Wetland carbon cycle monitoring and collecting system based on remote sensing technology

Abstract

The invention provides a wetland carbon cycle monitoring and acquisition system based on a remote sensing technology, and relates to the technical field of environment monitoring and data processing, and the system comprises a multi-source remote sensing data acquisition module which is used for collecting multi-source remote sensing data of a target wetland area; the data preprocessing module is used for preprocessing the collected multi-source remote sensing data; the carbon cycle parameter extraction module is used for extracting wetland carbon cycle related parameters based on the preprocessed data; the carbon cycle dynamic monitoring module is used for constructing a wetland carbon cycle dynamic model based on the wetland carbon cycle related parameters so as to analyze dynamic changes of wetland carbon sink and carbon emission and obtain monitoring data; the data storage and visualization module is used for storing the monitoring data and providing visual display; and the intelligent decision support module is used for providing decision support for wetland protection and carbon management based on the monitoring data. According to the method, the efficiency and accuracy of wetland carbon cycle monitoring can be remarkably improved, and powerful technical support is provided for wetland protection and sustainable management.

Description

Technical Field

[0001] The invention relates to the technical field of environmental monitoring and data processing, and in particular to a wetland carbon cycle monitoring and collection system based on remote sensing technology. Background Art

[0002] Wetlands are an important part of the global carbon cycle, and their carbon sink function is of great significance to climate change mitigation. However, the dynamic changes of wetland carbon cycle are complex, and traditional ground monitoring methods have limited coverage and low efficiency. Remote sensing technology has become an important means of wetland carbon cycle monitoring due to its large-scale and high-timeliness characteristics.

[0003] In existing technologies, the remote sensing data collection, processing and analysis systems are still imperfect, making it difficult to achieve efficient and accurate monitoring of wetland carbon cycles. Summary of the invention

[0004] In order to overcome the shortcomings of the prior art, the purpose of the present invention is to provide a wetland carbon cycle monitoring and collection system based on remote sensing technology, which significantly improves the efficiency and accuracy of wetland carbon cycle monitoring through efficient data collection, accurate data processing, comprehensive parameter extraction, scientific dynamic monitoring and analysis, intuitive data visualization and intelligent decision support, and provides strong technical support for wetland protection and sustainable management.

[0005] To achieve the above object, the present invention provides the following solutions:

[0006] A wetland carbon cycle monitoring and collection system based on remote sensing technology, including:

[0007] A multi-source remote sensing data acquisition module is used to collect multi-source remote sensing data of the target wetland area; the multi-source remote sensing data includes: optical images, thermal infrared images and radar images;

[0008] A data preprocessing module, used for preprocessing the collected multi-source remote sensing data to obtain preprocessed data;

[0009] A carbon cycle parameter extraction module, used for extracting wetland carbon cycle related parameters based on the preprocessed data;

[0010] A carbon cycle dynamic monitoring module is used to construct a wetland carbon cycle dynamic model based on the wetland carbon cycle related parameters to analyze the dynamic changes of wetland carbon sinks and carbon emissions and obtain monitoring data;

[0011] A data storage and visualization module, used to store the monitoring data and provide a visual display;

[0012] An intelligent decision support module is used to provide decision support for wetland protection and carbon management based on the monitoring data.

[0013] Preferably, the data preprocessing module includes:

[0014] A radiation correction unit, used to correct the optical image by an absolute radiation correction method, and correct the sensor noise and atmospheric absorption effect of the thermal infrared image to restore the surface temperature distribution, and correct the system noise and scattering effect of the radar image to restore the surface backscattering coefficient, so as to obtain the optical image, thermal infrared image and radar image after radiation correction;

[0015] An atmospheric correction unit, used to perform atmospheric correction on the optical image after radiation correction based on a preset atmospheric parameter correction method, correct the atmospheric radiation effect of the thermal infrared image after radiation correction, and correct the influence of atmospheric humidity on radar wave propagation of the radar image after radiation correction, so as to obtain the optical image, thermal infrared image and radar image after atmospheric correction;

[0016] The geometric correction unit is used to perform geometric correction on the optical image after atmospheric correction based on a geometric correction method of polynomial fitting, and to correct the radar image after atmospheric correction using an orthorectification method to eliminate the influence of terrain undulation and obtain the preprocessed data.

