Water body recognition method based on satellite remote sensing image block classification of supervised learning algorithm

Abstract

The application provides a satellite remote sensing image block classification water body identification method based on a supervised learning algorithm, which takes a Sentinel-1 SAR image as basic data, and realizes water body identification based on a satellite remote sensing image through the steps of image preprocessing, feature data preparation, sample point preparation, classification model construction and identification, and image post-processing. The application takes the VV, VH, angle, VV / VH and sum bands of the SAR image as the training feature input of the classification model, which is conducive to improving the accuracy of the classification result; the hexagon is taken as the minimum block unit to block and calculate the large-scale image, which can eliminate the edge effect in the block calculation, more accurately capture the features of the ground objects on the local scale, and reduce the calculation amount of a single local classification model; through the image post-processing step, the non-water body noise pixels in the classified image can be eliminated, and the influence of noise such as vegetation and buildings can be weakened. The method has the characteristics of high classification and identification precision, and is especially suitable for the identification of large-scale water bodies.

Description

Technical Field

[0001] The present invention belongs to the field of remote sensing technology and surface water body identification, and relates to a water body identification method based on satellite remote sensing image block classification based on a supervised learning algorithm. Background Art

[0002] In recent years, climate change and human activities have led to frequent extreme weather events, often causing widespread floods, droughts, and other disasters, significantly impacting socioeconomic development and ecological and environmental protection. Surface water characteristics are a crucial component of characterizing flood and drought disasters. Understanding the dynamics of surface water is crucial for flood and drought monitoring, risk analysis, and social development planning.

[0003] Using remote sensing satellite imagery with high temporal and spatial resolution, rapid and automated water body identification and extraction can be achieved. Currently, commonly used remote sensing imagery mainly includes optical satellite imagery and radar satellite imagery. Among them, SAR (Synthetic Aperture Radar) sensors are not affected by factors such as cloud cover and rainy weather, and can capture ground images around the clock. For example, the Sentinel-1 satellite can provide SAR imagery with a spatial resolution of 10 meters and a revisit period of 12 days. It can provide a data foundation for large-scale surface water body identification and monitoring, and has broad application prospects in analyzing the long-term evolution of surface water bodies, flood inundation monitoring, and drought characteristics.

[0004] Furthermore, machine learning algorithms are highly portable and computationally efficient, demonstrating great potential for identifying features in large-scale remote sensing imagery. For example, using Sentinel-1 SAR imagery to analyze flood disasters, Wang et al. employed a support vector machine classification algorithm to investigate flood inundation in the middle and lower reaches of the Yangtze River during the 2020 flood season. Dong et al. constructed a transformer-based change detection model and mapped the water evolution and flood inundation of Poyang Lake in 2020. Roth et al. processed imagery using the TU Wien algorithm to create a dynamic map of water body changes in Pakistan from August to September 2022, further capturing and analyzing the maximum inundation range of the flood. Although using machine learning algorithms to process Sentinel-1 SAR imagery is a common method for identifying large-scale water bodies, water body identification methods that are both universal and highly accurate are still lacking, especially in large-scale areas, where effective water body identification is difficult to achieve. Summary of the Invention

[0005] Since there is currently a lack of water body identification methods that are both universal and highly accurate, in order to achieve accurate identification of water bodies over a large scale, the present invention provides a water body identification method based on satellite remote sensing image block classification based on a supervised learning algorithm to improve the accuracy of water body identification, especially to improve the accuracy of water body identification over a large scale.

[0006] In order to achieve the above-mentioned object of the invention, the technical solution adopted by the present invention is as follows:

[0007] The water body identification method based on satellite remote sensing image block classification based on supervised learning algorithm includes the following steps:

[0008] (1) Image preprocessing

[0009] A Sentinel-1SAR Level-1 GRD dataset of the area to be classified within the target time period is obtained, which is denoted as the target SAR dataset. The target SAR dataset includes several SAR images. Each SAR image in the target SAR dataset is preprocessed to obtain several preprocessed SAR images. The preprocessing includes band screening, edge noise correction, terrain correction, noise filtering, and hand index masking.

