A method, system, device, medium and product for salt marsh vegetation classification
By generating vegetation index time series and constructing a random forest model, the problems of low efficiency and insufficient accuracy in salt marsh vegetation classification are solved, achieving low-cost and efficient salt marsh vegetation identification, which is suitable for large-scale classification.
Patent Information
- Application Number
- CN202411002345.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-24
- Publication Date
- 2025-11-04
- Estimated Expiration
- 2044-07-24
AI Technical Summary
Existing technologies are inefficient and costly when classifying salt marsh vegetation on a large scale. Furthermore, algorithms that rely on empirical judgment lack accuracy in long-term and large-scale spatial classification, making it difficult to effectively identify the specific types of coastal salt marsh vegetation.
By generating time series of vegetation indices, calculating the growth and decay periods using the double logistic function, and combining JM distance and random forest models, the vegetation indices with the most separable window periods are selected to construct a salt marsh vegetation classification model. The model is then efficiently classified using Landsat-8 and Sentinel-2 satellite data.
It achieves low-cost and efficient classification of salt marsh vegetation, improves the accuracy and scalability of large-scale salt marsh vegetation identification, and is applicable to salt marsh vegetation classification nationwide.
Smart Images
Figure CN119249257B_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The present application relates to the field of salt marsh vegetation classification, and particularly relates to a salt marsh vegetation classification method, system, device, medium and product. BACKGROUND
[0002] The coastal salt marsh wetland is located in the land-sea transition zone, mainly refers to the coastal line affected by the periodic or intermittent influence of the sea tide, the saline plant covered brackish water or brackish water muddy beach, and is called as a "blue carbon" ecological system together with the mangrove forest and the seagrass bed, and has a strong carbon sequestration capacity. However, the current method for classifying the coastal salt marsh vegetation has many defects. Limited by the complex environment of the coastal salt marsh wetland, the low-efficiency manual field investigation is often insufficient to carry out large-scale salt marsh vegetation investigation on the vast coastal zone in China, and the lack of samples leads to that most of the encapsulated general classification algorithms do not include the identification of specific types of coastal salt marsh vegetation. At present, most of the large-scale classification algorithms for salt marsh vegetation are threshold classification methods based on expert experience. Such algorithms which seriously depend on experience are not only poor in scalability in long-term classification, but also have the problem of insufficient accuracy in large-scale spatial classification. Although some studies attempt to use high-resolution, hyperspectral and radar data for classification, the availability and processing difficulty of such data are far less than that of mainstream remote sensing satellites such as Landsat and Sentinel, and it is difficult to carry out efficient and low-cost classification on a large scale. SUMMARY
[0003] The purpose of the present application is to provide a salt marsh vegetation classification method, system, device, medium and product, which can solve the problem that it is difficult to carry out efficient and low-cost classification on a large scale.
[0004] To achieve the above purpose, the present application provides the following solutions.
[0005] In a first aspect, the present application provides a salt marsh vegetation classification method, comprising:
[0006] generating a vegetation index time series according to satellite remote sensing data; the vegetation index time series includes a normalized difference vegetation index time series, an enhanced vegetation index time series, a normalized difference water index time series, and a corrected red edge chlorophyll vegetation index time series;
[0007] dividing the coastal zone latitude into several classification latitude bands as classification intervals, and calculating the monthly mean value index and the monthly variance index of each vegetation index in each classification interval, and the annual mean value index and the annual variance index of each vegetation index in the current year;
[0008] Based on the normalized difference vegetation index time series and the enhanced vegetation index time series, a growth period and a decline period of vegetation are calculated as candidate classification indexes by using a double logistic function;
[0009] In combination with different vegetation classification samples, J-M distances of each candidate classification index on different vegetation classification samples are calculated; each candidate index includes a monthly mean value index, a monthly variance index, an annual mean value index, an annual variance index, a growth period index, and a decline period index;
[0010] Based on the J-M distances of each candidate index on different vegetation classification samples and a preliminary classification contribution evaluation, a classification index with the most degree of separability is screened out, and a final random forest classification model is constructed;
[0011] According to the final random forest classification model, salt marsh vegetation in a coastal salt marsh wetland is classified.
