High-temporal-spatial-resolution image generation method based on temporal-spatial fusion algorithm and related device

Abstract

The invention discloses a high-temporal-spatial-resolution image generation method based on a temporal-spatial fusion algorithm and a related device, and relates to the technical field of remote sensing image fusion, and the method comprises the steps: carrying out the preliminary screening of any target grid pixel in a moving window with the position of the target grid pixel as the center according to the reflectivity difference, and carrying out the preliminary screening of the position of the target grid pixel; according to the spectral distance, the spatial distance, the time distance and corresponding thresholds, screening to obtain a final similar grid pixel set; when the spectral distance is calculated, different coefficient constant combinations are used according to different wavebands of the similar grid pixels, so that a more accurate spectral distance calculation result is obtained; then, according to the weight matrix obtained through calculation, remote sensing images with high time and low spatial resolution and high spatial and low time resolution are effectively fused, the defects of an existing remote sensing product in temporal and spatial resolution are overcome, and particularly, low-cost remote sensing products usually have the defects of high time resolution, low spatial resolution and low time resolution. Or the spatial resolution is high and the time resolution is low.

Description

Technical Field

[0001] The present application relates to the field of remote sensing image fusion technology, and in particular to a method and related device for generating high spatiotemporal resolution images based on a spatiotemporal fusion algorithm. Background Art

[0002] In recent years, global warming, water scarcity, land degradation, biodiversity loss, and frequent sandstorms have posed serious threats to the human environment and become a major constraint on sustainable social and economic development. Therefore, continuous spatiotemporal monitoring of land features such as vegetation, water bodies, and soil is crucial for environmental protection and governance. While traditional manual field sampling can provide high-precision data, its time-consuming and labor-intensive nature limits its application. Remote sensing, on the other hand, offers high-precision monitoring at a lower cost. However, existing satellite imagery (such as Sentinel-3), while offering high temporal resolution (1 day), has a spatial resolution of 300 meters, making it difficult to monitor small water bodies and land details, making it difficult to accurately capture changes in land features. In contrast, Sentinel-2 imagery (10, 20, and 60 meters) can more accurately monitor detailed changes in land features, but its 5-day temporal resolution limits its ability to track phenological changes in real time. Furthermore, while the GF series imagery offers high spatiotemporal resolution, its high cost prevents widespread application for long-term environmental monitoring, necessitating technological breakthroughs in this area.

[0003] Currently, spatiotemporal fusion models are an effective solution to addressing the shortage of high-resolution imagery. These models fuse high- and low-resolution imagery, generating high-resolution images from low-resolution images when high-resolution images are unavailable, thereby supplementing the data gap. Recently, deep learning methods have also been applied to spatiotemporal fusion models. While these methods can improve prediction accuracy, their extensive application is limited by the large amount of training data and time required. Therefore, combining the advantages of high-resolution imagery with high-resolution imagery while reducing reliance on training data is crucial for achieving continuous spatiotemporal monitoring. Summary of the Invention

[0004] The purpose of this application is to provide a high-temporal-resolution image generation method and related devices based on a spatiotemporal fusion algorithm, which can reduce the computational difficulty and improve the computational efficiency while ensuring the accuracy of the fused remote sensing monitoring image.

[0005] To achieve the above objectives, this application provides the following solutions:

[0006] In a first aspect, the present application provides a method for generating high spatiotemporal resolution images based on a spatiotemporal fusion algorithm, comprising the following steps:

[0007] The initial high spatial resolution images and the initial low spatial resolution images with similar dates are preprocessed in turn to obtain the preprocessed high spatial resolution images and the preprocessed low spatial resolution images; the projection coordinate system and grid size of the preprocessed high spatial resolution images and the preprocessed low spatial resolution images are the same; the initial high spatial resolution images and the initial low spatial resolution images are both remote sensing monitoring images including several bands.

[0008] For any target grid pixel of the preprocessed low spatial resolution image, a preliminary screening is performed according to the reflectivity difference in the moving window of the corresponding position in the preprocessed high spatial resolution image to obtain a set of pre-screened similar grid pixels of the target grid pixel; the moving window at the corresponding position is a moving window of a preset size centered on the position of the target grid pixel, and the moving window includes several grid pixels.

[0009] Based on the spectral distance, spatial distance, temporal distance and corresponding thresholds between each similar raster pixel in the initial screening similar raster pixel set and the target raster pixel, the final similar raster pixel set of the target raster pixel is screened out; when calculating the spectral distance between any similar raster pixel and the target raster pixel, different spectral distance calculation coefficients and spectral distance calculation constants are used according to the different bands of the similar raster pixels.