[0017] Preferably, the data preprocessing module further includes:

[0018] The image stitching unit is used to stitch the various pre-processed data to obtain complete pre-processed data covering the target wetland area.

[0019] Preferably, the image stitching unit comprises:

[0020] A weight calculation unit, used to calculate the weight of each of the pre-processed data at each pixel position;

[0021] A weighted fusion unit, used for fusing each individual pre-processed data into a complete pre-processed data according to a weight;

[0022] The edge smoothing unit is used to perform edge smoothing on the complete preprocessed data to eliminate the sudden change in the seam area.

[0023] Preferably, the calculation formula of the weight is: Wherein, k represents the type of image corresponding to the preprocessed data, wherein when k is 1, it means that the preprocessed data corresponds to an optical image, when k is 2, it means that the preprocessed data corresponds to a thermal infrared image, and when k is 3, it means that the preprocessed data corresponds to a radar image; represents the weight of the i-th image of type k at pixel (x, y), α k Assign weights to the importance of presets, is the spatial distance weight, is the spectral consistency weight, is the edge smoothing weight; is the average value of all images at pixel (x, y), γ is the parameter that controls the influence of spectral difference on weight, is the gradient amplitude of the i-th image, d i (x,y) is the distance from pixel (x,y) to the center of the i-th image, and σ is the parameter that controls the weight decay speed.

[0024] Preferably, the calculation formula for the complete preprocessed data is: in, is the pixel value of the i-th image, is the weight of the i-th image at pixel (x, y), N is the number of images involved in stitching, It is the pixel value of the complete pre-processed data at pixel (x, y) after stitching.

[0025] Preferably, the wetland carbon cycle related parameters include vegetation index, soil moisture and surface temperature.

[0026] Preferably, the wetland carbon cycle dynamic model is obtained by training a CNN-LSTM combined network.

[0027] The present invention discloses the following technical effects:

[0028] The present invention provides a wetland carbon cycle monitoring and collection system based on remote sensing technology, including: a multi-source remote sensing data collection module, used to collect multi-source remote sensing data of a target wetland area; the multi-source remote sensing data includes: optical images, thermal infrared images and radar images; a data preprocessing module, used to preprocess the collected multi-source remote sensing data to obtain preprocessed data; a carbon cycle parameter extraction module, used to extract wetland carbon cycle related parameters based on the preprocessed data; a carbon cycle dynamic monitoring module, used to construct a wetland carbon cycle dynamic model based on the wetland carbon cycle related parameters to analyze the dynamic changes of wetland carbon sinks and carbon emissions to obtain monitoring data; a data storage and visualization module, used to store the monitoring data and provide visual display; an intelligent decision support module, used to provide decision support for wetland protection and carbon management based on the monitoring data. The present invention significantly improves the efficiency and accuracy of wetland carbon cycle monitoring through efficient data collection, accurate data processing, comprehensive parameter extraction, scientific dynamic monitoring and analysis, intuitive data visualization and intelligent decision support, and provides strong technical support for wetland protection and sustainable management. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0030] Figure 1 A schematic diagram of the system structure provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0031] 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.

[0032] The purpose of the present invention is to provide a wetland carbon cycle monitoring and collection system based on remote sensing technology, which significantly improves the efficiency and accuracy of wetland carbon cycle monitoring through efficient data collection, accurate data processing, comprehensive parameter extraction, scientific dynamic monitoring and analysis, intuitive data visualization and intelligent decision support, and provides strong technical support for wetland protection and sustainable management.

