Wetland classification method, device, equipment and storage medium based on remote sensing data

CN114187507BActive Publication Date: 2024-12-03GUANGZHOU INST OF GEOGRAPHY GUANGDONG ACAD OF SCI
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202111408502.4
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2021-11-19
Publication Date
2024-12-03
Estimated Expiration
2041-11-19

AI Technical Summary

Technical Problem

The analysis of wetland type in the coastal wetland distribution area requires a lot of manpower and material costs, and the analysis cycle is long, and there is a lack of fast and accurate methods for obtaining wetland types, which affects wetland resource management and sustainable development.

Method used

By de-cloud processing of the remote sensing image data set, it is converted into a cloudless multi-spectral image data set, and sample extraction and feature index analysis are carried out to realize wetland classification and identify and display with electronic map data.

Benefits of technology

While ensuring the analysis rate, sufficient and accurate analysis of the remote sensing image data set is achieved, the manpower and material costs are reduced, and the accuracy and efficiency of wetland type identification are improved.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN114187507B_ABST
    Figure CN114187507B_ABST
Patent Text Reader

Abstract

The present invention relates to the field of data analysis, and in particular to a wetland classification method based on remote sensing data. The method comprises: obtaining a remote sensing image data set of a target area, obtaining a cloud-likeness value of each pixel of the remote sensing image, extracting a target pixel of the remote sensing image according to the cloud-likeness value, splicing the obtained target pixels, obtaining a cloud-free multispectral image data set of the target area, obtaining area proportion data corresponding to each wetland type in the cloud-free multispectral image data set, performing sample extraction on the cloud-free multispectral image data set according to a preset sample extraction algorithm, and obtaining a cloud-free multispectral image sample data set; obtaining a characteristic index of the cloud-free multispectral image, and performing wetland classification on the cloud-free multispectral image sample data set according to the characteristic index and the cloud-free multispectral image sample data set to obtain wetland classification data of the target area.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of data analysis, and in particular to a wetland classification method, device, equipment and storage medium based on remote sensing data. Background Art

[0002] The coastal wetland distribution area has the characteristic of poor accessibility. The analysis of wetland types in the coastal wetland distribution area requires a lot of manpower and material resources, and the analysis cycle is long. Therefore, quickly and accurately obtaining the wetland types of the coastal wetland distribution area is of great significance for the management of wetland resources, the development and utilization of coastal areas, and the sustainable development of resources. Summary of the invention

[0003] Based on this, the purpose of the present invention is to provide a wetland classification method, device, equipment and storage medium based on remote sensing data, by declouding the remote sensing image data set, converting it into a cloud-free multispectral image data set, and performing sample extraction to obtain a cloud-free multispectral image sample data set; classification and recognition analysis are performed based on the characteristic index extracted from the cloud-free multispectral image data set and the cloud-free multispectral image sample data set, thereby ensuring the analysis rate while achieving a full and accurate analysis of the remote sensing image data set.

[0004] In a first aspect, an embodiment of the present application provides a wetland classification method based on remote sensing data, comprising the following steps:

[0005] Acquire a remote sensing image dataset of a target area, wherein the target remote sensing image dataset includes a plurality of remote sensing images;

[0006] Obtaining a cloud-likeness value of each pixel of the remote sensing image, extracting a target pixel of the remote sensing image according to the cloud-likeness value, and splicing the obtained target pixels to obtain a cloud-free multispectral image dataset of the target area, wherein the cloud-free multispectral image dataset includes a plurality of cloud-free multispectral images;

[0007] Acquire area ratio data corresponding to each wetland type in the cloud-free multispectral image dataset, and extract samples from the cloud-free multispectral image dataset according to a preset sample extraction algorithm to obtain a cloud-free multispectral image sample dataset;

[0008] Acquire a characteristic index of the cloud-free multispectral image, and classify the cloud-free multispectral image sample data set according to the characteristic index and the cloud-free multispectral image sample data set to obtain wetland classification data of the target area;

[0009] Obtain display instructions, wherein the display instructions include a remote sensing image data set to be classified, obtain electronic map data, obtain corresponding wetland classification data based on the remote sensing image data set to be classified, obtain wetland identification of each area of ​​the electronic map data based on the wetland type of the wetland classification data and the wetland identification corresponding to the wetland type, and display and mark the wetland identification on the electronic map data.