[0010] According to the spectral distance, spatial distance and temporal distance between each similar grid pixel and the target grid pixel in the final similar grid pixel set, the weight matrix of each similar grid pixel is calculated.

[0011] According to the surface reflectance value of each similar grid pixel and the weight matrix of each similar grid pixel, the surface reflectance of the target grid pixel is calculated, and the high spatial resolution fusion of the preprocessed low spatial resolution image is completed to obtain a high spatiotemporal resolution image.

[0012] Optionally, the preprocessing performed on the initial high spatial resolution image and the initial low spatial resolution image with similar dates respectively includes: performing atmospheric correction processing using the ACOLITE atmospheric correction preprocessing software provided by the DSF atmospheric correction algorithm, selecting the WGS_1984_UTM_Zone_51N projection system for reprojection, and resampling using the cubic convolution interpolation method.

[0013] Optionally, when atmospheric correction is performed on initial high spatial resolution images and initial low spatial resolution images with similar dates, the surface reflectance is calculated according to the following formula:

[0014]

[0015] Among them, ρ s is the surface reflectivity, ρt is the reflectivity of the top of the atmosphere, t g is the atmospheric gas transmittance, ρ path is the atmospheric path reflectivity, ρ sky is the sky diffuse reflectivity, t du is the bidirectional diffuse atmospheric transmittance, s a is the spherical albedo of the atmosphere.

[0016] Optionally, preliminary screening is performed based on reflectivity differences in a moving window at a corresponding position of the preprocessed high spatial resolution image according to the following formula:

[0017]

[0018] Among them, H(x i ,y j ,t k ) is t k The position of the pre-processed high spatial resolution image at the moment is (x i ,y j ), t k The surface reflectance value of the central grid pixel of the high spatial resolution image after preprocessing at that moment, w is the row and column size of the moving window, σ is the standard deviation of the surface reflectance within the moving window at the corresponding position of the high spatial resolution image after preprocessing, and m is the number of ground object categories.

[0019] Optionally, the spectral distance between any similar grid pixel in the initial screening similar grid pixel set and the target grid pixel is calculated according to the following formula:

[0020] S ijk =|aH(x i ,y j ,t k )+bL(x i ,y j ,t k )|.

[0021] Among them, S ijk t k The position of the similar grid pixel set in the initial screening at the moment is (x i ,y j ) is the spectral distance between the similar grid pixel and the target grid pixel, a is the spectral distance calculation coefficient, b is the spectral distance calculation constant, H(x i ,y j ,t k ) is t k The position of the pre-processed high spatial resolution image at the moment is (x i ,y j) of the grid pixel surface reflectance value, L(x i ,y j ,t k ) is t k The position of the pre-processed low spatial resolution image at the moment is (x i ,y j ) raster pixels.

[0022] The temporal distance between any similar grid pixel in the initial screening similar grid pixel set and the target grid pixel is calculated according to the following formula:

[0023] T ijk =|L(x i ,y j ,t k )-L(x i ,y j ,t0)|.

[0024] Among them, T ijk t k The position of the similar grid pixel set in the initial screening at the moment is (x i ,y j ) is the time distance between the similar grid pixel and the target grid pixel, L(x i ,y j ,t0) is the position (x i ,y j ) raster pixels.

[0025] The spatial distance between any similar grid pixel in the initial screening similar grid pixel set and the target grid pixel is calculated according to the following formula:

[0026]

[0027] Among them, D ijk t k The position of the similar grid pixel set in the initial screening at the moment is (x i ,y j ) is the spatial distance between the similar grid pixels and the target grid pixels, and w is the row and column size of the moving window.

[0028] Optionally, the weight matrix of any similar grid pixel can be calculated according to the following formula:

[0029]

[0030] Among them, W ijk t k The position of the similar grid pixel set in the initial screening at the moment is (x i ,y j) is the weight matrix of similar grid pixels, S ijk 、T ijk and D ijk t k The position of the similar grid pixel set in the initial screening at the moment is (x i ,y j ) is the spectral distance, temporal distance and spatial distance between the similar grid pixels and the target grid pixels, w is the row and column size of the moving window, and n is the number of remote sensing monitoring image pairs selected in the calculation.

[0031] The surface reflectance of the target grid pixel is calculated according to the following formula:

[0032]

[0033] in, t k The surface reflectance of the target grid pixel at the moment, L(x i ,y j ,t0) is the position (x i ,y j ) of the grid pixel surface reflectance value, H(x i ,y j ,t k ) is t k The position of the pre-processed high spatial resolution image at the moment is (x i ,y j ) of the grid pixel surface reflectance value, L(x i ,y j ,t k ) is t k The position of the pre-processed low spatial resolution image at the moment is (x i ,y j ) raster pixels.

