Calculation method of hyperspectral karst carbon sink quantity

Abstract

The invention discloses a hyperspectral karst carbon sink quantity calculation method, and particularly relates to the field of ecological carbon sink detection, and the method comprises the following steps: S100, carrying out the data collection of a karst region through a hyperspectral sensor, obtaining the hyperspectral image data, and collecting the related geographic information, meteorological data and karst geological data auxiliary information; step S200, processing the hyperspectral image data, and extracting feature information related to karst carbon sink; step S300, based on the extracted feature information and the collected auxiliary information, establishing a hyperspectral karst carbon sink calculation model; and S400, verifying the established model by using an independent test data set, and evaluating the accuracy and reliability of the model. According to the method, the characteristic information related to karst carbon sink can be more accurately extracted, the calculation precision of the carbon sink amount is improved, a large amount of hyperspectral data and auxiliary information can be automatically processed, manual intervention is reduced, and the calculation efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the field of ecological carbon sink detection, and more specifically, to a method for calculating a high-spectral karst carbon sink. Background Art

[0002] Carbon sinks include forest carbon sinks, grassland carbon sinks, cultivated land carbon sinks, soil carbon sinks and ocean carbon sinks. At the same time, carbon sinks are also an indispensable manifestation of the carbon cycle process. Karst carbon sinks are often regarded as "missing sinks" in the carbon cycle system. The rocks in the southwestern karst area are highly soluble, and their geological composition presents a special above-ground and underground binary structure, which leads to the carbon migration path and transformation process in the southwestern karst area being often much more complicated than that in non-karst areas.

[0003] Karst areas play an important role in the global carbon cycle. Accurately calculating karst carbon sinks is of great significance for assessing the carbon balance of ecosystems and responding to climate change. Traditional methods for calculating karst carbon sinks often have problems such as low accuracy and complex operations. With the development of hyperspectral technology, a new way to more accurately calculate karst carbon sinks has been provided.

[0004] In order to solve the above problems, a technical solution is now provided. Summary of the invention

[0005] In order to overcome the above-mentioned defects of the prior art, an embodiment of the present invention provides a method for calculating a hyperspectral karst carbon sink to solve the problems raised in the above-mentioned background technology.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] A method for calculating a hyperspectral karst carbon sink comprises the following steps:

[0008] Step S100, using a hyperspectral sensor to collect data in the karst area, obtaining hyperspectral image data, and collecting relevant geographic information, meteorological data and karst geological data auxiliary information;

[0009] Step S200, processing the hyperspectral image data to extract characteristic information related to karst carbon sinks;

[0010] Step S300, establishing a hyperspectral karst carbon sink calculation model based on the extracted feature information and the collected auxiliary information;

[0011] Step S400, using an independent test data set to verify the established model and evaluate the accuracy and reliability of the model;

[0012] Step S500, input the new hyperspectral image data into the verified model to calculate the karst carbon sink, and visualize and analyze the calculation results according to different needs.

[0013] In a preferred embodiment, in step S100, a representative karst area is selected as a data collection area, and a preliminary survey is conducted on the collection area;

[0014] Collect hyperspectral image data at the selected time and route, organize and annotate the data, and record the collection time, location, sensor parameters and other information for subsequent processing and analysis;

[0015] Integrate hyperspectral image data and auxiliary information to establish a unified data management platform.

[0016] In a preferred embodiment, in step S200, according to the research objective of karst carbon sink, an area of ​​interest that may be related to carbon sink is determined on the hyperspectral image;

[0017] Perform principal component analysis on all bands of the hyperspectral image and convert multi-band data into a few principal components;

[0018] Extract independent spectral components from hyperspectral images and discover spectral features hidden in hyperspectral images

[0019] From the numerous extracted features, the features most relevant to the karst carbon sink are selected, and the features are screened and verified to ensure the relevance of the selected features to the karst carbon sink.

[0020] In a preferred embodiment, in step S300, the extracted hyperspectral feature information is integrated with the collected geographic information, meteorological data and karst geological data auxiliary information;

[0021] Conduct preliminary analysis on the integrated data set to understand the relationship and distribution characteristics of each variable;

[0022] Select a modeling method and divide the integrated data set into a training set and a test set. The training set is used for model training and parameter optimization, and the test set is used to evaluate the performance and generalization ability of the model.

