted for the four seasons of spring, summer, autumn and winter to obtain a global seasonal lookup table with regional specificity.

[0041] Preferably, in the multi-source cross-validation and consistency analysis in step S5, the satellite-borne lidar water color product, the passive remote sensing water color product, and the field measured data are cross-compared, and a weighted average method is used, with the weight being the data confidence, to generate a consistency report, identify systematic deviations, and mark low-confidence areas, wherein the weighted average is calculated as follows:

[0042] ,

[0043] Among them, bbp 融合 is the fused water particle backscatter product, W i is the weight of water color products, passive water color products and measured data, bbp i It is the backscatter coefficient of water color products, passive water color products and measured data;

[0044] If the absolute deviation between the single-source data and the fusion value exceeds 2 times its own uncertainty, the data is judged to have a systematic deviation and marked as an abnormal source; at the same time, the weighted variance of the multi-source data in the unit is calculated:

[0045] ,

[0046] when >0.2×10 -3Or when the amount of valid data is less than 2 groups, the area will be marked as a low confidence area, and its spatial position will be noted in the output result. Finally, a consistency report including the fusion value, deviation distribution and confidence level will be generated to provide a quantitative basis for conversion factor optimization and product iteration.

[0047] Preferably, when zoning the global ocean in step S6, at least 26 ocean dynamic zones are divided based on factors such as ocean circulation, water mass characteristics and bio-optical properties, and Kriging interpolation or other geostatistical methods are used to perform spatiotemporal interpolation prediction of the conversion factor χp(π).

[0048] Preferably, the specific method of step S7 is as follows:

[0049] First, define the probability distribution of the error source in each link:

[0050] The laser radar signal noise follows a Poisson distribution, and its standard deviation is determined by the signal-to-noise ratio (SNR). ;

[0051] The measurement error of the buoy measured data is set to be uniformly distributed based on the calibration certificate, ∈ buoy~ U(-0.02bbp,+0.02bbp);

[0052] The spatial offset error of spatiotemporal matching is modeled as a Gaussian distribution, ∈ space~ N(0,100 2 m 2 );

[0053] Band conversion coefficient γ The uncertainty of is assumed to be normally distributed. γ ~ N(0.78,0.05 2 );

[0054] The inversion algorithm parameter deviation is simulated by Gaussian noise, ∈ algo~ N(0,0.1 2 );

[0055] The Monte Carlo simulation is performed for 1000 independent iterations. In each iteration, the above error parameters are randomly sampled and the input data is perturbed, and the signal-noise superposition is performed in sequence:

[0056] LiDAR βp ( π ) noisy = βp ( π )+ ,

[0057] Buoy data band conversion bbp 532 =bbp 700 ×(532 / 700)-γ+δγ ,in δ, γ is the disturbance term,

[0058] Time and space position offset correction, matching position correction is x new = x +∈ space ,

[0059] and inversion algorithm parameter perturbations βp ( π ) final = βp ( π ) noisy ×(1+∈ algo );

[0060] Finally, the perturbed bbp value is generated through the inversion model;

[0061] Statistically calculate the probability distribution of all iterative results and the 95% confidence interval:

[0062] ,in μ is the mean, σ is the standard deviation;

[0063] And further quantify the contribution of each link to the total uncertainty through variance decomposition,

[0064] i=1, 2, 3, 4 correspond to data acquisition, spatiotemporal matching, band conversion, and inversion algorithm, respectively. The final output is a spatial error hotspot map and a link contribution report to support product quality grading and calibration strategy optimization.

[0065] Preferably, in step S9, the algorithm parameter selection logic is optimized by dynamically adjusting the data source weights and adaptively shrinking the spatiotemporal matching window, and a timeliness weighting factor is introduced to ensure that recent observation data dominates the inversion process.

[0066] Furthermore, the present invention also provides a spaceborne lidar ocean color product quality assessment and calibration system, which executes the method, including:

[0067] a) Data acquisition and management module, used for multi-source data acquisition and standardization;

[0068] b) Spatiotemporal matching optimization module, used to construct and select the optimal matching window based on the heterogeneity of the buoy and the lidar;

[0069] c) Abnormal data processing module, used to filter data whose AOD and wind speed thresholds exceed the preset range;

[0070] d) a seasonal conversion calculation module, configured to calculate and store the conversion factor χp(π);

[0071] e) Partition interpolation and uncertainty assessment module, used to perform spatiotemporal interpolation of conversion factors and output confidence assessment;

[0072] f) Day and night correction and dynamic feedback module, which is used to correct for day and night differences and feed back the correction results to the inversion process in real time to generate the final quality report.