[0033] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0034] Figure 1 A schematic diagram of the system structure provided by an embodiment of the present invention, such as Figure 1 As shown, the present invention provides a wetland carbon cycle monitoring and collection system based on remote sensing technology, comprising:

[0035] A multi-source remote sensing data acquisition module is used to collect multi-source remote sensing data of the target wetland area; the multi-source remote sensing data includes: optical images, thermal infrared images and radar images;

[0036] A data preprocessing module, used for preprocessing the collected multi-source remote sensing data to obtain preprocessed data;

[0037] A carbon cycle parameter extraction module, used for extracting wetland carbon cycle related parameters based on the preprocessed data;

[0038] A carbon cycle dynamic monitoring module is used to construct a wetland carbon cycle dynamic model based on the wetland carbon cycle related parameters to analyze the dynamic changes of wetland carbon sinks and carbon emissions and obtain monitoring data;

[0039] A data storage and visualization module, used to store the monitoring data and provide a visual display;

[0040] An intelligent decision support module is used to provide decision support for wetland protection and carbon management based on the monitoring data.

[0041] Specifically, the multi-source remote sensing data acquisition module of this embodiment is composed of multiple subsystems, including a satellite remote sensing system, an unmanned aerial vehicle remote sensing system, and a ground sensor network. The satellite remote sensing system uses high-resolution satellites (such as Sentinel-2, Landsat, etc.) to regularly acquire a wide range of optical images and thermal infrared images, which can cover a wide range of wetland areas and provide long-term series data. The unmanned aerial vehicle remote sensing system uses high-resolution cameras and sensors to fly at low altitudes to obtain higher-resolution image data, which is suitable for small-scale wetland monitoring. The ground sensor network uses sensors deployed in the wetland to monitor environmental parameters such as soil moisture, temperature, and gas concentration in real time to supplement the deficiencies in remote sensing data.

[0042] During the data collection process, the multi-source remote sensing data acquisition module first selects the appropriate remote sensing platform and sensor according to the preset monitoring plan and wetland characteristics. The satellite remote sensing system automatically acquires image data according to a fixed orbit and time interval, and the UAV system autonomously flies according to mission requirements to collect image and sensor data in real time. The ground sensor transmits the real-time monitoring data to the central database via a wireless network. All collected data are initially screened and formatted to ensure data integrity and consistency, and then stored in the data management system for subsequent processing and analysis.

[0043] In order to improve the accuracy and comprehensiveness of monitoring, the multi-source remote sensing data acquisition module also has data fusion and integration functions. By aligning data from different sources in time and space, the system can effectively integrate satellite images, drone images and ground sensor data. Using advanced data fusion algorithms (such as weighted average, principal component analysis, etc.), data of different resolutions and types are comprehensively processed to generate high-quality comprehensive images and environmental parameters. The data fusion of this embodiment not only improves the accuracy of wetland carbon cycle monitoring, but also provides a rich information basis for subsequent carbon cycle parameter extraction and dynamic monitoring.

[0044] Preferably, the data preprocessing module includes:

[0045] A radiation correction unit, used to correct the optical image by an absolute radiation correction method, and correct the sensor noise and atmospheric absorption effect of the thermal infrared image to restore the surface temperature distribution, and correct the system noise and scattering effect of the radar image to restore the surface backscattering coefficient, so as to obtain the optical image, thermal infrared image and radar image after radiation correction;

[0046] The main task of the radiation correction unit is to eliminate the radiation errors caused by sensor characteristics and observation conditions in remote sensing images. For optical images, the absolute radiation correction method is used to correct them by using the ground-measured radiation values ​​(such as reflectivity). The specific steps include: first, collecting ground observation data, establishing a radiation transfer model, and calculating the true reflectivity of each pixel. For thermal infrared images, the correction process involves the correction of sensor noise and atmospheric absorption effects, and using an atmospheric radiation transfer model (such as MODTRAN) to restore the surface temperature distribution. The correction of radar images analyzes system noise and scattering effects, and uses known ground features (such as surface roughness) to restore the surface backscattering coefficient. After these processes, radiation-corrected optical images, thermal infrared images, and radar images are obtained, providing high-quality input data for subsequent processing.