[0023] Use the training set data to train the selected modeling method, and adjust the model weights and parameters according to the model type and parameter settings so that the model can accurately fit the training data;

[0024] Improve the performance of the model by adjusting the parameters of the model.

[0025] In a preferred embodiment, the modeling method specifically includes the following contents:

[0026] Establishing a linear relationship between karst carbon sink and multiple independent variables, including hyperspectral features and auxiliary information;

[0027] The regression coefficient is estimated by the least squares method to minimize the sum of square errors between the model's predicted value and the actual observed value. The specific formula is:

[0028] Y=β0+β1X1+β2X2+…+β n X n +ε;

[0029] Among them, Y represents the karst carbon sink, X1, X2, …, X n represents n independent variables, β0 represents the intercept term, β1, β2, …, β n is the regression coefficient, and ε is the random error term.

[0030] In a preferred embodiment, in step S400, the error between the model prediction value and the actual observation value is calculated using the following formula:

[0031]

[0032] Where R represents the mean absolute deviation between the predicted value and the actual value, m represents the number of samples in the test data set, and y i represents the actual observed value, represents the model prediction value;

[0033] The test data set is further divided into multiple subsets, and multiple cross-validations are performed. The stability of the model performance is observed by calculating the evaluation indicators on different subsets.

[0034] In a preferred embodiment, in step S500, the hyperspectral image data and auxiliary information are input into the model, and the model is allowed to perform prediction to obtain the calculation result of the karst carbon sink;

[0035] Select a data visualization platform, integrate the hyperspectral image data and karst carbon sink data, and display them simultaneously on the data visualization platform;

[0036] Use data visualization platforms to create maps, display hyperspectral images as basemaps, and adjust the color, contrast, and transparency of images to highlight different features.

[0037] Overlay karst carbon sink data on the map and use different colors, symbols or charts to represent the size and distribution of the carbon sink.

[0038] In a preferred embodiment, a data visualization platform is used to perform spatial analysis to understand the spatial distribution characteristics of karst carbon sinks;

[0039] Conduct spatial interpolation analysis to estimate the carbon sink value in unknown areas based on the carbon sink data at known points;

[0040] Compare and analyze the carbon sink results in different regions or at different times to understand the differences and changes in karst carbon sinks, and use charts and maps to intuitively display the carbon sink situation under different circumstances.

[0041] The technical effects and advantages of the method for calculating the hyperspectral karst carbon sink of the present invention are as follows:

[0042] 1. Using hyperspectral technology, it is possible to more accurately extract characteristic information related to karst carbon sinks and improve the accuracy of carbon sink calculation;

[0043] 2. By establishing a machine learning model, a large amount of hyperspectral data and auxiliary information can be automatically processed, reducing manual intervention and improving computing efficiency;

[0044] 3. It has good versatility and scalability, and can be applied to the calculation of karst carbon sinks in different regions, providing strong support for ecological environmental protection and climate change research in karst areas. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 The figure is a flow chart of a method for calculating a hyperspectral karst carbon sink according to the present invention. DETAILED DESCRIPTION

[0046] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0047] Example 1

[0048] Figure 1 A method for calculating a hyperspectral karst carbon sink is provided, which specifically comprises the following steps:

[0049] Step S100, using a hyperspectral sensor to collect data in the karst area, obtaining hyperspectral image data, and collecting relevant geographic information, meteorological data and karst geological data auxiliary information;

[0050] Step S200, processing the hyperspectral image data to extract characteristic information related to karst carbon sinks;

[0051] Step S300, establishing a hyperspectral karst carbon sink calculation model based on the extracted feature information and the collected auxiliary information;

[0052] Step S400, using an independent test data set to verify the established model and evaluate the accuracy and reliability of the model;

[0053] Step S500, input the new hyperspectral image data into the verified model to calculate the karst carbon sink, and visualize and analyze the calculation results according to different needs.

[0054] It should be added that while using hyperspectral sensors to collect image data, auxiliary information such as geographic information, meteorological data, and karst geological data are also collected. This can provide a comprehensive understanding of the situation in the karst area from multiple perspectives and provide a rich data basis for the accurate calculation of carbon sinks. Different types of data can complement each other and improve the reliability of the calculation results.

[0055] Based on the extracted characteristic information and auxiliary information, a hyperspectral karst carbon sink calculation model is established to make the model more targeted and scientific. The model can make full use of the high resolution and multi-band characteristics of hyperspectral data, combined with other auxiliary information, to more accurately reflect the actual situation of karst carbon sink.