[0073] Preferably, the system further includes a multi-source cross-validation unit, which is used to automatically compare lidar water color products, passive remote sensing data and measured data during the product generation stage, output the confidence interval and consistency score of each indicator, and issue a calibration request instruction when the deviation exceeds the threshold, triggering the system's dynamic feedback process.

[0074] Furthermore, the present invention also provides a computer-readable storage medium having a computer program or instruction stored thereon, which implements the method when the computer program or instruction is executed by a processor.

[0075] Furthermore, the present invention also provides a computer program product, comprising a computer program or instructions, which implement the method when executed by a processor.

[0076] The present invention adopts the above technical solution, and the technical effects are as follows:

[0077] (1) This paper proposes for the first time a dynamically optimized multi-scale spatiotemporal matching strategy that can adaptively select the optimal matching window between the lidar and the measured data. It solves the problem of unstable verification conclusions and insufficient accuracy caused by the traditional manual experience-based window setting, and significantly improves the objectivity and accuracy of the quality evaluation of spaceborne lidar water color products.

[0078] (2) The present invention designs a method based on cross-validation and consistency analysis of multi-source data, integrating lidar water color products, passive remote sensing products and field measurement data. By using confidence-weighted averaging, the method effectively identifies and marks systematic deviations and low-confidence areas, thereby improving the reliability of the ocean water color parameter inversion results and the stability of the data products.

[0079] (3) This invention innovatively proposes a dynamic calculation and regional interpolation model for seasonal conversion factors, which fully considers the spatiotemporal variation of the optical properties of marine organisms, effectively reduces the inversion error caused by fixed conversion factors, significantly improves the accuracy of inverted water color parameters, and enhances the applicability and long-term monitoring capabilities of lidar water color products.

[0080] (4) The present invention solves the problem of systematic differences between day and night observations of satellite-borne lidar for the first time by establishing a day-night difference correction module, realizes the continuous, consistent and high-precision processing of day and night observation data, and ensures the continuity and accuracy of all-weather marine ecological environment monitoring.

[0081] (5) This paper constructs a full-link uncertainty quantification method, which comprehensively evaluates the errors in each link of data acquisition, spatiotemporal matching, band conversion and inversion algorithm through Monte Carlo simulation technology. It realizes the quantitative evaluation of the quality of lidar ocean color products for the first time, and provides an accurate credibility basis for the scientific application and decision support of data products.

[0082] In summary, the present invention significantly improves the accuracy and stability of spaceborne lidar ocean color products, breaks through the various bottlenecks and shortcomings of traditional verification and calibration methods, and can effectively support the high-reliability data needs of marine ecological environment monitoring, carbon flux assessment and climate change research, and has important scientific research and business application value. BRIEF DESCRIPTION OF THE DRAWINGS

[0083] Figure 1 This is a flow chart of the quality evaluation and calibration method for spaceborne lidar ocean color products.

[0084] Figure 2 Evaluation scores for different matching windows.

[0085] Figure 3 Comparison diagram before and after correction. DETAILED DESCRIPTION

[0086] The following is a clear and complete description of the technical solutions in the embodiments of the present invention in conjunction with the embodiments of the present invention. Obviously, the embodiments described are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts are within the scope of protection of the present invention.

[0087] like Figure 1 As shown, the present invention's method for evaluating and calibrating the quality of spaceborne lidar ocean color products specifically includes the following steps: Step 1: Multi-source data acquisition and standardization; Step 2: Dynamic spatiotemporal matching window optimization; Step 3: Abnormal data filtering and preprocessing; Step 4: Dynamic calculation of seasonal conversion factors; Step 5: Multi-source cross-validation and consistency analysis; Step 6: Construction of a spatiotemporal interpolation model for regional conversion factors; Step 7: Full-link uncertainty quantification; Step 8: Day-night difference correction module; Step 9: Dynamic feedback and product optimization. Each step will be described in detail below.