[0012] Optionally, based on the J-M distance and the above indexes, a window period vegetation index with the most degree of separability is screened out, and a random forest model is constructed, specifically including:
[0013] Indexes with J-M distance rankings in the top 50% are extracted as preliminary classification variables of the random forest model;
[0014] The contribution of each index to the random forest classification model is calculated, and the contribution is sorted;
[0015] The indexes with the top 10 contribution degrees are taken as the window period vegetation index with the most degree of separability;
[0016] The window period vegetation index with the most degree of separability is taken as an independent variable of a final random forest classification model, and the final random forest classification model is constructed.
[0017] Optionally, according to the satellite remote sensing data to generate the vegetation index time series, previously also includes: the satellite remote sensing data includes Landsat-8 image data and Sentinel-2 image data;
[0018] The pixels covered by clouds and shadows in the Landsat-8 image data are removed by using a QA band of Landsat-8, and the removed Landsat-8 image data is generated;
[0019] Pixels covered by clouds and shadows in the Sentinel-2 image data are removed by using a sentinel-2 cloud detector library and a S2_CLOUD_PROBABILITY dataset, and the removed Sentinel-2 image data is generated;
[0020] Taking the removed Sentinel-2 image data as a benchmark, the least square method is used to correct the difference between the removed Landsat-8 image data and the removed Sentinel-2 image data, and the optimal satellite remote sensing data is determined.
[0021] Optionally, the QA band of Landsat-8 is used to remove the pixels covered by clouds and shadows in the Landsat-8 image data to generate removed Landsat-8 image data, specifically including:
[0022] The QA band of Landsat-8 is used to determine the pixel occlusion information of the pixels covered by clouds and shadows in the Landsat-8 image data in a decimal manner;
[0023] The pixel occlusion information is converted into binary, and the pixel quality is determined based on the number of bits corresponding to the binary pixel occlusion information;
[0024] Based on the pixel quality, a mask function is used to remove the pixels covered by clouds and shadows in the Landsat-8 image data to generate removed Landsat-8 image data.
[0025] Optionally, the sentinel-2 cloud detector library and S2_CLOUD_PROBABILITY dataset are used to remove the pixels covered by clouds and shadows in the Sentinel-2 image data to generate removed Sentinel-2 image data, specifically including:
[0026] Based on the sentinel-2 cloud detector library, the pixel coverage probability of the pixels covered by clouds and shadows in the Sentinel-2 image data is determined according to the S2_CLOUD_PROBABILITY dataset;
[0027] Based on the pixel coverage probability, a filtering function is used to remove the pixels covered by clouds and shadows in the Sentinel-2 image data to generate removed Sentinel-2 image data.
[0028] Optionally, a vegetation index time series is generated according to satellite remote sensing data, specifically including:
[0029] The salt-and-pepper noise and extreme values appearing in the satellite remote sensing data collected in a set period are smoothed by using a Savitzky-Golay filtering algorithm, and the Sentinel-2 image data in the satellite remote sensing data are resampled by using a cubic convolution, so that the resampled Sentinel-2 data are consistent with the spatial resolution of the Landsat-8 image data in the satellite remote sensing data, and a vegetation index time series is generated; the vegetation index time series is displayed as a vegetation index annual characteristic distribution map.
[0030] In a second aspect, a salt marsh vegetation classification system comprises:
[0031] A vegetation index time series generation module is configured to generate a vegetation index time series according to satellite remote sensing data; the vegetation index time series comprises a normalized difference vegetation index time series, an enhanced vegetation index time series, a normalized difference water index time series, and a corrected red edge chlorophyll vegetation index time series.