[0010] In a second aspect, an embodiment of the present application provides a wetland classification device based on remote sensing data, comprising:

[0011] An acquisition module, used to acquire a remote sensing image dataset of a target area, wherein the target remote sensing image dataset includes a plurality of remote sensing images;

[0012] a stitching module, configured to obtain a cloud-likeness value of each pixel of the remote sensing image, extract a target pixel of the remote sensing image according to the cloud-likeness value, stitch the obtained target pixels, and obtain a cloud-free multispectral image dataset of the target area, wherein the cloud-free multispectral image dataset includes a plurality of cloud-free multispectral images;

[0013] A sample extraction module is used to obtain the area data corresponding to each wetland type in the cloud-free multispectral image dataset, and extract samples from the cloud-free multispectral image dataset according to a preset sample extraction algorithm to obtain a cloud-free multispectral image sample dataset;

[0014] a classification module, configured to obtain a characteristic index of the cloud-free multispectral image, and classify the cloud-free multispectral image sample data set into wetlands according to the characteristic index and the cloud-free multispectral image sample data set, so as to obtain wetland classification data of the target area;

[0015] A display module is used to obtain display instructions, wherein the display instructions include a remote sensing image data set to be classified, obtain electronic map data, obtain corresponding wetland classification data based on the remote sensing image data set to be classified, obtain wetland identification of each area of ​​the electronic map data based on the wetland type of the wetland classification data and the wetland identification corresponding to the wetland type, and display and mark the wetland identification on the electronic map data.

[0016] In a third aspect, an embodiment of the present application provides a device comprising: a processor, a memory, and a computer program stored in the memory and executable on the processor; when the computer program is executed by the processor, the steps of the wetland classification method based on remote sensing data as described in the first aspect are implemented.

[0017] In a fourth aspect, an embodiment of the present application provides a storage medium storing a computer program, which, when executed by a processor, implements the steps of the wetland classification method based on remote sensing data as described in the first aspect.

[0018] In an embodiment of the present application, a wetland classification method, apparatus, device and storage medium based on remote sensing data are provided, wherein a remote sensing image dataset is declouded to be converted into a cloud-free multispectral image dataset, and sample extraction is performed to obtain a cloud-free multispectral image sample dataset; classification and recognition analysis is performed based on feature indices extracted from the cloud-free multispectral image dataset and the cloud-free multispectral image sample dataset, thereby achieving a sufficient and accurate analysis of the remote sensing image dataset while ensuring the analysis rate.

[0019] For better understanding and implementation, the present invention is described in detail below with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 A schematic diagram of a wetland classification method based on remote sensing data provided in one embodiment of the present application;

[0021] Figure 2 A schematic diagram of the process of S2 in a wetland classification method based on remote sensing data provided in one embodiment of the present application;

[0022] Figure 3 A schematic diagram of the flow of S201 in a wetland classification method based on remote sensing data provided in one embodiment of the present application;

[0023] Figure 4 A schematic diagram of the process of S4 in a wetland classification method based on remote sensing data provided in one embodiment of the present application;

[0024] Figure 5 A schematic flow chart of a wetland classification method based on remote sensing data provided in another embodiment of the present application;

[0025] Figure 6 A schematic diagram of the process of S6 in a wetland classification method based on remote sensing data provided in one embodiment of the present application;

[0026] Figure 7 A schematic diagram of the structure of a wetland classification device based on remote sensing data provided by an embodiment of the present application;

[0027] Figure 8 A schematic diagram of the structure of a device provided for one embodiment of the present application. DETAILED DESCRIPTION

[0028] Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The implementations described in the following exemplary embodiments do not represent all implementations consistent with the present application. Instead, they are merely examples of devices and methods consistent with some aspects of the present application as detailed in the appended claims.

[0029] The terms used in this application are for the purpose of describing specific embodiments only and are not intended to limit this application. The singular forms of "a", "said" and "the" used in this application and the appended claims are also intended to include plural forms unless the context clearly indicates other meanings. It should also be understood that the term "and / or" used herein refers to and includes any or all possible combinations of one or more associated listed items.

[0030] It should be understood that, although the terms first, second, third, etc. may be used in the present application to describe various information, these information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other. For example, without departing from the scope of the present application, the first information may also be referred to as the second information, and similarly, the second information may also be referred to as the first information. Depending on the context, the words "if" / "if" as used herein may be interpreted as "at the time of" or "when" or "in response to determination".

[0031] See also Figure 1 , Figure 1 A schematic flow chart of a wetland classification method based on remote sensing data provided in one embodiment of the present application, the method comprising the following steps:

[0032] S1: Acquire a remote sensing image dataset of a target area, wherein the target remote sensing image dataset includes a plurality of remote sensing images.

[0033] The executor of the wetland classification method based on remote sensing data is a classification device of the wetland classification method based on remote sensing data (hereinafter referred to as the classification device). In an optional embodiment, the classification device can be a computer device, a server, or a server cluster composed of multiple computer devices.

[0034] Due to the differences in sensors between the early Landsat 1, 2, 3, 4 and subsequent satellites, only the cloud-free multispectral and thermal infrared bands of Landsat 5TM, 7 ETM+ and 8 OLI / TIRS were selected as data sources. Among them, Landsat 5 selected all bands; Landsat 7 has two thermal infrared bands (high gain and low gain). Since there is no ultra-high temperature or ultra-low temperature surface in wetlands (monitored in low gain mode), and under normal temperature conditions, the inversion accuracy of high gain is higher than that of low gain, so the thermal infrared band in high gain mode is selected; Unlike Landsat 7, the two thermal infrared bands of Landsat 8 (bands 10 and 11) are located in different atmospheric windows. Since the radiation calibration error of band 11 is large, band 10 of Landsat 8 is selected as the thermal infrared band data source.