[0088] Step 1: Multi-source data collection and standardization;

[0089] Multi-source data collection and standardization integrates spaceborne lidar ocean color products, field-measured data, passive remote sensing ocean color products, and auxiliary data (wind speed, aerosol optical depth (AOD), and sea surface temperature). Data are then time-aligned, spatially registered, and standardized. Here, the spaceborne lidar ocean particle backscatter coefficient (BBP) product is used as an example, validated using field-measured data from buoys and passive ocean color remote sensing MODIS products.

[0090] To address the time asynchrony problem of different observation methods such as satellite-borne lidar, buoy measurements, and passive remote sensing, the cubic spline interpolation method is used to align the buoy profile data with the satellite transit time. The spatial registration process uses WGS84 coordinate system conversion and bilinear interpolation technology to positionally couple the MODIS kilometer-level grid data with the lidar sub-hundred-meter light spots, ensuring that the spatial offset between the buoy measurement point and the satellite observation trajectory is controlled within a hundred-meter error. Unit standardization focuses on resolving the differences in physical quantities between sensors. Based on the radiation transfer model, the buoy backscatter coefficient is unified to the lidar wavelength, and auxiliary parameters such as wind speed and aerosols are dimensionlessly processed to form a multi-source data set that can be directly compared, providing a consistent benchmark data input for subsequent dynamic matching and calibration.

[0091] Step 2: Dynamic spatiotemporal matching window optimization. Based on the spatiotemporal heterogeneity of the buoy and the lidar, a multi-scale matching window is constructed. The optimal window is automatically selected through statistical scoring. The calculation formula for the statistical score is:

[0092] Overall score = 0.4 × R 2 + 0.3×(1-RMSE) + 0.3×(1-MAPE)

[0093] Among them, R 2 is the coefficient of determination, RMSE is the root mean square error, and MAPE is the mean absolute percentage error; the window with the highest score is selected as the optimal matching strategy.

[0094] The time window here is 3 to 24 hours, and the space window is 9-50 kilometers. Figure 2 Evaluate scores for different matching windows.

[0095] Step 3: Abnormal data filtering and preprocessing: remove outliers, and filter out atmospheric interference and bubble noise by combining the thresholds of AOD>0.3 and wind speed ≥9 m / s.

[0096] Step 4: Dynamic calculation of the seasonal conversion factor: Based on the measured data of the particle backscatter coefficient bbp and the 180° scattering coefficient βp(π) obtained by lidar inversion, the regional specific conversion factor χp(π) = bbp / (2πβp(π)) is calculated according to the season to generate a global seasonal lookup table.

[0097] Step 5: In multi-source cross-validation and consistency analysis, the lidar water color product, passive water color product, and measured data are cross-compared, and a weighted average method (weights are data confidence levels) is used to generate a consistency report, identify systematic deviations, and mark low-confidence areas.

[0098] The weighted average is calculated as follows:

[0099] ,

[0100] Among them, bbp 融合 is the fused water particle backscatter product, W i is the weight of water color products, passive water color products and measured data, bbp i It is the backscatter coefficient of water color products, passive water color products and measured data;

[0101] If the absolute deviation between the single-source data and the fusion value exceeds twice its own uncertainty (e.g. the BBP error of the lidar is ±0.5×10 -³ m -¹ When the deviation threshold is set to 1.0×10 -³ m -¹ ), the data is judged to have systematic deviation and marked as an abnormal source; at the same time, the weighted variance of the multi-source data in the unit is calculated:

[0102] ,

[0103] when >0.2×10 -3 Or when the amount of valid data is less than 2 groups, the area will be marked as a low confidence area, and its spatial position will be noted in the output result. Finally, a consistency report including the fusion value, deviation distribution and confidence level will be generated to provide a quantitative basis for conversion factor optimization and product iteration.

[0104] Step 6: In the construction of the spatiotemporal interpolation model of regional conversion factors, the global ocean is divided into 26 sea areas based on the differences in ocean dynamic zoning and bio-optical characteristics. The seasonal spatial distribution model of χp(π) is established in combination with the Kriging interpolation algorithm to predict the parameters of the unsampled areas.