[0047] The atmospheric correction unit is used to perform atmospheric correction on the optical image after radiation correction based on the correction method of preset atmospheric parameters, and correct the atmospheric radiation effect of the thermal infrared image after radiation correction, and correct the influence of atmospheric humidity on radar wave propagation of the radar image after radiation correction, so as to obtain the optical image, thermal infrared image and radar image after atmospheric correction; the function of the atmospheric correction unit is to further eliminate the influence of the atmosphere on the remote sensing image. The unit first performs atmospheric correction on the optical image after radiation correction according to the preset atmospheric parameters (such as aerosol optical thickness, humidity, temperature, etc.). In the specific implementation, the atmospheric correction algorithm (such as 6S model) is used to calculate the influence of the atmosphere on the optical image, and make corresponding corrections. For thermal infrared images, the correction process includes eliminating the atmospheric radiation effect to ensure the accuracy of the restored surface temperature. The atmospheric correction of radar images focuses on the influence of atmospheric humidity on radar wave propagation, and corrects the propagation path of radar signals through model calculation to ensure the accuracy of radar images. Finally, the optical images, thermal infrared images and radar images processed by atmospheric correction will provide more accurate basic data for subsequent geometric correction.

[0048] The geometric correction unit is used to perform geometric correction on the optical image after atmospheric correction based on the geometric correction method of polynomial fitting, and to correct the radar image after atmospheric correction using the orthorectification method to eliminate the influence of terrain undulation on the image, so as to obtain the preprocessed data. The main task of the geometric correction unit is to eliminate the geometric distortion in the remote sensing image and ensure the accurate matching of the image with the geographic coordinate system. The unit first performs geometric correction on the optical image after atmospheric correction, and uses the polynomial fitting method to establish the relationship between the image and the geographic coordinates by selecting the ground control points (GCP). The specific steps include: extracting feature points in the image, matching them with the ground control points of known coordinates, and using the polynomial model to perform geometric transformation to eliminate the geometric distortion of the image. For radar images, the orthorectification method is used in combination with the digital elevation model (DEM) to eliminate the influence of terrain undulation on the image. Through these correction steps, the final preprocessed data will have high spatial accuracy and consistency, providing reliable basic data for subsequent carbon cycle parameter extraction and dynamic monitoring.

[0049] Preferably, the data preprocessing module further includes:

[0050] The image stitching unit is used to stitch the various pre-processed data to obtain complete pre-processed data covering the target wetland area.

[0051] Preferably, the image stitching unit comprises:

[0052] A weight calculation unit, used to calculate the weight of each of the pre-processed data at each pixel position;

[0053] A weighted fusion unit, used for fusing each individual pre-processed data into a complete pre-processed data according to a weight;

[0054] The edge smoothing unit is used to perform edge smoothing on the complete preprocessed data to eliminate the sudden change in the seam area.

[0055] Preferably, the calculation formula of the weight is: Wherein, k represents the type of image corresponding to the preprocessed data, wherein when k is 1, it means that the preprocessed data corresponds to an optical image, when k is 2, it means that the preprocessed data corresponds to a thermal infrared image, and when k is 3, it means that the preprocessed data corresponds to a radar image; represents the weight of the i-th image of type k at pixel (x, y), α k Assign weights to the importance of presets, is the spatial distance weight, is the spectral consistency weight, is the edge smoothing weight; is the average value of all images at pixel (x, y), γ is the parameter that controls the influence of spectral difference on weight, is the gradient amplitude of the i-th image, d i (x,y) is the distance from pixel (x,y) to the center of the i-th image, and σ is the parameter that controls the weight decay speed.

[0056] Preferably, the calculation formula for the complete preprocessed data is: in, is the pixel value of the i-th image, is the weight of the i-th image at pixel (x, y), N is the number of images involved in stitching, It is the pixel value of the complete pre-processed data at pixel (x, y) after stitching.