[0032] A monthly and annual index calculation module is configured to divide a coastal zone latitude into a plurality of classification latitude bands as classification intervals, and calculate a monthly mean value index and a monthly variance index of each vegetation index in each classification interval, and an annual mean value index and an annual variance index of each vegetation index in each classification interval.
[0033] A growth period candidate index and a decline period candidate index calculation module is configured to calculate a growth period and a decline period of vegetation as candidate classification indexes by using a double logistic function based on the normalized difference vegetation index time series and the enhanced vegetation index time series.
[0034] A J-M distance calculation module is configured to calculate a J-M distance of each candidate classification index on different vegetation classification samples in combination with different vegetation classification samples; each candidate index comprises a monthly mean value index, a monthly variance index, an annual mean value index, an annual variance index, a growth period index, and a decline period index.
[0035] Based on the J-M distance of each candidate index on different vegetation classification samples and a preliminary classification contribution evaluation, the most separable classification index is screened out to construct a final random forest classification model.
[0036] A classification module is configured to classify salt marsh vegetation in a coastal salt marsh wetland according to the final random forest classification model.
[0037] In a third aspect, the present application provides a computer device, comprising a memory, a processor, a computer program stored in the memory and executable on the processor, and the processor executes the computer program to implement the steps of the salt marsh vegetation classification method in any one of the above aspects.
[0038] In a fourth aspect, the present application provides a computer readable storage medium, having stored thereon a computer program, which, when executed by a processor, implements the steps of the salt marsh vegetation classification method according to any one of the preceding aspects.
[0039] In a fifth aspect, the present application provides a computer program product comprising a computer program which, when executed by a processor, implements the steps of the salt marsh vegetation classification method according to any one of the preceding aspects.
[0040] According to the specific embodiments provided by the present application, the present application discloses the following technical effects: the present application generates a vegetation index time series using satellite remote sensing data, then synthesizes and smooths the monthly indicators and the current indicators of multiple vegetation indices, and calculates the growth period indicators and the decline period indicators of the vegetation using a double logistic function, then combines different vegetation classification samples, selects the most separable window period vegetation index according to the J-M distance, constructs a random forest model, and finally classifies the salt marsh vegetation in the coastal salt marsh wetland using the constructed random forest model, thereby completing the coastal plant classification task at low cost and high efficiency. BRIEF DESCRIPTION OF DRAWINGS
[0041] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings needed in the embodiments will be briefly introduced as follows. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can also be obtained by those skilled in the art without any creative effort on the basis of these drawings.
[0042] Figure 1 A flow chart of a salt marsh vegetation classification method according to an embodiment of the present application;
[0043] Figure 2 A distribution map of annual characteristics of key plant indices of different vegetation types according to an embodiment of the present application; Figure 2 (a) in the figure is a NDVI sequence diagram, Figure 2 (b) in the figure is an EVI sequence diagram, Figure 2 (c) in the figure is a NDWI sequence diagram; Figure 2 (d) in the figure is a MRECI sequence diagram;
[0044] Figure 3 A technical line chart of a salt marsh vegetation classification method according to an embodiment of the present application;
[0045] Figure 4 A distribution and succession characteristic map of Chinese salt marsh wetland vegetation according to an embodiment of the present application; Figure 4 (a) in the figure is a latitude distribution map of the average number of changes in surface coverage of the salt marsh wetland in the past six years, Figure 4(b) is a latitude distribution map of the proportion of change in the surface coverage of salt marshes in the past six years, Figure 4 (c) is a schematic diagram of the Liaohe Estuary, Figure 4 (d) is a schematic diagram of the Yellow River Estuary, Figure 4 (e) is a schematic diagram of the Yangtze River Estuary, Figure 4 (f) is a schematic diagram of the Pearl River Estuary. DETAILED DESCRIPTION
[0046] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, but not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work fall within the scope of protection of the present application.
[0047] The above purposes, features and advantages of the present application can be more obvious and easy to understand. The present application will be described in further detail below with reference to the drawings and specific embodiments.