[0105] Step 7: In the full-link uncertainty quantification, Monte Carlo simulation is used to quantify the errors in data acquisition, spatiotemporal matching, band conversion, and inversion algorithms, generating a 95% confidence interval map for the lidar water color product.

[0106] First, define the probability distribution of the error source in each link:

[0107] The laser radar signal noise follows a Poisson distribution, and its standard deviation is determined by the signal-to-noise ratio (SNR). ;

[0108] The measurement error of the buoy measured data is set to be uniformly distributed based on the calibration certificate, ∈ buoy~ U(-0.02bbp,+0.02bbp);

[0109] The spatial offset error of spatiotemporal matching is modeled as a Gaussian distribution, ∈ space~ N(0,100 2 m 2 );

[0110] Band conversion coefficient γ The uncertainty of is assumed to be normally distributed. γ~ N(0.78,0.05 2 );

[0111] The inversion algorithm parameter deviation is simulated by Gaussian noise, ∈ algo~ N(0,0.1 2 );

[0112] The Monte Carlo simulation is performed for 1000 independent iterations. In each iteration, the above error parameters are randomly sampled and the input data is perturbed, and the signal-noise superposition is performed in sequence:

[0113] LiDAR βp ( π ) noisy = βp ( π )+ ,

[0114] Buoy data band conversion bbp 532 =bbp 700 ×(532 / 700) -γ+δγ ,in δ, γ is the disturbance term,

[0115] Time and space position offset correction, matching position correction is x new = x +∈ space ,

[0116] and inversion algorithm parameter perturbations βp ( π ) final = βp ( π ) noisy ×(1+∈ algo );

[0117] Finally, the perturbed bbp value is generated through the inversion model;

[0118] Statistically calculate the probability distribution of all iterative results and the 95% confidence interval:

[0119] ,in μ is the mean, σ is the standard deviation;

[0120] And further quantify the contribution of each link to the total uncertainty through variance decomposition,

[0121] i =1, 2, 3, 4 correspond to data acquisition, space-time matching, band conversion, and inversion algorithm respectively, and ultimately output a spatial error hotspot map and a link contribution report to support product quality grading and calibration strategy optimization.

[0122] Step 8: In the day-night difference correction module, separate the daytime and nighttime observation data, calculate the day-night difference of χp(π), design the time-sharing calibration coefficient and embed it into the inversion algorithm.

[0123] Step 9: In dynamic feedback and product optimization, the calibrated χp(π) and error parameters are fed back to the lidar inversion process in real time. By dynamically adjusting the data source weight (e.g., increasing the priority of measured data to 0.9 in densely populated areas with buoys) and adaptively shrinking the spatiotemporal matching window (spatial window in highly dynamic sea areas is reduced to 5 km, and time window is compressed to ±3 hours), the algorithm parameter selection logic is optimized (core parameters with error contribution less than 20% are preferred), and a timeliness weighting factor is introduced ( α ( t )= e- 0.1Δ t ) ensures that recent observational data dominates the inversion process, ultimately generating global BBP products and supporting quality reports, covering R², RMSE, 95% confidence intervals, and spatial distribution maps of low-confidence areas, to support data credibility grading for scientific research and operational applications.

[0124] Figure 3 This is the difference map before and after correction. After correction, RMSE dropped from 0.009 to 0.003, and the inversion accuracy was significantly improved.

[0125] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware. Furthermore, the present application may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0126] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0127] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0128] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 The steps for the function specified in one or more boxes.

[0129] In a typical configuration, a computing device includes one or more processors (CPUs), input / output interfaces, network interfaces, and memory.

[0130] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0131] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can be implemented using any method or technology for information storage. Information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer storage media include, but are not limited to, phase-change RAM (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassettes, magnetic disk storage or other magnetic storage devices, or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer-readable media does not include transitory computer-readable media, such as modulated data signals and carrier waves.