[0010] Band filtering refers to removing the HH and HV band information from the SAR image while retaining the VV, VH, and angle band information in the SAR image; edge noise correction refers to performing edge noise correction on the SAR image after band filtering; terrain correction refers to using the DEM (Digital Elevation Model) data of the area to be classified to mask the overlap and shadow areas in the SAR image after edge noise correction; noise filtering refers to using the Refined Lee filter processing algorithm to process the terrain-corrected SAR image to reduce the clutter in the SAR image; and hand index masking refers to calculating the hand index based on the noise-filtered SAR image and masking out areas with a hand index greater than 20.

[0011] (2) Feature data preparation

[0012] Based on each preprocessed SAR image, VV / VH and sum are calculated. VV / VH is the ratio of VV to VH, and sum is the sum of VV and VH. VV, VH, angle, VV / VH, and sum of each preprocessed SAR image are used as feature data.

[0013] (3) Sample point preparation

[0014] The RGB channels were set using the VV, VH, and VV / VH of the preprocessed SAR images. Each preprocessed SAR image was visualized, and the visualized SAR images were mosaicked to obtain a base map. Several water body sample points and non-water body sample points were randomly selected from the base map, and all sample points were divided into a training data set and a validation data set.

[0015] (4) Construction and identification of classification models

[0016] The base map of step (3) is divided into blocks by using block calculation to generate a number of block units that completely cover the area to be classified, and a local classification model is constructed in each block unit. The local classification model is constructed using the random forest algorithm as a classifier, and each local classification model is trained in its own block unit and its adjacent block units; during training, the feature data is input as the feature data of the local classification model, and the data in the training data set is input as the training data; each local classification model after training is used to identify the water body and non-water body in the block unit where it is located, and obtain a preliminary recognition result image;

[0017] (5) Image post-processing

[0018] Acquire all Sentinel-2MSI optical images of the area to be classified within the target time period in which the percentage of cloud pixels is less than 10%, use the cloud removal function to remove the cloud pixels in each Sentinel-2MSI optical image, calculate the MNDWI index and EVI index of each Sentinel-2MSI optical image with cloud pixels removed, construct the MNDWI maximum value image and the EVI minimum value image, the pixels with the maximum MNDWI value < 0 in the MNDWI maximum value image or the pixels with the minimum EVI value > 0 in the EVI minimum value image are non-water body pixels, and the non-water body pixels in the preliminary recognition result image obtained in step (4) are removed in combination with the spatial coordinates of the non-water body pixels to obtain an image after non-water pixel filtering processing; then remove the water body pixels in the image after non-water pixel filtering processing in which the number of adjacent pixels is less than a preset value to obtain the water body distribution image of the area to be classified, thus completing water body recognition.

[0019] In step (4) of the above technical solution, the number of decision trees in each local classification model is preferably 60 to 100.

[0020] In step (4) of the above technical solution, the block unit is preferably a regular hexagonal block unit. Further, the side length of the regular hexagonal block unit in step (4) is preferably 0.4° to 0.5°, where the unit "°" refers to the geographic coordinate system unit.

[0021] In step (3) of the above technical solution, all sample points can usually be divided into a training data set and a validation data set according to the division ratio of (7-8):(3-2).

[0022] In step (5) of the above technical solution, when removing water pixels whose number of adjacent pixels in the image after non-water pixel filtering is less than a preset value, the preset value of the number of adjacent pixels is determined according to actual application requirements. When it is necessary to obtain water body distribution information of a larger scale in the area to be classified, a relatively large preset value can be selected. For example, when it is necessary to identify water bodies with an area greater than 0.3 hectares in the classified area, the preset value can be set to 30, that is, it is necessary to remove images of water bodies with an area not exceeding 0.3 hectares in the image after non-water pixel filtering.

[0023] The method of the present invention is particularly suitable for identifying water bodies in a large scale range and can produce higher recognition accuracy. The large scale here is usually an area of ​​more than 3,000 square kilometers.