[0035] In this embodiment, the classification device obtains a remote sensing image dataset of the 10th band of Landsat 8 in the target area, wherein the target remote sensing image dataset includes a plurality of remote sensing images.

[0036] S2: Obtain cloud similarity values ​​for each pixel of the remote sensing image, extract target pixels of the remote sensing image according to the cloud similarity values, and stitch the obtained target pixels to obtain a cloud-free multispectral image dataset of the target area, wherein the cloud-free multispectral image dataset includes a plurality of cloud-free multispectral images.

[0037] The cloud-free multispectral image is an image formed by the reflection and transmission of electromagnetic waves of any band by an object, wherein the band is mostly in the visible light region;

[0038] In order to reduce the impact of interference caused by cloud shadows and thus accurately classify the wetland types in the target area, in this embodiment, the classification device obtains a cloud-free multispectral image dataset of the target area by declouding the remote sensing image.

[0039] Specifically, the classification device extracts target pixels of the remote sensing image in the remote sensing image data set, wherein the cloud similarity value of the target pixel is smaller than a preset cloud similarity threshold, and the cloud similarity value indicates the possibility that the pixel is contaminated by clouds. The smaller the cloud similarity value, the less affected the pixel is by cloud shadow interference.

[0040] The acquired target pixels are stitched to obtain a cloud-free multispectral image, and the cloud-free multispectral images corresponding to the plurality of remote sensing images are combined into a cloud-free multispectral image data set.

[0041] See also Figure 2 , Figure 2A schematic flow chart of S2 in a wetland classification method based on remote sensing data provided in an embodiment of the present application includes steps S201 to S202:

[0042] S201: Acquire the radiance, temperature and NDSI index of the plurality of remote sensing images, and calculate the cloud similarity value of each pixel of the remote sensing image according to the radiance, temperature, NDSI index and cloud similarity value calculation algorithm.

[0043] Radiance generally refers to radiation brightness. Radiance is referred to as radiance. Radiance represents the physical quantity of the radiation intensity of a point on a surface radiation source in a certain direction, that is, the total radiation energy in the target area displayed by the remote sensing image.

[0044] Temperature (LST) refers to the surface temperature. After the sun's heat energy is radiated to the ground, part of it is reflected and part of it is absorbed by the ground, causing the ground to heat up. The temperature obtained after measuring the temperature of the ground is the surface temperature, that is, the surface temperature in the target area shown in the remote sensing image.

[0045] NDSI is the Normalized Snow Cover Index, which uses a combination of visible and shortwave infrared bands to highlight the snow cover in the remote sensing image.

[0046] In this embodiment, the classification device obtains the radiance, temperature and NDSI index of the plurality of remote sensing images, and calculates the cloud similarity value of each pixel of the remote sensing image according to the radiance, temperature, NDSI index and cloud similarity value calculation algorithm, wherein the cloud similarity value calculation algorithm is:

[0047] S c =B p ×T p ×NDSI

[0048] In the formula, S c is the cloud similarity value of the remote sensing image, B p is the radiance, T p For temperature.

[0049] See also Figure 3 , Figure 3 A schematic flow chart of S201 in a wetland classification method based on remote sensing data provided in one embodiment of the present application includes steps S2011 to S2012:

[0050] S2011: Obtaining pixel values, gain coefficients, bias coefficients, temperature coefficients and band information parameters of the remote sensing image.

[0051] The pixel value is a numerical value obtained by quantifying the radiation received by the satellite, which is used to represent the pixel brightness value of the remote sensing image and records the grayscale value of the objects in the target area.

[0052] The gain coefficient and the bias coefficient are Landsat series radiation calibration parameters of the remote sensing image, wherein the gain coefficient is a coefficient for proportionally amplifying or reducing the data of the remote sensing image as a whole, and the bias coefficient is a coefficient for adjusting the zero point corresponding position of the data of the remote sensing image, which are used to obtain the radiance of the remote sensing image and realize cloud removal processing of the remote sensing image. In an optional embodiment, the gain coefficient and the bias coefficient can be obtained from the header file information of the remote sensing image.

[0053] The temperature coefficient is used to obtain the temperature of the remote sensing image based on the Landsat series of the remote sensing image and the temperature calculation algorithm. In an optional embodiment, the temperature coefficient can be obtained from the header file information of the remote sensing image.

[0054] The band information parameters are information parameters of electromagnetic waves in any band in the remote sensing image.

[0055] In an optional embodiment, the classification device can obtain the pixel value, gain coefficient, bias coefficient, temperature coefficient and band information parameters of the remote sensing image input by the user. In another optional embodiment, the pixel value, gain coefficient, bias coefficient, temperature coefficient and band information parameters of the remote sensing image may be data pre-stored by the classification device.