[0048] In an exemplary embodiment, as shown in Figure 1 a salt marsh vegetation classification method is provided, comprising:
[0049] Step 101: generating a vegetation index time series according to satellite remote sensing data; the vegetation index time series includes a normalized difference vegetation index time series, an enhanced vegetation index time series, a normalized difference water index time series, and a corrected red edge chlorophyll vegetation index time series.
[0050] Step 102: dividing the coastal zone latitude into several classification latitude bands as classification intervals, and calculating the monthly mean value index and the monthly variance index of each vegetation index in each classification interval, and the annual mean value index and the annual variance index of each vegetation index in the current year.
[0051] Step 103: based on the normalized difference vegetation index time series and the enhanced vegetation index time series, using a double logistic function to calculate the growth period and the decline period of the vegetation as candidate classification indicators.
[0052] Step 104: combining different vegetation classification samples, calculating the J-M distance of each candidate classification indicator on different vegetation classification samples.
[0053] Step 105: based on the J-M distance of each candidate indicator on different vegetation classification samples and the preliminary classification contribution evaluation, screening out the most separable classification indicators to construct a final random forest classification model.
[0054] Step 106: classifying the salt marsh vegetation in the coastal salt marsh wetland according to the final random forest classification model.
[0055] In another example embodiment of the present application, in order to accurately construct the random forest model to achieve the effect of classifying the salt marsh vegetation in different coastal salt marsh wetlands, the step 105 is replaced by steps 201-204 as follows:
[0056] The step 105 specifically comprises:
[0057] Step 201: extracting the indicators ranked in the top 50% of J-M distance as the preliminary classification variables of the random forest model.
[0058] Step 202: calculating the contribution of each indicator to the random forest classification model and ranking the contribution.
[0059] Step 203: taking the indicators ranked in the top 10 of the contribution as the most separable window period vegetation indices.
[0060] Step 204: taking the most separable window period vegetation indices as the independent variables of the final random forest classification model to construct the final random forest classification model.
[0061] In another example embodiment of the present application, in order to ensure the consistency of the input satellite remote sensing data, the input satellite remote sensing data can be preliminarily processed, and before the step 101, the method can further comprise steps 301-303. Among them:
[0062] Before the step 101, the satellite remote sensing data comprises Landsat-8 image data and Sentinel-2 image data.
[0063] Step 301: removing the pixels covered by clouds and shadows in the Landsat-8 image data using the QA band of Landsat-8 to generate the removed Landsat-8 image data.
[0064] Step 302: removing the pixels covered by clouds and shadows in the Sentinel-2 image data using the sentinel-2 cloud detector library and S2_CLOUD_PROBABILITY dataset to generate the removed Sentinel-2 image data.
[0065] Step 303: using the removed Sentinel-2 image data as a reference, using the least squares method to correct the differences between the removed Landsat-8 image data and the removed Sentinel-2 image data to determine the optimal satellite remote sensing data.
[0066] The step 301 specifically comprises: determining pixel occlusion information of the Landsat-8 image data covered by clouds and shadows in a decimal manner by using a QA band of Landsat-8; converting the pixel occlusion information into binary, and determining pixel quality based on a corresponding bit number of the binary pixel occlusion information; and removing pixels covered by clouds and shadows in the Landsat-8 image data based on the pixel quality by using a mask function, to generate removed Landsat-8 image data.
[0067] The step 302 specifically comprises: determining a pixel coverage probability of the Sentinel-2 image data covered by clouds and shadows based on a sentinel-2 cloud detector library and according to the S2_CLOUD_PROBABILITY data set; and removing pixels covered by clouds and shadows in the Sentinel-2 image data based on the pixel coverage probability by using a filtering function, to generate removed Sentinel-2 image data.
[0068] In another exemplary embodiment of the present application, for salt marsh plants, the following four indexes are selected as key parameters that can reflect the characteristics of the ground vegetation after literature research and local pre-classification results, and the step 101 comprises the following steps.