[0132] The above is a description of the embodiments of the present invention. The above description of the disclosed embodiments will enable professionals in the field to implement or use the present invention. Various modifications to these embodiments will be apparent to professionals in the field. The general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but should conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for evaluating and calibrating the quality of ocean color products from spaceborne lidar, characterized in that: The following steps are involved: Step S1: Multi-source data collection and standardization Integrate spaceborne lidar ocean color products, in situ measurements, passive remote sensing ocean color products, and auxiliary data, including wind speed, aerosol optical depth (AOD), and sea surface temperature, and perform temporal alignment, spatial registration, and unit standardization on heterogeneous data sources. Step S2: Dynamic spatiotemporal matching window optimization Based on the spatiotemporal heterogeneity of buoys and lidars, a multi-scale matching window is constructed, and the optimal matching strategy is automatically selected through comprehensive scoring. The comprehensive scoring calculation formula of the multi-scale matching window is: Overall score = 0.4 × R 2 + 0.3×(1−RMSE) + 0.3×(1−MAPE) Among them, R 2 is the coefficient of determination, RMSE is the root mean square error, and MAPE is the mean absolute percentage error; the matching window with the highest comprehensive score is selected as the optimal strategy; Step S3: Abnormal data filtering and preprocessing Remove outliers and filter out atmospheric and bubble noise based on aerosol optical depth (AOD) and wind speed thresholds; Step S4: Dynamic calculation of seasonal conversion factors, Based on the measured particle backscatter coefficient (bbp) and the 180° scattering coefficient (βp(π)) obtained from lidar inversion, a regionally specific conversion factor (χp(π)) was calculated by season, and a global seasonal lookup table was generated. The conversion factor (χp(π)) was calculated as: χp(π)=bbp / (2πβp(π)). Regional statistics were performed for each of the four seasons, spring, summer, autumn, and winter, to generate a global seasonal lookup table with regional specificity. Step S5: Multi-source cross-validation and consistency analysis Cross-compare lidar water color products, passive water color products, and measured data, and use the weighted average method to generate a consistency report to identify systematic deviations and mark low-confidence areas; Step S6: Construction of spatiotemporal interpolation model for regional conversion factors Based on the differences in ocean dynamic zoning and bio-optical properties, the global ocean is divided into several sea areas, and the conversion factor χp(π) is predicted seasonally and spatiotemporally using an interpolation algorithm. Step S7: Full-link uncertainty quantification Monte Carlo simulations are used to quantify the errors in data acquisition, spatiotemporal matching, band conversion, and inversion algorithms, and to generate confidence interval maps of lidar water color products. The specific method is as follows: First, define the probability distribution of the error source in each link: The laser radar signal noise follows a Poisson distribution, and its standard deviation is determined by the signal-to-noise ratio (SNR). ; The measurement error of the buoy measured data is set to be uniformly distributed based on the calibration certificate. ϵ buoy ∼U(−0.02bbp,+0.02bbp); The spatial offset error of spatiotemporal matching is modeled as a Gaussian distribution, ϵ space ∼N(0,100 2 m 2 ); Band conversion coefficient γ The uncertainty of is assumed to be normally distributed. γ ∼N(0.78,0.05 2 ); The inversion algorithm parameter deviation is simulated by Gaussian noise. ϵ algo ∼N(0,0.1 2 ); The Monte Carlo simulation is performed for 1000 independent iterations. In each iteration, the above error parameters are randomly sampled and the input data is perturbed, and the signal-noise superposition is performed in sequence: LiDAR βp ( π ) noisy = βp ( π )+ σ SNR , Buoy data band conversion bbp 532 =bbp 700 ×(532 / 700) −γ+δγ ,in δ, γ is the disturbance term, Time and space position offset correction, matching position correction is x new = x + ϵ space , and inversion algorithm parameter perturbations βp ( π ) final = βp ( π ) noisy ×(1+ ϵ algo ); Finally, the perturbed bbp value is generated through the inversion model; Statistically calculate the probability distribution of all iterative results and the 95% confidence interval: ,in μ is the mean, σ is the standard deviation; The contribution of each link to the total uncertainty is further quantified through variance decomposition: ， i =1, 2, 3, 4 correspond to data acquisition, spatiotemporal matching, band conversion, and inversion algorithm, respectively. The final output is a spatial error heat map and a link contribution report to support product quality grading and calibration strategy optimization; Step S8: Day-night difference correction Separate daytime and nighttime observation data, calculate the daytime and nighttime differences of χp(π), design time-sharing calibration coefficients and embed them into the inversion algorithm; Step S9: Dynamic feedback and product optimization, The calibrated conversion factor χp(π) and error parameters are fed back to the lidar inversion process in real time to update the matching logic and parameter priorities, and output a final water color product quality report containing R², RMSE, and confidence interval indicators.