[0056] S2012: Acquire the radiance of the remote sensing image according to the pixel value, gain coefficient, bias coefficient and radiance calculation algorithm of the remote sensing image.

[0057] The radiance calculation algorithm is:

[0058] B p =DN×Gain+offset

[0059] In the formula, B p is the radiance, DN is the pixel value, Gain is the gain coefficient, and offset is the offset coefficient;

[0060] In this embodiment, the classification device obtains the radiance of the remote sensing image using a radiance calculation algorithm according to the pixel value, gain coefficient and bias coefficient of the acquired remote sensing image.

[0061] S2013: Acquire the temperature of the remote sensing image according to the temperature coefficient and the temperature calculation algorithm.

[0062] The temperature calculation algorithm is:

[0063]

[0064] Where, T p is temperature, k1 is the first value based on the temperature coefficient and the temperature calculation algorithm, and k2 is the second value based on the temperature coefficient and the temperature calculation algorithm.

[0065] In this embodiment, the classification device obtains the temperature of the remote sensing image by using a temperature calculation algorithm according to the first temperature coefficient and the temperature calculation algorithm and the second temperature coefficient and the temperature calculation algorithm of the acquired remote sensing image.

[0066] S2014: Acquire the NDSI index of the remote sensing image according to the band information parameters and the NDSI index calculation algorithm.

[0067] The NDSI index calculation algorithm is:

[0068]

[0069] Wherein, Green is the green band information parameter of the remote sensing image, SWIR1 is the near infrared 1 band information parameter of the remote sensing image, and SWRI2 is the near infrared 2 band information parameter of the remote sensing image.

[0070] In this embodiment, the classification device obtains the NDSI index of the remote sensing image by using an NDSI index calculation algorithm according to the band information parameters of the acquired remote sensing image.

[0071] S202: acquiring target pixels of the remote sensing image according to the cloud similarity value of each pixel of the remote sensing image and a preset cloud similarity threshold, stitching the target pixels to acquire a plurality of cloud-free multispectral images, and combining the images into a cloud-free multispectral image dataset.

[0072] In an optional embodiment, the cloud likelihood value obtained by the classification device is between [0, 100], and a value less than 10 indicates an acceptable lower cloud likelihood value. In order to minimize the influence of interference caused by cloud shadows and accurately classify the wetland type in the target area, the cloud likelihood threshold is set to 40.

[0073] In this embodiment, the classification device obtains pixels greater than the cloud-likeness threshold according to the cloud-likeness values ​​of each pixel of the remote sensing image and a preset cloud-likeness threshold, as target pixels of the remote sensing image, and splices the target pixels to obtain a cloud-free multispectral image.

[0074] S3: Obtain area ratio data corresponding to each wetland type in the cloud-free multispectral image dataset, and extract samples from the cloud-free multispectral image dataset according to a preset sample extraction algorithm to obtain a cloud-free multispectral image sample dataset.

[0075] The cloud-free multispectral image dataset includes several wetland types. Specifically, the wetland types include rivers, offshore and coastal wetlands, aquaculture, lakes and reservoirs, shallow sea waters, mangroves, and tidal flats.

[0076] The sample extraction algorithm is:

[0077]

[0078] Wherein, N is the sample size of a certain wetland type in the cloud-free multispectral image; Z is the statistic, when the confidence level is 95%, Z=1.96, when the confidence level is 90%, Z=1.64; P is the area ratio data of a certain wetland type; ε is the preset error value.

[0079] In this embodiment, the classification device obtains the sample quantity and area ratio data of a certain wetland type in the cloud-free multispectral image each year, and uses a sample extraction algorithm to extract samples from the cloud-free multispectral image data set according to preset statistics and error values, obtains a plurality of cloud-free multispectral image samples, and forms the cloud-free multispectral image sample data set.

[0080] S4: Acquire a characteristic index of the cloud-free multispectral image, and classify the cloud-free multispectral image sample data set into wetlands according to the characteristic index and the cloud-free multispectral image sample data set to obtain wetland classification data of the target area.

[0081] The characteristic index is a linear or nonlinear combination of different remote sensing spectral bands, wherein the characteristic index includes a water body index and a vegetation index. The water body index is an index used to reflect the difference between the reflection of the water body in the visible light and near-infrared bands and the soil background; the vegetation index is an index used to reflect the difference between the reflection of the vegetation in the visible light and near-infrared bands and the soil background, and can be used to quantitatively describe the growth status of the vegetation under certain conditions.

[0082] The classification device obtains characteristic indexes of the cloud-free multispectral image according to the band information parameters of the cloud-free multispectral image, wherein the characteristic indexes include mNDWI (modified normalized difference water index), NDBI (normalized building index), IR (infrared ratio index), SBL (soil background line index), etc.

[0083] In this embodiment, the classification device classifies the multispectral image sample data set into wetlands according to the acquired characteristic index and the cloud-free multispectral image sample data set through a random forest algorithm, obtains the wetland classification data of the target area, and saves it to the corresponding storage space.