[0069] The step 101 specifically comprises: smoothing salt and pepper noise and extreme values appearing in satellite remote sensing data collected in a set period by using a Savitzky-Golay filtering algorithm, and using cubic convolution resampling to resample Sentinel-2 image data in the satellite remote sensing data, so that the resampled Sentinel-2 data is consistent with the spatial resolution of Landsat-8 image data in the satellite remote sensing data, to generate a vegetation index time series; and the vegetation index time series is displayed as a vegetation index annual characteristic distribution map.
[0070] The present application also provides an application scenario, which applies the salt marsh vegetation classification method described above. Specifically: Figure 3As shown, taking storage of all Landsat-8 Collection2 Level-2 OLI (TOA) data and Sentinel-2 Level-2A-MSI (TOA) data in the remote sensing cloud computing platform as an example, the Landsat-8 series data includes all available Landsat 8 Operational Land Imager (OLI) images from the United States Geological Survey, with a spatial resolution of 30 m and a ground revisit time of 8 days; the Sentinel-2 data includes all available Sentinel-2A and Sentinel-2B multispectral imager (MSI) images from the European Space Agency, with a spatial resolution of 10 m and a ground revisit time of 5 days.
[0071] Secondly, the satellite remote sensing data is preprocessed: since the remote sensing data set in the remote sensing cloud computing platform has been subjected to atmospheric correction / topographic correction, the pixels covered by clouds and shadows in the Landsat-8 image data and the Sentinel-2 image data are removed using the QA band and the S2_CLOUD_PROBABILITY data set of the Sentinel2 cloud detector library provided by Landsat-8, respectively.
[0072] Among them, the Landsat-QA band provides information on whether the pixel is blocked by clouds or fog in decimal form, which is converted to binary, and the pixel quality is determined according to the corresponding number of bits, and the corresponding removal is performed using the mask function provided by the remote sensing cloud computing platform, while the S2_CLOUD_PROBABILITY directly provides the possibility of whether the image pixel in Sentinel-2 is covered by clouds, so that the filtering function can be used to remove the pixels that are very likely to be contaminated by clouds with a threshold of 80%. At the same time, in order to eliminate the differences between the sensor bands of different satellites, taking the Sentinel-2 satellite with more images as a reference, the least square method is used to correct the differences between different sensor data, and finally the optimal remote sensing data image collection (i.e. the optimal satellite remote sensing data) is obtained.
[0073] Further, the vegetation index time series is generated according to the satellite remote sensing data: for salt marsh plants, after literature research and local pre-classification results, the following four indicators are selected as key parameters that can reflect the characteristics of the surface vegetation, including:
[0074] 1) Normalized Differential Vegetation Index (NDVI): calculated by the difference in absorption of near-infrared and infrared bands by vegetation, which can reflect the health of vegetation and the growth of vegetation.
[0075] 2) Enhanced Vegetation Index (EVI): NDVI index modified to reduce atmospheric and soil noise, which can well reflect the dense plant growth conditions.
[0076] 3) Normalized Difference Water Index (NDWI): Through the absorption difference of water in the green and near-infrared bands, it can reflect the surface soil moisture and vegetation coverage.
[0077] 4) Modified Red Edge Chlorophyll Index (MRECI): The chlorophyll content in the leaves nourished by nitrogen is reflected, which can show the photosynthetic activity of the canopy.
[0078] The index calculation is shown in formula (1), (2), (3) and (4) respectively. The results are resampled by three convolutions to unify the resolution of each image, so that four indicators in each pixel of a single image are obtained.
[0079]
[0080] Among them, NIR is the near-infrared band (845-885nm); RED is the red band (630-688nm); BLUE is the blue band (450-515nm); GREEN is the green band (530-590nm).