[0057] Specifically, based on previous research and experience, an initial value σ is selected. For most wetland remote sensing images, σ can be set to a proportion of the image resolution (such as 1 / 2 or 1 / 3 of the image resolution) to ensure that the weight in the center area of ​​the image is higher, while the weight in the edge area gradually decreases. In this way, the stitching effect can be observed in preliminary experiments and adjusted according to the results. During the stitching process, an adaptive method is used to dynamically adjust the value of γ. Specifically, the spectral difference of each pixel can be calculated in real time according to the spectral distribution of the image, and γ can be adjusted according to the size of the difference. When the spectral difference is small, the value of γ is increased to increase the weight; when the spectral difference is large, the value of γ is reduced to reduce the weight. The method of this embodiment can more flexibly cope with the spectral changes between different images and improve the quality of the stitching results.

[0058] Furthermore, the preset importance allocation weights α of various types of images in this embodiment are k The sum is 1. For example, in this embodiment, α1 of the optical image is 0.5, α2 of the thermal infrared image is 0.3, and α3 of the radar image is 0.2.

[0059] Preferably, the wetland carbon cycle related parameters include vegetation index, soil moisture and surface temperature.

[0060] Specifically, the vegetation index of this embodiment is the normalized difference vegetation index NDVI, which is an important indicator for evaluating the growth status and carbon absorption capacity of wetland vegetation. The calculation of NDVI is based on the reflectivity of the red and near-infrared bands of optical images. Soil moisture is an important parameter of wetland carbon cycle, which directly affects plant growth and carbon storage. This module extracts soil moisture by analyzing thermal infrared images and ground sensor data. In the specific implementation, the surface temperature data in the thermal infrared image is combined with the soil heat conduction model to calculate the moisture content of the soil. Commonly used soil moisture estimation methods include models based on the temperature-humidity relationship and regression analysis. By regressing the temperature data of the thermal infrared image with the soil moisture data measured on the ground, a relationship model between soil moisture and temperature is established, thereby achieving accurate estimation of wetland soil moisture. Surface temperature is an important parameter for evaluating the health of wetland ecosystems and carbon cycle, and this embodiment extracts it through thermal infrared images. The carbon cycle parameter extraction module uses the thermal infrared image after radiation correction, and applies an atmospheric correction model (such as MODTRAN) to eliminate the atmospheric impact and restore the surface temperature. The specific steps include: first, converting thermal infrared images into brightness temperature, and then calculating the actual surface temperature based on the atmospheric correction model. The system can also be verified and corrected in conjunction with data from ground meteorological stations to improve the accuracy of surface temperature. Ultimately, the extracted surface temperature data will provide an important basis for the dynamic monitoring and analysis of wetland carbon cycles together with vegetation indices and soil moisture.

[0061] Preferably, the wetland carbon cycle dynamic model is obtained by training a CNN-LSTM combined network.

[0062] Specifically, the wetland carbon cycle dynamic model of this embodiment is trained using a CNN-LSTM combined network to make full use of the spatial characteristics and time series characteristics of remote sensing data. First, the input of the model includes extracted wetland carbon cycle related parameters, such as vegetation index, soil moisture and surface temperature, etc., which are usually organized in the form of time series. The CNN (convolutional neural network) part is responsible for extracting the spatial characteristics of the input data, and extracts features from the input multidimensional data through multiple convolutional layers and pooling layers to capture the spatial distribution characteristics of the wetland. Then, the extracted features are passed to the LSTM (long short-term memory network) part, which can process time series data and learn the dynamic patterns of wetland carbon cycle parameters changing over time. During the training process, historical monitoring data is used as a training set, and the model parameters are optimized through the back propagation algorithm to minimize the error between the predicted value and the actual observed value.

[0063] After training, the wetland carbon cycle dynamic model can analyze the dynamic changes of wetland carbon sinks and carbon emissions in real time. The model first receives new wetland carbon cycle related parameters as input, which can be real-time monitoring data or prediction data. The CNN part extracts spatial features of the input data again, generates feature maps, and then passes these feature maps to the LSTM part. The LSTM predicts future carbon sinks and carbon emissions based on historical data and current inputs. The monitoring data output by the model includes the changing trend of wetland carbon sinks, the dynamic changes of carbon emissions, and possible abnormal situations. These results provide a scientific basis for wetland management and protection, and help decision makers formulate corresponding ecological protection measures and carbon management strategies.