[0084] See also Figure 4 , Figure 4 A schematic flow chart of S4 in a wetland classification method based on remote sensing data provided in an embodiment of the present application includes steps S401 to S402:

[0085] S401: Acquire a plurality of training samples from the cloud-free multispectral image sample dataset, and acquire a plurality of training features from the feature index.

[0086] In this embodiment, the classification device obtains a number of training samples from the cloud-free multispectral image sample data set in a sampling with replacement manner, and also obtains a number of training features from the feature index in a sampling with replacement manner.

[0087] S402: Obtain a plurality of classification trees according to the training samples, training features and a preset number of iterations, and obtain wetland classification data of the target area according to the classification trees.

[0088] In this embodiment, the classification device divides the training sample T into T1 and T2 according to the certain training feature C, and calculates the Gini index of the training sample T under the feature variable C according to the Gini index algorithm, wherein the Gini index algorithm is:

[0089]

[0090]

[0091] In the formula, Gini represents the Gini index, K represents the number of classification categories, and p i represents the probability of belonging to the i-th category.

[0092] For the training feature C, calculate the Gini(T|C) value after dividing the training sample into two parts under any value, and take the minimum Gini(T|C) value as the optimal binary solution under the training feature C.

[0093] Repeat the above steps to obtain the optimal binary solution for all training features, take the solution corresponding to the minimum Gini index as the optimal binary solution for the training sample and the overall training features, and use it as the two child nodes of the classification tree to obtain several classification trees.

[0094] The classification tree is a weak classifier with low classification accuracy. Each classification tree outputs a classification result, which is the wetland classification result of each pixel of the training sample. According to the classification tree, K classification results are obtained. Among the classification results, the one with the largest proportion is taken as the wetland classification result of the pixel, and the wetland classification data of the target area is obtained.

[0095] See also Figure 5 , Figure 5 A schematic flow chart of a wetland classification method based on remote sensing data provided in another embodiment of the present application further includes step S6, which is before step S5 and is specifically as follows:

[0096] S6: Correcting the wetland classification data of the target area to obtain corrected wetland classification data.

[0097] See also Figure 6 , Figure 6 The schematic flow chart of S6 in the wetland classification method based on remote sensing data provided in one embodiment of the present application includes steps S601 to S602, which are specifically as follows:

[0098] S601: Acquire remote sensing images of the target area for several consecutive years, correct wetland classification data corresponding to the years according to pixels of the remote sensing images, and obtain corrected wetland classification data.

[0099] The classification device obtains remote sensing images of the target area for several consecutive years, and obtains corrected wetland classification data based on pixels of the remote sensing images, wherein the type of the pixel is a wetland type or a water body type, and the wetland types include mangrove wetland types, tidal flat wetland types and shallow sea water wetland types.

[0100] Taking the correction of the mangrove wetland type in the i-th year as an example, the classification device extracts the remote sensing images of the target area two years before and after the i-th year from the remote sensing images, and constructs several time series, where the expression of the time series is:

[0101] [y i-2 ,y i-1 ,y i ,y i+1 ,y i+2 ]

[0102] In the formula, y i is the remote sensing image of the i-th year, y i+1 is the remote sensing image of the i+1th year, y i-1 is the remote sensing image of year i-1.

[0103] The time series is analyzed. When the types of each pixel of the remote sensing image in the time series are not the mangrove wetland type, a classification error result is obtained. According to the classification error result, the wetland type with the largest proportion of pixels of a certain wetland type in the remote sensing image of the time series is obtained, and the wetland type with the largest proportion is used as the corrected wetland type. The wetland type of the remote sensing image of the time series is replaced, thereby correcting the wetland classification data corresponding to the year and obtaining the corrected wetland classification data.

[0104] S602: Obtain the type of pixels of the remote sensing image of the target area, and correct the wetland classification data after the primary correction according to the type of pixels of the remote sensing image to obtain the wetland classification data after the secondary correction, wherein the pixel type includes wetland type and water body type.

[0105] In this embodiment, the classification device obtains the type of pixels of the remote sensing image of the target area, corrects the wetland classification data after the primary correction according to the type of pixels of the remote sensing image, and obtains the wetland classification data after the secondary correction.

[0106] In an optional embodiment, the classification device selects a unit pixel from the pixels of the remote sensing image. The unit pixel may be a single pixel or an image patch. The image patch is a wetland area where pixels of the same wetland type are spatially clustered together and interconnected.

[0107] Obtain neighboring pixels centered on the unit pixel, wherein the neighboring pixels are pixels whose distance from the unit pixel is less than a preset pixel distance threshold, form the neighboring pixels into a pixel sequence, and judge whether the type of each pixel in the pixel sequence is a water body type. When the type of each pixel in the pixel sequence is not a water body type, obtain a classification error result, obtain the wetland type of the neighboring pixel with the largest proportion of a certain wetland type in the pixel sequence, and use it as the second corrected wetland type to replace the wetland type of the neighboring pixels of the pixel sequence, thereby correcting the wetland classification data after the first correction, and obtaining the wetland classification data after the second correction.