[0034] In a second aspect, the present application provides a high spatiotemporal resolution image generation system based on a spatiotemporal fusion algorithm, comprising:

[0035] The remote sensing monitoring image preprocessing module is used to preprocess the initial high spatial resolution images and the initial low spatial resolution images with similar dates in sequence to obtain the preprocessed high spatial resolution images and the preprocessed low spatial resolution images; the projection coordinate system and grid size of the preprocessed high spatial resolution images and the preprocessed low spatial resolution images are the same; the initial high spatial resolution images and the initial low spatial resolution images are both remote sensing monitoring images including several bands.

[0036] The similar grid pixel preliminary screening module is used to perform preliminary screening according to the reflectivity difference for any target grid pixel in the preprocessed low spatial resolution image in the moving window at the corresponding position in the preprocessed high spatial resolution image, and obtain a set of preliminary screened similar grid pixels of the target grid pixel; the moving window at the corresponding position is a moving window of a preset size centered on the position of the target grid pixel, and the moving window includes a number of grid pixels.

[0037] The similar grid pixel final screening module is used to screen the final similar grid pixel set of the target grid pixel based on the spectral distance, spatial distance, temporal distance and corresponding threshold between each similar grid pixel in the initial screening similar grid pixel set and the target grid pixel; when calculating the spectral distance between any similar grid pixel and the target grid pixel, different spectral distance calculation coefficients and spectral distance calculation constant terms are used according to the different bands of the similar grid pixels.

[0038] The pixel weight matrix technology module is used to calculate the weight matrix of each similar grid pixel based on the spectral distance, spatial distance and temporal distance between each similar grid pixel and the target grid pixel in the final similar grid pixel set.

[0039] The surface reflectance fusion calculation module is used to calculate the surface reflectance of the target grid pixel based on the surface reflectance value of each similar grid pixel and the weight matrix of each similar grid pixel, complete the high spatial resolution fusion of the preprocessed low spatial resolution image, and obtain a high spatiotemporal resolution image.

[0040] In a third aspect, the present application provides a computer device comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the method for generating high spatiotemporal resolution images based on the spatiotemporal fusion algorithm described above.

[0041] In a fourth aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method for generating high spatiotemporal resolution images based on the spatiotemporal fusion algorithm described above.

[0042] 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 method for generating high spatiotemporal resolution images based on the spatiotemporal fusion algorithm described above.

[0043] According to the specific embodiments provided in this application, this application discloses the following technical effects:

[0044] The present application provides a method and related device for generating high spatiotemporal resolution images based on a spatiotemporal fusion algorithm. In this method, an initial high spatial resolution image and an initial low spatial resolution image with similar dates are first preprocessed in sequence so that the projection coordinate system and grid size of the two are the same, which is convenient for subsequent processing; then, for any target grid pixel, a final set of similar grid pixels is screened in a moving window centered on the position of the target grid pixel according to reflectance difference, spectral distance, spatial distance, temporal distance and corresponding threshold; when calculating the spectral distance, different coefficient constant combinations are used according to the different bands of each similar grid pixel to obtain a more accurate calculation result; then, a corresponding weight matrix is ​​calculated based on the obtained spectral distance, spatial distance and temporal distance; finally, based on the surface reflectance value of each similar grid pixel and its weight matrix calculation, the high spatial resolution fusion of the preprocessed low spatial resolution image is completed to obtain a high spatiotemporal resolution image. This application effectively integrates remote sensing images with high temporal and low spatial resolution and high spatial and low temporal resolution to obtain high-precision remote sensing images with high temporal and spatial resolution. While ensuring the accuracy of the predicted image, the calculation difficulty is low and the calculation efficiency is high, which solves the shortcomings of existing remote sensing products in temporal and spatial resolution, especially the limitations of low-cost remote sensing products that usually have high temporal resolution and low spatial resolution, or high spatial resolution and low temporal resolution. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0046] Figure 1 A flowchart of a method for generating high spatiotemporal resolution images based on a spatiotemporal fusion algorithm is provided in one embodiment of the present application.

[0047] Figure 2 This is a schematic diagram of the regional division of Zhukui Reservoir, a study area in a specific embodiment of the present application.

[0048] Figure 3 This is a comparison diagram of the preprocessing result and the initial remote sensing image of the Zhukui Reservoir in a specific embodiment of the present application.