[0024] Compared with the prior art, the technical solution of the present invention produces the following beneficial technical effects:

[0025] The present invention provides a method for identifying water bodies by using satellite remote sensing imagery block classification based on a supervised learning algorithm. The method uses Sentinel-1 SAR images as basic data and, after image preprocessing, feature data preparation, sample point preparation, classification model construction and identification, and image postprocessing, achieves water body identification based on satellite remote sensing images. The present invention optimizes the VV, VH, angle, VV / VH, and sum bands of SAR images as training feature inputs for the classification model, thereby improving the accuracy of classification results. The high accuracy of the classification results also demonstrates the effectiveness of the training features, providing a reference for SAR image water body identification technology. The present invention uses hexagons as the minimum block unit to perform block calculations on large-scale images. The hexagonal block unit facilitates eliminating edge effects that are prone to occur during block calculations, thereby improving the consistency of classification and identification results. Furthermore, block calculations can more accurately capture features of objects at a local scale, thereby overcoming the problem of large-scale spatial heterogeneity. Furthermore, block calculations can significantly reduce the computational complexity of a single local classification model and increase the speed of classification calculations. During the image post-processing step, the present invention utilizes images with the maximum MNDWI or minimum EVI values ​​to effectively eliminate noise pixels from non-water bodies in the classified image, effectively reducing the impact of noise from vegetation, buildings, and other factors. Compared to existing technologies, the method described in this invention boasts high classification and recognition accuracy. It is suitable for identifying large-scale water bodies in areas exceeding 3,000 square kilometers, providing a reliable technical solution for monitoring disasters such as floods and droughts at river basin, national, or intercontinental scales. This is of great significance for disaster prevention, mitigation, and policy formulation. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 It is a schematic diagram of the identification process of the method of the present invention.

[0027] Figure 2 This is a flowchart of image preprocessing.

[0028] Figure 3 It is a flowchart of local classification model training and recognition using the trained local classification model.

[0029] Figure 4 It is the spatial location and topographic distribution map of the area to be classified in Example 1.

[0030] Figure 5 This is the distribution of sample points in Example 1, where blue sample points are water bodies and red sample points are non-water bodies.

[0031] Figure 6 This is the block division of the area to be classified in Example 1.

[0032] Figure 7 This is the classification and recognition result of Example 1, where blue represents water bodies and gray represents non-water bodies. DETAILED DESCRIPTION

[0033] The following examples further illustrate the method for identifying water bodies by using satellite remote sensing imagery block classification based on a supervised learning algorithm provided by the present invention. It is necessary to point out that the following examples are only used to further illustrate the present invention and are not to be construed as limiting the scope of protection of the present invention. Based on the above-mentioned invention, those skilled in the art may make some non-essential improvements and adjustments to the present invention for specific implementation, which still fall within the scope of protection of the invention.

[0034] Example 1

[0035] Take a certain area in Beijing, Tianjin and Hebei as the area to be classified, its spatial location and terrain distribution are as follows Figure 4 As shown in the figure, there are a large number of surface water bodies such as reservoirs and rivers in the area. Analyzing the distribution of water bodies in the area is of great significance for monitoring flood and drought disasters and water resource changes in the area. Specifically, this embodiment uses the method of the present invention to identify the surface water bodies in the area between June 13 and June 18, 2023. The identification process diagram is shown in the figure below. Figure 1 shown.

[0036] 1. Image preprocessing

[0037] The Sentinel-1SAR Level-1 GRD dataset (available at: https: / / developers.google.com / earth-engine / datasets / catalog / COPERNICUS_S1_GRD) is used as the image source. Based on the raw Sentinel-1SAR imagery, this dataset undergoes orbit calibration, GRD boundary noise removal, thermal noise removal, radiometric calibration, and terrain correction. The image band parameters of the Sentinel-1SAR Level-1 GRD dataset are shown in Table 1. All data in the Sentinel-1SAR Level-1 GRD dataset include spatial coordinate information, and the SAR images obtained in the following processing also have spatial coordinate information.