[0081] Due to the cloud and fog mask, pixel loss problems often occur in a single remote sensing image, accompanied by natural salt and pepper noise problems. In view of this, all remote sensing images in the range are aggregated to fill the lost pixels due to clouds and fog, and further smoothed using the Savitzky-Golay filtering algorithm which can smooth the data and maintain the original trend of change. Salt and pepper noise and extreme values appear, and the resolution of Sentinel-2 data is resampled by three convolutions to ensure that it is consistent with the spatial resolution of Landsat remote sensing images. Finally, a high-quality annual vegetation index annual characteristic distribution map with a time resolution of 15 days and a spatial resolution of 30m is obtained, which is the vegetation index time series.
[0082] In 2022, detailed field surveys and corresponding classification sample collection work were completed in multiple coastal salt marsh vegetation distribution areas in China. Taking the vegetation in the Yangtze River Estuary as an example, by combining field investigation classification sample data, the present application draws key plant index annual characteristic distribution maps of different regions and different vegetation, such asFigure 2 are shown.
[0083] Due to the large latitude span of China's coastal zone, a number of classification latitude zones are divided based on a large number of ground samples obtained from long-term coastal zone surveys, with a division of 3-5 degrees (flexible interval to ensure continuity of some areas of the coastal zone).
[0084] For each classification latitude zone, the monthly mean and monthly variance indicators of each calculated vegetation index are first synthesized monthly, and the annual mean and annual variance indicators of each vegetation index are calculated.
[0085] Based on the annual vegetation index time series, the Start of Growing Season (SOS) and End of Growing Season (EOS) phenological indicators of different vegetation are further fitted using a double logistic function for the NDVI sequence and the EVI sequence.
[0086] At this point, a total of 106 selected classification indicators are calculated on each pixel, including the monthly / annual mean and variance of the four vegetation indices calculated above, and the SOS and EOS phenological indicators generated based on the NDVI sequence and the EVI sequence.
[0087] Using different vegetation classification samples obtained from field surveys, the Jeffries-Matusita distance (J-M distance) of each vegetation index on different vegetation classification samples is calculated to represent the index separability. The J-M distance is a transformation of the Bhattacharyya distance, and is a widely used measure of spectral separability distance between two class density functions, and is usually used as a measure of class separability.
[0088] First, the indicators with J-M distance ranking in the top 50% are extracted as the preliminary classification variables of the smileRandomForest model, and then the contribution of each indicator to vegetation classification is determined. After sorting, the top 10 indicators with the highest contribution are selected as the most separable window period vegetation indices, and the final smileRandomForest model is constructed for final classification.
[0089] The vegetation classification samples are divided into 8:2, and the 10-fold cross-validation method is used to perform iterative training combined with the true vegetation classification samples obtained from field surveys. Based on the optimal parameters (the accuracy of each classification indicator is above 0.9), the smileRandomForest model is deployed in the salt marshes nationwide, using the above 10 most separable window period vegetation indices as independent variables, and the smileRandomForest model (machine learning model) to complete the classification of the national salt marsh vegetation.
[0090] For precision verification, the indices of User's Accuracy (UA), Producer's Accuracy (PA), Overall Accuracy (OA), F-score and Kappa coefficient are used for verification.
[0091]
[0092] wherein TP is the positive sample correctly classified by the model, TN is the negative sample correctly classified by the model, FP is the positive sample incorrectly classified by the model, FN is the negative sample incorrectly classified by the model, N is the total number of pixels, a i is the total number of actual observations of a certain sample type, and b i is the total number of model classification values of a certain sample type.
[0093] The technical solution of the present application is verified on the entire Chinese eastern coastal belt. Landsat 8 TOA and Sentinel-2 TOA data from 2017 to 2022 are obtained in Google Earth Engine, and necessary geometric correction, cloud mask, vegetation index calculation and other basic image preprocessing are completed to synthesize the vegetation index time series.