[0108] S5: Obtain display instructions, wherein the display instructions include a remote sensing image data set to be classified, obtain electronic map data, obtain corresponding wetland classification data based on the remote sensing image data set to be classified, obtain wetland identification of each area of ​​the electronic map data based on the wetland type of the wetland classification data and the wetland identification corresponding to the wetland type, and display and mark the wetland identification on the electronic map data.

[0109] The wetland display instruction is issued by a user and received by a classification device.

[0110] In this embodiment, the classification device obtains the wetland display instruction sent by the user, responds, and obtains the electronic map data.

[0111] The classification device obtains the remote sensing image data set to be classified according to the wetland display instruction, obtains the corresponding wetland classification data, obtains the wetland identification of each area of ​​the electronic map data according to the wetland type of the wetland classification data and the wetland identification corresponding to the wetland type, and returns it to the display interface of the detection device, and displays and marks the wetland identification on the electronic map data.

[0112] Please refer to Figure 7 , Figure 7 A schematic diagram of the structure of a wetland classification device based on remote sensing data provided by an embodiment of the present application. The device can implement all or part of the wetland classification device based on remote sensing data through software, hardware, or a combination of both. The device 7 includes:

[0113] An acquisition module 71 is used to acquire a remote sensing image dataset of a target area, wherein the target remote sensing image dataset includes a plurality of remote sensing images;

[0114] A stitching module 72 is used to obtain a cloud-likeness value of each pixel of the remote sensing image, extract a target pixel of the remote sensing image according to the cloud-likeness value, and stitch the obtained target pixels to obtain a cloud-free multispectral image dataset of the target area, wherein the cloud-free multispectral image dataset includes a plurality of cloud-free multispectral images;

[0115] The sample extraction module 73 is used to obtain the area data corresponding to each wetland type in the cloud-free multispectral image dataset, and extract samples from the cloud-free multispectral image dataset according to a preset sample extraction algorithm to obtain a cloud-free multispectral image sample dataset;

[0116] A classification module 74 is used to obtain a characteristic index of the cloud-free multispectral image, and classify the cloud-free multispectral image sample data set into wetlands according to the characteristic index and the cloud-free multispectral image sample data set to obtain wetland classification data of the target area;

[0117] The display module 75 is used to obtain display instructions, wherein the display instructions include a remote sensing image data set to be classified, obtain electronic map data, obtain corresponding wetland classification data based on the remote sensing image data set to be classified, obtain wetland identification of each area of ​​the electronic map data based on the wetland type of the wetland classification data and the wetland identification corresponding to the wetland type, and display and mark the wetland identification on the electronic map data.

[0118] In an embodiment of the present application, a remote sensing image dataset of a target area is acquired through an image acquisition module, wherein the target remote sensing image dataset includes a plurality of remote sensing images; a cloud-likeness value of each pixel of the remote sensing image is acquired through a stitching module, a target pixel of the remote sensing image is extracted according to the cloud-likeness value, and the acquired target pixel is stitched to acquire a cloud-free multispectral image dataset of the target area, wherein the cloud-free multispectral image dataset includes a plurality of cloud-free multispectral images; area data corresponding to each wetland type in the cloud-free multispectral image dataset is acquired, and sample extraction is performed on the cloud-free multispectral image dataset according to a preset sample extraction algorithm to acquire a cloud-free multispectral image dataset. Image sample data set; through the classification module, the characteristic index of the cloudless multispectral image is obtained, and the wetland classification of the cloudless multispectral image sample data set is performed on the cloudless multispectral image sample data set according to the characteristic index and the cloudless multispectral image sample data set, so as to obtain the wetland classification data of the target area; through obtaining display instructions, wherein the display instructions include the remote sensing image data set to be classified, the electronic map data is obtained, and the corresponding wetland classification data is obtained according to the remote sensing image data set to be classified, and the wetland identification of each area of ​​the electronic map data is obtained according to the wetland type of the wetland classification data and the wetland identification corresponding to the wetland type, and the wetland identification is displayed and marked on the electronic map data. The remote sensing image data set is declouded, converted into a cloudless multispectral image data set, and sample extraction is performed to obtain a cloudless multispectral image sample data set; classification and recognition analysis is performed according to the characteristic index extracted from the cloudless multispectral image data set and the cloudless multispectral image sample data set, so as to achieve a full and accurate analysis of the remote sensing image data set while ensuring the analysis rate.

[0119] Please refer to Figure 8 , Figure 8 The schematic diagram of the structure of the device provided in one embodiment of the present application, the device 8 comprises: a processor 81, a memory 82 and a computer program 83 stored in the memory 82 and executable on the processor 81; the computer device may store a plurality of instructions, the instructions being suitable for being loaded and executed by the processor 81 Figures 1 to 6 The method steps of the embodiment shown in the figure can be found in the specific implementation process. Figures 1 to 6 The specific description of the illustrated embodiment will not be repeated here.