[0064] Furthermore, the data storage and visualization module is responsible for managing and displaying the monitoring data obtained from the carbon cycle dynamic monitoring module. The monitoring data include wetland carbon cycle related parameters (such as vegetation index, soil moisture, surface temperature), dynamic changes of carbon sinks and carbon emissions, etc. These data are first stored in an efficient database to ensure the security and accessibility of the data. The module adopts a combination of structured and unstructured data storage to facilitate subsequent query and analysis. At the same time, the module provides a user-friendly visualization interface, which displays the monitoring data to users in an intuitive way through charts, maps and dashboards, helping users quickly understand the status and changing trends of wetland ecosystems.

[0065] The intelligent decision support module uses the stored monitoring data to provide scientific decision-making basis for wetland protection and carbon management. The module analyzes the monitoring data, identifies the health status and potential risks of the wetland ecosystem, and generates corresponding management recommendations. For example, based on the dynamic changes of carbon sinks and carbon emissions, the module can evaluate the carbon storage capacity of wetlands and propose corresponding protection measures. In addition, the module can also combine machine learning algorithms to predict future wetland change trends and help decision makers formulate long-term management strategies. By interacting with users, the module can dynamically adjust the content and form of decision support according to user needs and feedback to ensure that the recommendations provided are targeted and practical.

[0066] The beneficial effects of the present invention are as follows:

[0067] (1) By integrating optical images, thermal infrared images and radar images, the system can comprehensively obtain multi-dimensional information of wetlands, improving the richness and accuracy of data. Remote sensing technology can cover a wide range of wetland areas, overcome the limitations of traditional ground monitoring methods, and reduce manpower and time costs.

[0068] (2) The present invention ensures the accuracy and consistency of the data through preprocessing steps such as radiation correction, atmospheric correction and geometric correction, providing a reliable basis for subsequent analysis; the present invention effectively removes sensor noise and environmental interference, improves the signal-to-noise ratio of the data, and enhances the credibility of the monitoring results.

[0069] (3) The present invention can extract a variety of wetland carbon cycle related parameters, such as vegetation net primary productivity (NPP), soil organic carbon content and greenhouse gas emissions, and provide comprehensive carbon cycle information; through time series analysis, the system can monitor the dynamic changes of wetland carbon cycle and timely reflect the health status of wetland ecosystems.

[0070] (4) Based on the extracted carbon cycle parameters, the system can construct a dynamic model of wetland carbon cycle, analyze the changing trends of carbon sinks and carbon emissions, and support scientific research and policy making; it can timely identify abnormal changes in wetland carbon cycle and provide early warning for ecological protection and management.

[0071] (5) Through the data storage and visualization module, the system of the present invention can efficiently manage and store a large amount of remote sensing data, which is convenient for subsequent analysis and query, and provides a user-friendly visualization interface, making the monitoring results easy to understand and disseminate, and supporting the participation of decision makers and the public.

[0072] (6) Based on monitoring data, the system of the present invention can provide scientific decision-making support for wetland protection and carbon management, help formulate effective management strategies, provide data support for governments and relevant agencies, promote the formulation and implementation of wetland protection policies, and promote sustainable development.

[0073] (7) Through monitoring and visualization, the present invention can enhance the public and decision makers’ awareness of the importance of wetland ecosystems, promote ecological protection actions, provide data support for wetland ecological restoration, help identify priority restoration areas, and optimize resource allocation.

[0074] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0075] The principles and implementation methods of the present invention are described in this article using specific examples. The description of the above embodiments is only used to help understand the method and core idea of ​​the present invention. At the same time, for those skilled in the art, according to the idea of ​​the present invention, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting the present invention.