[0049] Figure 4 This is a true color comparison diagram of the fusion results in a method for generating high spatiotemporal resolution images based on a spatiotemporal fusion algorithm provided in a specific embodiment of the present application.

[0050] Figure 5This is a boxplot showing the band-by-band comparison of fusion results in a method for generating high spatiotemporal resolution images based on a spatiotemporal fusion algorithm provided in one embodiment of the present application.

[0051] Figure 6 A comparison chart of the fusion results of a high spatiotemporal resolution image generation method based on a spatiotemporal fusion algorithm provided in one embodiment of the present application and the fusion results of other methods.

[0052] Figure 7 A scatter plot comparing inversion results in a method for generating high spatiotemporal resolution images based on a spatiotemporal fusion algorithm provided in one embodiment of the present application.

[0053] Figure 8 A schematic diagram of the time series of inversion results in a method for generating high spatiotemporal resolution images based on a spatiotemporal fusion algorithm provided in one embodiment of the present application.

[0054] Figure 9 A schematic diagram of the functional modules of a high spatiotemporal resolution image generation system based on a spatiotemporal fusion algorithm provided in one embodiment of the present application.

[0055] Figure 10 A schematic diagram of the structure of a computer device provided in one embodiment of the present application. DETAILED DESCRIPTION

[0056] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0057] In order to make the above-mentioned purposes, features and advantages of the present application more obvious and easy to understand, the present application is further described in detail below with reference to the accompanying drawings and specific implementation methods.

[0058] The present application provides a method for generating high spatiotemporal resolution images based on a spatiotemporal fusion algorithm. In an exemplary embodiment, Figure 1 As shown, the following steps are included:

[0059] S1. Preprocess the initial high-resolution image and the initial low-resolution image with similar dates to obtain a preprocessed high-resolution image and a preprocessed low-resolution image. The projection coordinate system and grid size of the preprocessed high-resolution image and the preprocessed low-resolution image are the same. Both the initial high-resolution image and the initial low-resolution image are remote sensing monitoring images that include multiple bands. Specifically, the initial high-resolution image uses image data collected by the Sentinel-2MSI multispectral imager, and the initial low-resolution image uses image data collected by the Sentinel-3OLCI ocean and land color instrument.

[0060] In a specific embodiment, the preprocessing performed on the initial high spatial resolution image and the initial low spatial resolution image with similar dates in step S1 includes: screening for cloud-free lighting conditions, performing atmospheric correction processing using the ACOLITE atmospheric correction preprocessing software provided by the DSF atmospheric correction algorithm, selecting the WGS_1984_UTM_Zone_51N projection system for reprojection, and resampling using the cubic convolution interpolation method.

[0061] Specifically, cloud-free images within the study area were screened, and primary products with poor lighting conditions were removed. Atmospheric correction was then uniformly performed using the ACOLITE atmospheric correction preprocessing software provided by the Dark Spectrum Fitting (DSF) atmospheric correction algorithm. During the calculation process, specific values ​​(such as water vapor transmittance, Rayleigh optical depth, and atmospheric path reflectivity) were selected from a lookup table (LUT) generated using 6SV. The specific calculation method for surface reflectivity is as follows:

[0062]

[0063] Among them, ρ s is the surface reflectivity, ρ t is the reflectivity of the top of the atmosphere, t g is the atmospheric gas transmittance, ρ path is the atmospheric path reflectivity, ρ sky is the sky diffuse reflectivity, t du is the bidirectional diffuse atmospheric transmittance, s a is the spherical albedo of the atmosphere. The above method can eliminate the influence of flares on images.

[0064] To ensure a one-to-one correspondence between the raster pixels of the two remote sensing images, this embodiment further performs reprojection and resampling operations on the remote sensing images. After comparative analysis, this embodiment first selects the WGS_1984_UTM_Zone_51N projection system to reproject the two remote sensing images. Then, cubic convolution interpolation is used to complete the image resampling. This ensures that the projection coordinate system of the two remote sensing images is the same, and the grid size is 10×10m, thus ensuring the spatial consistency of the image data.

[0065] The above preprocessing steps provide a solid data foundation for spatiotemporal fusion analysis of images and remote sensing information extraction, significantly improving the accuracy and reliability of the fusion results and meeting the needs of various remote sensing application scenarios.