[0120] Among them, the processor 81 may include one or more processing cores. The processor 81 uses various interfaces and lines to connect various parts in the server, and executes various functions and processes data of the wetland classification device 7 based on remote sensing data by running or executing instructions, programs, code sets or instruction sets stored in the memory 82, and calling the data in the memory 82. Optionally, the processor 81 can be implemented in at least one hardware form of digital signal processing (Digital Signal Processing, DSP), field programmable gate array (Field-Programmable Gate Array, FPGA), and programmable logic array (Programble Logic Array, PLA). The processor 81 can integrate one or more combinations of a central processing unit 81 (Central Processing Unit, CPU), an image processor 81 (Graphics Processing Unit, GPU) and a modem. Among them, the CPU mainly processes the operating system, user interface and application programs; the GPU is responsible for rendering and drawing the content to be displayed on the touch display; the modem is used to process wireless communications. It can be understood that the above-mentioned modem may not be integrated into the processor 81, and it can be implemented by a single chip.

[0121] Among them, the memory 82 may include a random access memory 82 (Random Access Memory, RAM), and may also include a read-only memory 82 (Read-Only Memory). Optionally, the memory 82 includes a non-transitory computer-readable storage medium. The memory 82 can be used to store instructions, programs, codes, code sets or instruction sets. The memory 82 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for at least one function (such as touch instructions, etc.), instructions for implementing the above-mentioned various method embodiments, etc.; the data storage area may store data involved in the above-mentioned various method embodiments, etc. The memory 82 may also be optionally at least one storage device located away from the aforementioned processor 81.

[0122] The present application also provides a storage medium that can store multiple instructions, which are suitable for the processor to load and execute the above-mentioned Figures 1 to 6 The method steps of the embodiment shown in the figure can be found in the specific implementation process. Figures 1 to 6 The specific description of the illustrated embodiment will not be repeated here.

[0123] The technicians in the relevant field can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example for illustration. In practical applications, the above-mentioned function allocation can be completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiment can be integrated in a processing unit, or each unit can exist physically separately, or two or more units can be integrated in one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of this application. The specific working process of the units and modules in the above-mentioned system can refer to the corresponding process in the aforementioned method embodiment, which will not be repeated here.

[0124] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0125] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraint algorithm of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0126] In the embodiments provided by the present invention, it should be understood that the disclosed devices / terminal equipment and methods can be implemented in other ways. For example, the device / terminal equipment embodiments described above are only schematic. For example, the division of the modules or units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0127] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0128] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0129] If the integrated module / unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present invention implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and when the computer program is executed by a processor, the steps of the above-mentioned method embodiments can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form, etc.

[0130] The present invention is not limited to the above-mentioned embodiments. If various changes or modifications to the present invention do not depart from the spirit and scope of the present invention, and if these changes and modifications fall within the scope of the claims and equivalent technologies of the present invention, the present invention is also intended to include these changes and modifications.

Claims

1. A wetland classification method based on remote sensing data, characterized in that: The following steps are involved: Acquire a remote sensing image dataset of a target area, wherein the remote sensing image dataset of the target area includes a plurality of remote sensing images; The radiance, temperature and NDSI index of the remote sensing images are obtained, and the cloud similarity value of each pixel of the remote sensing image is calculated according to the radiance, temperature, NDSI index and cloud similarity value calculation algorithm, wherein the cloud similarity value calculation algorithm is: S c =B p ×T p ×NDSI In the formula, S c is the cloud similarity value of the remote sensing image, B p is the radiance, T p is temperature; Extracting target pixels of the remote sensing image according to the cloud-likeness value, and splicing the extracted target pixels to obtain a cloud-free multispectral image dataset of the target area, wherein the cloud-free multispectral image dataset includes a plurality of cloud-free multispectral images; Acquire area ratio data corresponding to each wetland type in the cloud-free multispectral image dataset, and extract samples from the cloud-free multispectral image dataset according to a preset sample extraction algorithm to obtain a cloud-free multispectral image sample dataset; Acquire a characteristic index of the cloud-free multispectral image, and classify the cloud-free multispectral image sample data set according to the characteristic index and the cloud-free multispectral image sample data set to obtain wetland classification data of the target area; Obtain display instructions, wherein the display instructions include a remote sensing image data set to be classified, obtain electronic map data, obtain corresponding wetland classification data based on the remote sensing image data set to be classified, obtain wetland identification of each area of ​​the electronic map data based on the wetland type of the wetland classification data and the wetland identification corresponding to the wetland type, and display and mark the wetland identification on the electronic map data.