[0094] A large number of field vegetation classification samples obtained from long-term coastal belt field surveys from 2017 to 2022 and ultra-high resolution satellite remote sensing images are used to screen key classification window periods and classification indicators with high separability in multiple vegetation indices, to establish a random forest model and perform parameter tuning. The final constructed random forest model is used to classify the salt marsh vegetation of the Chinese coastal belt from 2017 to 2022, and the classification results and accuracy are as shown in Figure 4 and Table 1.
[0095] Table 1 Model classification result accuracy evaluation (last five years)
[0096]
[0097] Based on the same inventive concept, the embodiments of the present application also provide a salt marsh vegetation classification system, comprising:
[0098] A vegetation index time series generation module is configured to generate a vegetation index time series according to satellite remote sensing data. The vegetation index time series includes a normalized difference vegetation index time series, an enhanced vegetation index time series, a normalized difference water index time series, and a corrected red edge chlorophyll vegetation index time series.
[0099] The monthly index and the current year index calculation module is configured to divide the coastal latitude into a plurality of classified latitude bands as classified intervals, and to calculate a monthly average index and a monthly variance index of each vegetation index in each classified interval, and a yearly average index and a yearly variance index of each vegetation index in the current year.
[0100] The growth period candidate index and the decline period candidate index calculation module is configured to calculate the growth period and the decline period of the vegetation as candidate classification indexes by using a double logistic function based on the normalized difference vegetation index time series and the enhanced vegetation index time series.
[0101] The J-M distance calculation module is configured to calculate the J-M distance of each candidate classification index on different vegetation classification samples in combination with different vegetation classification samples.
[0102] Based on the J-M distance of each candidate index on different vegetation classification samples and the preliminary classification contribution evaluation, the most separable classification index is screened out to construct a final random forest classification model.
[0103] The classification module is configured to classify the salt marsh vegetation in the coastal salt marsh wetland according to the final random forest classification model.
[0104] In an exemplary embodiment, a computer device is provided, which can be a server or a terminal. The computer device includes a processor, a memory, an input / output interface (I / O) and a communication interface. The processor, the memory and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. The processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The input / output interface of the computer device is configured to exchange information between the processor and external devices. The communication interface of the computer device is configured to communicate with external terminals through network connection. The computer program is executed by the processor to implement a salt marsh vegetation classification method.
[0105] In an exemplary embodiment, a computer device is also provided, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the steps in the above method embodiments.
[0106] In an exemplary embodiment, a computer readable storage medium is provided, which stores a computer program. The computer program is executed by a processor to implement the steps in the above method embodiments.
[0107] In an example embodiment, a computer program product is provided, including a computer program which, when executed by a processor, implements the steps of any of the above method embodiments.
[0108] It should be noted that the user information (including but not limited to user equipment information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) involved in the present application are all information and data authorized by the user or authorized by all parties, and the collection, use and processing of related data need to comply with relevant regulations.
[0109] It can be understood by those skilled in the art that all or part of the processes in the above-mentioned embodiments can be completed by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer readable storage medium. When the computer program is executed, it can include the processes of the above-mentioned embodiments. Any reference to memory, database or other medium used in the embodiments provided by the present application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical storage, high-density embedded non-volatile memory, resistive memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration but not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc.
[0110] The database involved in the embodiments provided by the present application can include at least one of a relational database and a non-relational database. The non-relational database can include a distributed database based on blockchain, etc., without being limited thereto. The processor involved in the embodiments provided by the present application can be a general-purpose processor, a central processing unit, a graphics processing unit, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., without being limited thereto.
[0111] Any technical features in the above embodiments can be combined, and for the sake of brevity, not all possible combinations are described above, however, it should be understood that the application encompasses all possible combinations of the technical features unless such a combination is not technically possible.
[0112] The principles and implementation manners of the present application are described herein by using specific examples, and the above embodiments are only used to help understand the method of the present application and its core idea; meanwhile, according to the idea of the present application, the specific implementation manners and application scopes will be changed by those skilled in the art. In conclusion, the content of the present specification should not be understood as a limitation of the present application.