2. The method according to claim 1, characterized in that In step S1, the buoy profile data is aligned with the satellite transit time using a cubic spline interpolation method to address the time asynchrony problem among different observation methods, such as satellite-borne lidar water color products, passive remote sensing water color products, and field measurement data. The spatial registration process uses WGS84 coordinate system conversion and bilinear interpolation technology to positionally couple the kilometer-level grid data of MODIS with the sub-hundred-meter light spots of the lidar, ensuring that the spatial offset between the buoy's measured point and the satellite observation trajectory is controlled within a hundred-meter error. Unit standardization focuses on resolving the differences in physical quantities between sensors. Based on the radiation transfer model, the buoy backscatter coefficient is unified to the lidar wavelength, and auxiliary parameters including wind speed, aerosol optical depth AOD and sea surface temperature are dimensionlessly processed to form a multi-source data set that can be directly compared, providing a consistent benchmark data input for subsequent dynamic matching and calibration.

3. The method according to claim 1, characterized in that In step S3, observation samples with AOD>0.3 and wind speed≥9m / s are removed or marked to prevent inversion anomalies caused by atmospheric interference and sea bubble noise.

4. The method according to claim 1, wherein In the construction of the spatiotemporal interpolation model of the conversion factor χp(π) in step S6, the global ocean is divided into 26 sea areas based on the differences in ocean dynamic zoning and bio-optical characteristics. The seasonal spatial distribution model of χp(π) is established in combination with the Kriging interpolation algorithm to predict the parameters of the unsampled area.

5. The method according to claim 1, wherein In the multi-source cross-validation and consistency analysis in step S5, the satellite-borne lidar water color product, the passive remote sensing water color product, and the field measured data are cross-compared, and a weighted average method is used, with the weight being the data confidence level, to generate a consistency report, identify systematic deviations, and mark low-confidence areas. The weighted average is calculated as follows: ， Among them, bbp 融合 is the fused water particle backscatter product, W i The weights of spaceborne lidar water color products, field measurement data, and passive remote sensing water color products, bbp i It is the backscatter coefficient of water color products, passive water color products and measured data; If the absolute deviation between the single-source data and the fusion value exceeds 2 times its own uncertainty, the data is judged to have a systematic deviation and marked as an abnormal source; at the same time, the weighted variance of the multi-source data in the unit is calculated: ， when >0.2×10 -3 Or when the amount of valid data is less than 2 groups, the area will be marked as a low confidence area, and its spatial position will be noted in the output result. Finally, a consistency report including the fusion value, deviation distribution and confidence level will be generated to provide a quantitative basis for conversion factor optimization and product iteration.

6. The method according to claim 1, characterized in that In step S9, the algorithm parameter selection logic is optimized by dynamically adjusting the data source weights and adaptively shrinking the spatiotemporal matching window, and a timeliness weighting factor is introduced to ensure that recent observation data dominates the inversion process.

7. A spaceborne lidar ocean color product quality evaluation and calibration system, characterized by: The system implements the method according to any one of claims 1 to 6, comprising: Data collection and management module, used for multi-source data collection and standardization; The spatiotemporal matching optimization module is used to construct and select the optimal matching window based on the heterogeneity of the buoy and the lidar; Abnormal data processing module, used to filter data whose AOD and wind speed thresholds exceed the preset range; A seasonal conversion calculation module, used for calculating and storing the conversion factor χp(π); Partition interpolation and uncertainty assessment module, used to perform spatiotemporal interpolation of conversion factors and output confidence assessment; The day and night correction and dynamic feedback module is used to correct for day and night differences, and feed the correction results back to the inversion process in real time to generate a final quality report.

8. A computer-readable storage medium having a computer program or instruction stored thereon, characterized in that: When the computer program or instruction is executed by a processor, the method according to any one of claims 1 to 6 is implemented.

9. A computer program product comprising a computer program or instructions, characterized in that When the computer program or instruction is executed by a processor, the method according to any one of claims 1 to 6 is implemented.

Images