2. The wetland classification method based on remote sensing data according to claim 1 is characterized in that: The step of extracting target pixels of the remote sensing image according to the cloud-likeness value, splicing the obtained target pixels, and acquiring a cloud-free multispectral image dataset of the target area comprises the following steps: According to the cloud-likeness value of each pixel of the remote sensing image and a preset cloud-likeness threshold, a target pixel of the remote sensing image is obtained, the target pixel is spliced, a plurality of cloud-free multispectral images are obtained, and the cloud-free multispectral images are combined into a cloud-free multispectral image data set.

3. The wetland classification method based on remote sensing data according to claim 2 is characterized in that: The step of obtaining the radiance, temperature and NDSI index of the plurality of remote sensing images comprises the following steps: Obtaining pixel values, gain coefficients, bias coefficients, temperature coefficients and band information parameters of the remote sensing image; The radiance of the remote sensing image is obtained according to the pixel value, gain coefficient, bias coefficient and radiance calculation algorithm of the remote sensing image, wherein the radiance calculation algorithm is: B p =DN×Gain+offset In the formula, B p is the radiance, DN is the pixel value, Gain is the gain coefficient, and offset is the offset coefficient; The temperature of the remote sensing image is obtained according to the temperature coefficient and the temperature calculation algorithm, wherein the temperature calculation algorithm is: Where, T p is temperature, k1 is the first temperature coefficient, k2 is the second temperature coefficient; According to the band information parameters and the NDSI index calculation algorithm, the NDSI index of the remote sensing image is obtained, wherein the NDSI index calculation algorithm is: Wherein, Green is the green band information parameter of the remote sensing image, SWIR1 is the near infrared 1 band information parameter of the remote sensing image, and SWRI2 is the near infrared 2 band information parameter of the remote sensing image.

4. The wetland classification method based on remote sensing data according to claim 1 is characterized in that: The step of performing wetland classification on the cloud-free multispectral image sample data set according to the characteristic index and the cloud-free multispectral image sample data set to obtain wetland classification data of the target area comprises the following steps: Acquire a plurality of training samples from the cloud-free multispectral image sample data set, and acquire a plurality of training features from the feature index; According to the training samples, training features and a preset number of iterations, several classification trees are obtained, and according to the classification trees, wetland classification data of the target area is obtained.

5. The wetland classification method based on remote sensing data according to claim 1, characterized in that: Also includes the steps: The wetland classification data of the target area is corrected to obtain corrected wetland classification data.

6. The wetland classification method based on remote sensing data according to claim 5 is characterized in that: The step of correcting the wetland classification data of the target area to obtain the corrected wetland classification data comprises the following steps: Acquire remote sensing images of the target area for several consecutive years, and correct the wetland classification data corresponding to the years according to the pixels of the remote sensing images to obtain corrected wetland classification data; The type of pixels of the remote sensing image of the target area is obtained, and according to the type of pixels of the remote sensing image, the wetland classification data after the primary correction is corrected to obtain the wetland classification data after the secondary correction, wherein the type of pixels includes wetland type and water body type.

7. A wetland classification device based on remote sensing data, characterized in that: include: An acquisition module is used to acquire a remote sensing image dataset of a target area, wherein the remote sensing image dataset of the target area includes a plurality of remote sensing images; The splicing module is used to obtain the radiance, temperature and NDSI index of the plurality of remote sensing images, and calculate the cloud similarity value of each pixel of the remote sensing image according to the radiance, temperature, NDSI index and cloud similarity value calculation algorithm, wherein the cloud similarity value calculation algorithm is: S c =B p ×T p ×NDSI In the formula, S c is the cloud similarity value of the remote sensing image, B p is the radiance, T p is temperature; Extracting target pixels of the remote sensing image according to the cloud-likeness value, and splicing the extracted target pixels to obtain a cloud-free multispectral image dataset of the target area, wherein the cloud-free multispectral image dataset includes a plurality of cloud-free multispectral images; A sample extraction module is used to obtain the area data corresponding to each wetland type in the cloud-free multispectral image dataset, and to extract samples from the cloud-free multispectral image dataset according to a preset sample extraction algorithm to obtain a cloud-free multispectral image sample dataset; a classification module, configured to obtain a characteristic index of the cloud-free multispectral image, and classify the cloud-free multispectral image sample data set into wetlands according to the characteristic index and the cloud-free multispectral image sample data set, so as to obtain wetland classification data of the target area; A display module is used to obtain display instructions, wherein the display instructions include a remote sensing image data set to be classified, obtain electronic map data, obtain corresponding wetland classification data based on the remote sensing image data set to be classified, obtain wetland identification of each area of ​​the electronic map data based on the wetland type of the wetland classification data and the wetland identification corresponding to the wetland type, and display and mark the wetland identification on the electronic map data.

8. A device, characterized in that include: A processor, a memory, and a computer program stored in the memory and executable on the processor; when the computer program is executed by the processor, the steps of the wetland classification method based on remote sensing data as described in any one of claims 1 to 6 are implemented.

9. A storage medium, characterized in that: The storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the wetland classification method based on remote sensing data as described in any one of claims 1 to 6 are implemented.