Claims

1. A wetland carbon cycle monitoring and collection system based on remote sensing technology, characterized in that: include: Multi-source remote sensing data acquisition module, used to collect multi-source remote sensing data of target wetland areas; The multi-source remote sensing data includes: optical images, thermal infrared images and radar images; A data preprocessing module, used for preprocessing the collected multi-source remote sensing data to obtain preprocessed data; A carbon cycle parameter extraction module, used to extract wetland carbon cycle related parameters based on the preprocessed data; A carbon cycle dynamic monitoring module is used to construct a wetland carbon cycle dynamic model based on the wetland carbon cycle related parameters to analyze the dynamic changes of wetland carbon sinks and carbon emissions and obtain monitoring data; A data storage and visualization module, used to store the monitoring data and provide a visual display; An intelligent decision support module is used to provide decision support for wetland protection and carbon management based on the monitoring data.

2. The wetland carbon cycle monitoring and collection system based on remote sensing technology according to claim 1 is characterized in that: The data preprocessing module comprises: A radiation correction unit, used to correct the optical image by an absolute radiation correction method, and correct the sensor noise and atmospheric absorption effect of the thermal infrared image to restore the surface temperature distribution, and correct the system noise and scattering effect of the radar image to restore the surface backscattering coefficient, so as to obtain the optical image, thermal infrared image and radar image after radiation correction; An atmospheric correction unit, used to perform atmospheric correction on the optical image after radiation correction based on a preset atmospheric parameter correction method, correct the atmospheric radiation effect of the thermal infrared image after radiation correction, and correct the influence of atmospheric humidity on radar wave propagation of the radar image after radiation correction, so as to obtain the optical image, thermal infrared image and radar image after atmospheric correction; The geometric correction unit is used to perform geometric correction on the optical image after atmospheric correction based on a geometric correction method of polynomial fitting, and to correct the radar image after atmospheric correction using an orthorectification method to eliminate the influence of terrain undulation and obtain the preprocessed data.

3. The wetland carbon cycle monitoring and collection system based on remote sensing technology according to claim 1 is characterized in that: The data preprocessing module also includes: The image stitching unit is used to stitch the various pre-processed data to obtain complete pre-processed data covering the target wetland area.

4. The wetland carbon cycle monitoring and collection system based on remote sensing technology according to claim 3 is characterized in that: The image stitching unit comprises: A weight calculation unit, used to calculate the weight of each of the pre-processed data at each pixel position; A weighted fusion unit, used for fusing each individual pre-processed data into a complete pre-processed data according to a weight; The edge smoothing unit is used to perform edge smoothing on the complete preprocessed data to eliminate the sudden change in the seam area.

5. The wetland carbon cycle monitoring and collection system based on remote sensing technology according to claim 4 is characterized in that: The calculation formula of the weight is: Wherein, k represents the type of image corresponding to the preprocessed data, wherein when k is 1, it means that the preprocessed data corresponds to an optical image, when k is 2, it means that the preprocessed data corresponds to a thermal infrared image, and when k is 3, it means that the preprocessed data corresponds to a radar image; represents the weight of the i-th image of type k at pixel (x, y), α k Assign weights to the importance of presets, is the spatial distance weight, is the spectral consistency weight, is the edge smoothing weight; is the average value of all images at pixel (x, y), γ is the parameter that controls the influence of spectral difference on weight, G i (x,y) is the gradient amplitude of the i-th image, d i (x,y) is the distance from pixel (x,y) to the center of the i-th image, and σ is the parameter that controls the weight decay speed.

6. The wetland carbon cycle monitoring and collection system based on remote sensing technology according to claim 5 is characterized in that: The calculation formula for the complete preprocessed data is: in, is the pixel value of the i-th image, is the weight of the i-th image at pixel (x, y), N is the number of images involved in stitching, It is the pixel value of the complete pre-processed data at pixel (x, y) after stitching.

7. The wetland carbon cycle monitoring and collection system based on remote sensing technology according to claim 1 is characterized in that: The wetland carbon cycle related parameters include vegetation index, soil moisture and surface temperature.

8. The wetland carbon cycle monitoring and collection system based on remote sensing technology according to claim 1 is characterized in that: The wetland carbon cycle dynamic model is obtained by training the CNN-LSTM combined network.

Images