Claims
1. A method of classifying salt marsh vegetation, characterized by, The salt marsh vegetation classification method comprises: According to the satellite remote sensing data, a vegetation index time series is generated, specifically comprising: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises:
2. The salt marsh vegetation classification method of claim 1, wherein, The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification method comprises: The salt marsh vegetation classification The Sentinel-2 image data is removed from pixels covered by clouds and shadows by using a sentinel-2 cloud detector library and a S2_CLOUD_PROBABILITY dataset to generate removed Sentinel-2 image data. The removed Landsat-8 image data and the removed Sentinel-2 image data are corrected by using a least square method to correct differences between the removed Landsat-8 image data and the removed Sentinel-2 image data based on the removed Sentinel-2 image data to determine optimal satellite remote sensing data.
3. The salt marsh vegetation classification method of claim 2, wherein, The removed Landsat-8 image data is generated by removing pixels covered by clouds and shadows in the Landsat-8 image data by using a QA band of the Landsat-8, and specifically includes: The pixel occlusion information is converted into binary, and the pixel quality is determined based on the number of bits corresponding to the binary pixel occlusion information. The removed Landsat-8 image data is generated by removing pixels covered by clouds and shadows in the Landsat-8 image data by using a mask function based on the pixel quality. The removed Sentinel-2 image data is generated by removing pixels covered by clouds and shadows in the Sentinel-2 image data by using a sentinel-2 cloud detector library and a S2_CLOUD_PROBABILITY dataset, and specifically includes:
4. The salt marsh vegetation classification method of claim 2, wherein, The pixel coverage probability is determined based on the sentinel-2 cloud detector library and the S2_CLOUD_PROBABILITY dataset. The removed Sentinel-2 image data is generated by removing pixels covered by clouds and shadows in the Sentinel-2 image data by using a filter function based on the pixel coverage probability. The salt marsh vegetation classification system adopts the salt marsh vegetation classification method of any one of claims 1-4, and the salt marsh vegetation classification system includes:
5. A salt marsh vegetation classification system, characterized by, A vegetation index time series generation module for generating a vegetation index time series based on satellite remote sensing data; the vegetation index time series includes a normalized difference vegetation index time series, an enhanced vegetation index time series, a normalized difference water index time series, and a corrected red edge chlorophyll vegetation index time series; A monthly index and annual index calculation module for dividing a coastal belt latitude into a plurality of classification latitude bands as classification intervals, and calculating a monthly average index and a monthly variance index of each vegetation index in each classification interval, and an annual average index and an annual variance index of each vegetation index in each classification interval; A growth period candidate index and decline period candidate index calculation module for calculating a growth period and a decline period of vegetation as candidate classification indexes based on the normalized difference vegetation index time series and the enhanced vegetation index time series by using a double logistic function; A growth period candidate index and decline period candidate index calculation module for calculating a growth period and a decline period of vegetation as candidate classification indexes based on the normalized difference vegetation index time series and the enhanced vegetation index time series by using a double logistic function; A J-M distance calculation module is configured to calculate J-M distances of each candidate classification index on different vegetation classification samples in combination with different vegetation classification samples; each candidate index includes a monthly mean index, a monthly variance index, an annual mean index, an annual variance index, a growth period index, and a decline period index; Based on the J-M distances of each candidate index on different vegetation classification samples and the preliminary classification contribution evaluation, the most distinguishable classification index is screened out to construct a final random forest classification model; A classification module is configured to classify salt marsh vegetation in a coastal salt marsh wetland according to the final random forest classification model.
6. A computer device comprising: A memory and a processor to store a computer program stored on the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the salt marsh vegetation classification method of any one of claims 1-4.
7. A computer-readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to implement the steps of the salt marsh vegetation classification method of any one of claims 1-4.
8. A computer program product comprising a computer program, characterized in that, The computer program is executed by the processor to implement the steps of the salt marsh vegetation classification method of any one of claims 1-4.