[0056] In step S100, a representative karst area is selected as a data collection area, and a preliminary survey is conducted on the collection area;

[0057] Collect hyperspectral image data at the selected time and route, organize and annotate the data, and record the collection time, location, sensor parameters and other information for subsequent processing and analysis;

[0058] Integrate hyperspectral image data and auxiliary information to establish a unified data management platform.

[0059] It should be added that by selecting representative karst areas as data collection areas, it can be ensured that the collected samples can better reflect the characteristics of the entire karst area. Representative areas usually cover different types of karst landforms, vegetation coverage and geological conditions, making subsequent analysis and calculation results more universal and reliable.

[0060] Collecting hyperspectral image data at a selected time and route can ensure data consistency and comparability. For example, selecting a period of clear weather and sufficient light for collection can reduce the impact of atmospheric interference and light changes on image quality, so that data collected at different times have similar lighting conditions, which is convenient for subsequent analysis and processing.

[0061] Integrating hyperspectral image data and auxiliary information and establishing a unified data management platform can realize centralized storage and management of data. This can avoid the problem that data is scattered in different storage devices or folders, which is easy to lose or difficult to find. The unified data management platform can also facilitate data backup, recovery and sharing, improving data security and availability.

[0062] In step S200, according to the research objectives of karst carbon sink, an area of ​​interest that may be related to carbon sink is determined on the hyperspectral image;

[0063] Perform principal component analysis on all bands of the hyperspectral image and convert multi-band data into a few principal components;

[0064] Extract independent spectral components from hyperspectral images and discover spectral features hidden in hyperspectral images

[0065] From the numerous extracted features, the features most relevant to the karst carbon sink are selected, and the features are screened and verified to ensure the relevance of the selected features to the karst carbon sink.

[0066] It should be added that determining the areas of interest that may be related to carbon sinks according to the research objectives of karst carbon sinks can make the research more focused. This helps to concentrate on analyzing specific areas that have an important impact on karst carbon sinks and avoid wasting time and resources on irrelevant areas.

[0067] The principal components are linear combinations of the original bands, which are independent of each other and contain different spectral information. By analyzing the characteristics of the principal components, the main spectral features related to karst carbon sinks can be extracted.

[0068] For example, the first principal component usually reflects the overall brightness and contrast of the image, while subsequent principal components may be related to specific object types or spectral characteristics.

[0069] Different from principal component analysis, independent component analysis does not require the components to be orthogonal to each other, but rather pursues statistical independence between components.

[0070] Through independent component analysis, some spectral features hidden in hyperspectral images can be discovered, which may be related to specific processes or factors of karst carbon sinks.

[0071] In step S300, the extracted hyperspectral feature information is integrated with the collected geographic information, meteorological data and karst geological data auxiliary information;

[0072] Conduct preliminary analysis on the integrated data set to understand the relationship and distribution characteristics of each variable;

[0073] Select a modeling method and divide the integrated data set into a training set and a test set. The training set is used for model training and parameter optimization, and the test set is used to evaluate the performance and generalization ability of the model.

[0074] Use the training set data to train the selected modeling method, and adjust the model weights and parameters according to the model type and parameter settings so that the model can accurately fit the training data;

[0075] Improve the performance of the model by adjusting the parameters of the model.

[0076] It should be added that the integration of the extracted hyperspectral feature information with auxiliary information such as geographic information, meteorological data and karst geological data can fully utilize the advantages of multi-source data. Different types of data can reflect the characteristics of karst areas and carbon sink-related factors from different angles. Combining this information can more comprehensively and accurately describe the situation of karst carbon sinks.

[0077] The modeling method specifically includes the following contents:

[0078] Establishing a linear relationship between karst carbon sink and multiple independent variables, including hyperspectral features and auxiliary information;

[0079] The regression coefficient is estimated by the least squares method to minimize the sum of square errors between the model's predicted value and the actual observed value. The specific formula is:

[0080] Y=β0+β1X1+β2X2+…+β n X n +ε;

[0081] Among them, Y represents the karst carbon sink, X1, X2, …, X n represents n independent variables, β0 represents the intercept term, β1, β2, …, β n is the regression coefficient, and ε is the random error term.