[0038] Table 1 Band parameter information of Sentinel-1SAR Level-1 GRD dataset images

[0039]

[0040]

[0041] The Sentinel-1SARLevel-1GRD dataset of the area to be classified within the target period (June 13 to June 18, 2023) is obtained and recorded as the target SAR dataset. The target SAR dataset includes several SAR images. Each SAR image in the target SAR dataset is preprocessed on the Google Earth Engine cloud platform to obtain several preprocessed SAR images. The image preprocessing process is shown in the figure. Figure 2 The specific operations are as follows:

[0042] (1) Filter bands

[0043] The HH and HV band information in each SAR image is removed, and the VV, VH and angle band information in each SAR image is retained.

[0044] (2) Edge noise correction

[0045] Each SAR image processed in step (1) is subjected to edge noise correction to remove noise pixels at the edge of the image.

[0046] (3) Terrain correction

[0047] The DEM data of the area to be classified (available at: https: / / developers.google.com / earth-engine / datasets / catalog / CGIAR_SRTM90_V4) is used to mask the overlapped and shadowed areas in each SAR image processed in step (2) to avoid misidentification of water bodies caused by these areas.

[0048] (4) Noise filtering

[0049] The Refined Lee filter processing algorithm is applied to perform noise filtering on each SAR image processed in step (3) to reduce the clutter in the SAR image and weaken the uneven brightness distribution in the SAR image caused by interference and scattering effects.

[0050] (5)HAND mask

[0051] The HAND (height above nearest drainage) index characterizes the relative height of a pixel from the local water system and can be applied to disaster mapping, landform classification, and other fields. The HAND index is calculated based on each SAR image processed in step (4). Pixels with a HAND index greater than 20 are considered to be non-floodable pixels, and these areas are masked out to obtain several pre-processed SAR images.

[0052] 2. Feature Data Preparation

[0053] Based on each preprocessed SAR image, calculate VV / VH and sum. VV / VH is the ratio of VV to VH, and sum is the sum of VV and VH. These two bands clearly reflect the difference between water pixels and other pixels. VV, VH, angle, VV / VH, and sum of each preprocessed SAR image are used as feature data for the subsequent classification model.

[0054] 3. Sample Point Preparation

[0055] Since the surface of the water body is smooth, the backscatter coefficient of the water pixel in the SAR image is low. When visualizing the SAR image, the color of the water pixel will be darker than that of the non-water pixel. Based on this feature, the RGB channels are set with VV, VH, and VV / VH of the preprocessed SAR image, that is, VV is assigned to the red channel, VH is assigned to the green channel, and VV / VH is assigned to the blue channel. Each preprocessed SAR image is visualized, and the visualized SAR images are mosaicked to obtain the base map. 895 water sample points and 1429 non-water sample points are randomly selected from the base map. These 2324 sample points are divided into a training data set (1787 sample points) and a validation data set (537 sample points) according to a division ratio of 8:2. The distribution of the sample points is shown as follows. Figure 5 As shown, Figure 5 In the figure, blue sample points are water bodies and red sample points are non-water bodies.

[0056] 4. Construction and identification of classification models

[0057] In large-scale classification, block-based computation (which involves dividing the target area into a number of smaller blocks before performing computations) can yield faster and more accurate classification results. On the one hand, block-based computation avoids the spatial heterogeneity inherent in large-scale, global classification. Classification models trained using local features can more accurately capture the differences between water bodies and non-water bodies within the area, enhancing the accuracy of the classification model. On the other hand, after the target area is divided into blocks, computational speed can be accelerated through parallel processing and loops, improving overall computational efficiency. The operations in this step are performed in Google Earth Engine. The parameters for each local classification model include: the minimum number of samples per leaf node (default value 1), the sample ratio used in each tree training (default value 0.5), and the random seed (default value 0). These parameters do not affect the accuracy of the classification model and are generally set to their default values. In this step, the target area is first divided into blocks for computation. Then, a local classification model is established. The trained local classification model is then used to identify water bodies and non-water bodies. The specific operations are as follows:

[0058] A regular hexagon with a side length of 0.5° (geographic coordinate system unit) is used as a block unit to generate 23 block units that completely cover the area to be classified, such as Figure 6As shown. A local classification model is constructed in each block unit. The local classification model is constructed using the random forest algorithm, which is a supervised classification algorithm, as a classifier. Each local classification model is trained in its own block unit and its adjacent block units. During training, the feature data in step 2 is input as the feature data of the local classification model, and the data in the training data set in step 3 is input as the training data. The number of decision trees for each local classification model is set to 60. Each trained local classification model is used to identify the water bodies and non-water bodies in the block unit where it is located. That is, 23 trained local classification models are used to identify the water bodies and non-water bodies in the block unit where each local classification model is located. All the recognition results are mosaicked to obtain a preliminary recognition result image.

[0059] Figure 3 The process of training each local classification model and using the trained local classification model for recognition is shown. In each block unit ( Figure 3 Red hexagon) to build a local classification model, each local model is in its own block unit and the adjacent block unit of the block unit ( Figure 3 The training is carried out on the dotted hexagon in the figure. This increases the sample size of the training and enhances the classification ability of the local classification model. The local classification model after training is used to train the block unit ( Figure 3 water bodies and non-water bodies (red hexagons) are identified.

[0060] 5. Image post-processing

[0061] Adjacent objects often exhibit the same spectral characteristics, meaning that different types of objects in remote sensing images share the same spectral signature. This phenomenon manifests itself in classification results where non-water bodies are identified as water. To eliminate this misidentification, optimize the classification results, and remove small patches of noise, the preliminary identification image obtained in step 4 undergoes post-processing. This process involves filtering non-water pixels and smoothing them.

[0062] Acquire all 10m-resolution Sentinel-2MSI optical images of the area to be classified during the target time period. Select those with less than 10% cloud pixels from these images. Apply a declouding function to remove cloud pixels from each Sentinel-2MSI image. Then, calculate the MNDWI (Water Depth Index) and EVI (Enhanced Vegetation Index) for each cloud-removed Sentinel-2MSI image, and construct the MNDWI maximum and EVI minimum images. All Sentinel-2MSI image data includes spatial coordinate information, and the constructed MNDWI maximum and EVI minimum images also include spatial coordinate information. Pixels with a maximum MNDWI value less than 0 in the MNDWI maximum image or a minimum EVI value greater than 0 in the EVI minimum image are considered non-water pixels. Combined with their spatial coordinate information, these non-water pixels are removed from the preliminary identification image obtained in step 4, resulting in an image filtered for non-water pixels.

[0063] Since this embodiment aims to identify water bodies of larger scale in the area to be classified, the area of ​​0.3 hectares is used as the boundary. Water pixels with less than 30 adjacent pixels are removed from the image obtained after non-water pixel filtering. That is, only water bodies with an area larger than 0.3 hectares are retained. Small water patches with an area of ​​no more than 0.3 hectares are removed from the identification results to achieve a smoothing effect. This results in an image of the water distribution in the area to be classified, and the water body identification is completed. The results are shown in Figure 2. Figure 7 As shown, Figure 7 Medium blue represents water bodies and gray represents non-water bodies.

[0064] 6. Accuracy Verification

[0065] In order to test the accuracy and reliability of the classification and recognition results, the confusion matrix of the classification results is calculated using the sample points (537 sample points) in the validation data set. The calculation results are shown in Table 2.

[0066] Table 2 Confusion matrix

[0067]

[0068]

[0069] As shown in Table 2, of the 537 sample points in the validation dataset, there are 190 water body points and 347 non-water body points. Of the 190 water body points, 24 were identified as non-water bodies after the classification and recognition steps described above, while only one of the 347 non-water body points was identified as a water body after the classification and recognition steps described above. The overall classification accuracy (OA) of the classification results was 0.95, achieving a high classification accuracy. Furthermore, the Kappa index, which tests the consistency of the classification results, reached 0.90, indicating that the classification model's predictions are almost completely consistent with the actual classification results. Furthermore, the Producer's Accuracy (PA) for water bodies reached 0.87, indicating that 87% of the test samples were correctly classified. The User's Accuracy (UA) for water bodies reached 0.94, indicating that 94% of the samples classified as water bodies were correctly classified. In general, the method of the present invention has shown a high accuracy in water body identification based on satellite remote sensing images.