[0082] It should be added that the linear relationship between karst carbon sink and multiple independent variables (hyperspectral features and auxiliary information) is clarified, so that the model has a clear mathematical expression. This clear relationship helps to understand which factors affect the karst carbon sink and the extent of the influence, and provides a theoretical basis for in-depth research on the mechanism of karst carbon sink.

[0083] In step S400, the error between the model prediction value and the actual observation value is calculated using the following formula:

[0084]

[0085] Where R represents the mean absolute deviation between the predicted value and the actual value, m represents the number of samples in the test data set, and y irepresents the actual observed value, represents the model prediction value;

[0086] The test data set is further divided into multiple subsets, and multiple cross-validations are performed. The stability of the model performance is observed by calculating the evaluation indicators on different subsets.

[0087] It should be added that one can get an intuitive understanding of the average degree of difference between the model predictions and the actual observed values.

[0088] Further dividing the test data set into multiple subsets for cross-validation can more comprehensively evaluate the performance stability of the model. Different subsets represent different data distributions. By evaluating on multiple subsets, we can observe the performance of the model under different data conditions.

[0089] In step S500, the hyperspectral image data and auxiliary information are input into the model, and the model is allowed to make predictions to obtain the calculation results of the karst carbon sink;

[0090] Select a data visualization platform, integrate the hyperspectral image data and karst carbon sink data, and display them simultaneously on the data visualization platform;

[0091] Use data visualization platforms to create maps, display hyperspectral images as basemaps, and adjust the color, contrast, and transparency of images to highlight different features.

[0092] Overlay karst carbon sink data on the map and use different colors, symbols or charts to represent the size and distribution of the carbon sink.

[0093] Use the data visualization platform to conduct spatial analysis to understand the spatial distribution characteristics of karst carbon sinks;

[0094] Conduct spatial interpolation analysis to estimate the carbon sink value in unknown areas based on the carbon sink data at known points;

[0095] Compare and analyze the carbon sink results in different regions or at different times to understand the differences and changes in karst carbon sinks, and use charts and maps to intuitively display the carbon sink situation under different circumstances.

[0096] It should be added that by inputting hyperspectral image data and auxiliary information into a verified model for prediction, the calculation results of karst carbon sink can be quickly obtained. The use of mature models can ensure the accuracy and reliability of the results and provide a scientific basis for subsequent analysis and decision-making.

[0097] Choosing a data visualization platform for data integration and display can present complex data in an intuitive form.

[0098] Displaying hyperspectral images as basemaps and adjusting color, contrast, and transparency can highlight different land features and help users better identify and understand the landforms and ecological environment in karst areas. This provides an intuitive visual reference for analyzing the relationship between carbon sinks and land features.

[0099] Using different colors, symbols or charts to represent the size and distribution of carbon sinks can clearly show the spatial trend of carbon sinks. Users can quickly identify high and low carbon sink areas, providing an intuitive basis for formulating targeted protection and management measures.

[0100] Using the data visualization platform for spatial analysis can provide a deep understanding of the spatial distribution characteristics of karst carbon sinks. For example, calculating statistical indicators such as mean and standard deviation, analyzing the concentration trend and dispersion of carbon sinks, can help identify hot spots and weak links in carbon sinks, and provide guidance for resource allocation and key protection.

[0101] Through spatial interpolation analysis, the carbon sink value of unknown areas can be inferred based on the carbon sink data of known points. This is of great significance for filling data gaps and fully understanding the carbon sink situation in karst areas, and providing more complete data support for regional planning and ecological restoration.

[0102] Comparative analysis of carbon sink results in different regions or at different times can intuitively show the differences and changes in carbon sinks. By comparing charts and maps, we can quickly discover the changing trends and influencing factors of carbon sinks, providing strong support for evaluating the effectiveness of protection measures and formulating long-term ecological management strategies.

[0103] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters and thresholds in the formula are set by technicians in this field according to actual conditions.

[0104] The above embodiments can be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website site, computer, server or data center to another website site, computer, server or data center by wired (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that contains one or more available media sets. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state hard disk.

[0105] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0106] In the several embodiments provided in the present application, it should be understood that the disclosed methods can be implemented in other ways. For example, the device embodiments described above are only schematic, for example, the division of the units is only a logical function division, and there may be other division methods in actual implementation. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed may be through some interfaces, indirect coupling or communication connection of devices or units, which may be electrical, mechanical or other forms.

[0107] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, and may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0108] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0109] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0110] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

[0111] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for calculating hyperspectral karst carbon sink, characterized in that: The steps include: Step S100, using a hyperspectral sensor to collect data in the karst area, obtaining hyperspectral image data, and collecting relevant geographic information, meteorological data and karst geological data auxiliary information; Step S200, processing the hyperspectral image data to extract characteristic information related to karst carbon sinks; Step S300, establishing a hyperspectral karst carbon sink calculation model based on the extracted feature information and the collected auxiliary information; Step S400, using an independent test data set to verify the established model and evaluate the accuracy and reliability of the model; Step S500, input the new hyperspectral image data into the verified model to calculate the karst carbon sink, and visualize and analyze the calculation results according to different needs.

2. The method for calculating the hyperspectral karst carbon sink according to claim 1, characterized in that: In step S100, a representative karst area is selected as a data collection area, and a preliminary survey is conducted on the collection area; Collect hyperspectral image data at the selected time and route, organize and annotate the data, and record the collection time, location, sensor parameters and other information for subsequent processing and analysis; Integrate hyperspectral image data and auxiliary information to establish a unified data management platform.

3. The method for calculating hyperspectral karst carbon sink according to claim 1, characterized in that: In step S200, according to the research objectives of karst carbon sink, an area of ​​interest that may be related to carbon sink is determined on the hyperspectral image; Perform principal component analysis on all bands of the hyperspectral image and convert multi-band data into a few principal components; Extract independent spectral components from hyperspectral images and discover spectral features hidden in hyperspectral images From the numerous extracted features, the features most relevant to the karst carbon sink are selected, and the features are screened and verified to ensure the relevance of the selected features to the karst carbon sink.

4. The method for calculating hyperspectral karst carbon sink according to claim 1, characterized in that: In step S300, the extracted hyperspectral feature information is integrated with the collected geographic information, meteorological data and karst geological data auxiliary information; Conduct preliminary analysis on the integrated data set to understand the relationship and distribution characteristics of each variable; Select a modeling method and divide the integrated data set into a training set and a test set. The training set is used for model training and parameter optimization, and the test set is used to evaluate the performance and generalization ability of the model. Use the training set data to train the selected modeling method, and adjust the model weights and parameters according to the model type and parameter settings so that the model can accurately fit the training data; Improve the performance of the model by adjusting the parameters of the model.

5. The method for calculating hyperspectral karst carbon sink according to claim 4, characterized in that: The modeling method specifically includes the following contents: Establishing a linear relationship between karst carbon sink and multiple independent variables, including hyperspectral features and auxiliary information; The regression coefficient is estimated by the least squares method to minimize the sum of square errors between the model's predicted value and the actual observed value. The specific formula is: Y=β0+β1X1+β2X2+…+β n X n +e; Among them, Y represents the karst carbon sink, X1, X2, …, X n represents n independent variables, β0 represents the intercept term, β1, β2, …, β n is the regression coefficient, and ε is the random error term.

6. The method for calculating hyperspectral karst carbon sink according to claim 1, characterized in that: In step S400, the error between the model prediction value and the actual observation value is calculated using the following formula: Where R represents the mean absolute deviation between the predicted value and the actual value, m represents the number of samples in the test data set, and y i represents the actual observed value, represents the model prediction value; The test data set is further divided into multiple subsets, and multiple cross-validations are performed. The stability of the model performance is observed by calculating the evaluation indicators on different subsets.

7. The method for calculating hyperspectral karst carbon sink according to claim 1, characterized in that: In step S500, the hyperspectral image data and auxiliary information are input into the model, and the model is allowed to make predictions to obtain the calculation results of the karst carbon sink; Select a data visualization platform, integrate the hyperspectral image data and karst carbon sink data, and display them simultaneously on the data visualization platform; Use data visualization platforms to create maps, display hyperspectral images as basemaps, and adjust the color, contrast, and transparency of images to highlight different features. Overlay karst carbon sink data on the map and use different colors, symbols or charts to represent the size and distribution of the carbon sink.

8. The method for calculating hyperspectral karst carbon sink according to claim 7, characterized in that: Use the data visualization platform to conduct spatial analysis to understand the spatial distribution characteristics of karst carbon sinks; Conduct spatial interpolation analysis to estimate the carbon sink value in unknown areas based on the carbon sink data at known points; Compare and analyze the carbon sink results in different regions or at different times to understand the differences and changes in karst carbon sinks, and use charts and maps to intuitively display the carbon sink situation under different circumstances.

Images