Claims

1. A water body identification method based on satellite remote sensing image block classification based on supervised learning algorithm, characterized by: The following steps are involved: (1) Image preprocessing A Sentinel-1SAR Level-1 GRD dataset of the area to be classified within the target time period is obtained, which is denoted as the target SAR dataset. The target SAR dataset includes several SAR images. Each SAR image in the target SAR dataset is preprocessed to obtain several preprocessed SAR images. The preprocessing includes band screening, edge noise correction, terrain correction, noise filtering, and hand index masking. Band screening refers to removing the HH and HV band information in the SAR image and retaining the VV, VH and angle band information in the SAR image; edge noise correction refers to performing edge noise correction on the SAR image after band screening; terrain correction refers to using the DEM data of the area to be classified to mask the overlap and shadow areas in the SAR image after edge noise correction; noise filtering refers to using the Refined Lee filter processing algorithm to process the SAR image after terrain correction to reduce the clutter in the SAR image; HAND index masking refers to calculating the HAND index based on the noise-filtered SAR image and masking out areas with a HAND index greater than 20; (2) Feature data preparation Based on each preprocessed SAR image, VV / VH and sum are calculated. VV / VH is the ratio of VV to VH, and sum is the sum of VV and VH. The VV, VH, angle, VV / VH and sum of each preprocessed SAR image are used as feature data; (3) Sample point preparation The RGB channels were set using the VV, VH, and VV / VH of the preprocessed SAR images. Each preprocessed SAR image was visualized, and the visualized SAR images were mosaicked to obtain a base map. Several water body sample points and non-water body sample points were randomly selected from the base map, and all sample points were divided into a training data set and a validation data set. (4) Construction and identification of classification models The base map of step (3) is divided into blocks by using block calculation to generate a plurality of block units that completely cover the area to be classified, and a local classification model is constructed in each block unit. The local classification model is constructed using a random forest algorithm as a classifier, and each local classification model is trained in its own block unit and its adjacent block units; during training, the feature data is input as the feature data of the local classification model, and the data in the training data set is input as the training data; The trained local classification models are used to identify the water and non-water bodies in the block units to obtain the preliminary recognition result images. (5) Image post-processing Acquire all Sentinel-2MSI optical images of the area to be classified within the target time period in which the percentage of cloud pixels is less than 10%, use the cloud removal function to remove the cloud pixels in each Sentinel-2MSI optical image, calculate the MNDWI index and EVI index of each Sentinel-2MSI optical image with cloud pixels removed, construct the MNDWI maximum value image and the EVI minimum value image, the pixels with the maximum MNDWI value < 0 in the MNDWI maximum value image or the pixels with the minimum EVI value > 0 in the EVI minimum value image are non-water body pixels, and the non-water body pixels in the preliminary recognition result image obtained in step (4) are removed in combination with the spatial coordinates of the non-water body pixels to obtain an image after non-water pixel filtering processing; then remove the water body pixels in the image after non-water pixel filtering processing in which the number of adjacent pixels is less than a preset value to obtain the water body distribution image of the area to be classified, thus completing water body recognition.

2. The method for water body identification based on satellite remote sensing image block classification based on supervised learning algorithm according to claim 1 is characterized in that: In step (4), the number of decision trees in each local classification model is 60 to 100.

3. The method for water body identification by block classification of satellite remote sensing images based on supervised learning algorithm according to claim 1 is characterized in that: In step (4), the block unit is a regular hexagonal block unit.

4. The method for water body identification by block classification of satellite remote sensing images based on supervised learning algorithm according to claim 3 is characterized in that: The side length of the regular hexagonal block unit in step (4) is 0.4° to 0.5°.

5. The method for identifying water bodies by block classification of satellite remote sensing images based on a supervised learning algorithm according to any one of claims 1 to 4, characterized in that: In step (3), all sample points are divided into a training data set and a validation data set according to the division ratio of (7-8):(3-2).

Images