[0066] Next, we use the improved STARFM model to fuse remote sensing image data. First, we need to determine the STARFM model's input parameters, including the side length of the moving window, the spatial impact factor, and the number of feature types. After comparative analysis, this example selected a moving window length of 3000m, approximately equal to the size of 10×10 Sentinel-3 pixels, which ensures both prediction accuracy and computational efficiency. We also selected a spatial impact factor of 500 to balance the relationship between the strictness of spatial distance in pixel selection and the number of pixels. Furthermore, we set the number of feature types to 4 to ensure an appropriate balance between matching accuracy and pixel number.

[0067] S2. For any target raster pixel in the preprocessed low spatial resolution image, perform a preliminary screening based on reflectivity differences in a moving window at the corresponding position in the preprocessed high spatial resolution image to obtain a pre-screened set of similar raster pixels to the target raster pixel; the moving window at the corresponding position is a moving window of a preset size centered on the position of the target raster pixel, and includes a number of raster pixels. Specifically, in this embodiment, preliminary screening based on reflectivity differences is performed in the moving window at the corresponding position in the preprocessed high spatial resolution image according to the following formula:

[0068]

[0069] Among them, H(x i ,y j ,t k ) is t k The position of the pre-processed high spatial resolution image at the moment is (x i ,y j ), t kThe surface reflectance value of the central grid pixel of the high spatial resolution image after preprocessing at that moment, w is the row and column size of the moving window, σ is the standard deviation of the surface reflectance within the moving window at the corresponding position of the high spatial resolution image after preprocessing, and m is the number of ground object categories.

[0070] S3. Based on the spectral distance, spatial distance, temporal distance, and corresponding thresholds between each similar grid pixel in the initial screening set of similar grid pixels and the target grid pixel, a final set of similar grid pixels of the target grid pixel is screened and obtained; when calculating the spectral distance between any similar grid pixel and the target grid pixel, different spectral distance calculation coefficients and spectral distance calculation constants are used depending on the wavebands of the similar grid pixels. Specifically, in this embodiment, the spectral distance between any similar grid pixel in the initial screening set of similar grid pixels and the target grid pixel is calculated according to the following formula:

[0071] S ijk =|aH(x i ,y j ,t k )+bL(x i ,y j ,t k )|.

[0072] Among them, S ijk t k The position of the similar grid pixel set in the initial screening at the moment is (x i ,y j ) is the spectral distance between the similar grid pixel and the target grid pixel, a is the spectral distance calculation coefficient, b is the spectral distance calculation constant, H(x i ,y j ,t k ) is t k The position of the pre-processed high spatial resolution image at the moment is (x i ,y j ) of the grid pixel surface reflectance value, L(x i ,y j ,t k ) is t k The position of the pre-processed low spatial resolution image at the moment is (x i ,y j ) raster pixels.

[0073] In this embodiment, by modeling the corresponding bands of the two remote sensing images and performing linear regression calculations, the spectral distance calculation coefficient a and the constant term b of each band are determined. The calculation results are shown in the table below.

[0074] Table 1 Spectral distance calculation coefficients and constant terms for each band

[0075] Central wavelength (nm) a b 443 1.13 0.0002 490 0.98 -0.0001 560 0.85 0.0035 665 0.87 0.0019 705 0.87 0.0025 740 1.16 0.0011 783 1.04 0.0011 842 1.14 0.0006 865 1.15 0.0005

[0076] The temporal distance between any similar grid pixel in the initial screening similar grid pixel set and the target grid pixel is calculated according to the following formula:

[0077] T ijk =|L(x i ,y j ,t k )-L(x i ,y j ,t0)|.

[0078] Among them, T ijk t k The position of the similar grid pixel set in the initial screening at the moment is (x i ,y j ) is the time distance between the similar grid pixel and the target grid pixel, L(x i ,y j ,t0) is the position (x i ,y j ) raster pixels.

[0079] The spatial distance between any similar grid pixel in the initial screening similar grid pixel set and the target grid pixel is calculated according to the following formula:

[0080]

[0081] Among them, D ijk t k The position of the similar grid pixel set in the initial screening at the moment is (x i ,y j ) is the spatial distance between the similar grid pixels and the target grid pixels, and w is the row and column size of the moving window.

[0082] When the spectral distance, temporal distance, and spatial distance are less than their corresponding thresholds, they are identified as similar pixels and participate in subsequent calculations.

[0083] S4. Calculate the weight matrix of each similar grid pixel based on the spectral distance, spatial distance, and temporal distance between each similar grid pixel in the final similar grid pixel set and the target grid pixel. In this embodiment, the weight matrix of any similar grid pixel is calculated according to the following formula:

[0084]

[0085] Among them, W ijk t k The position of the similar grid pixel set in the initial screening at the moment is (x i ,y j) is the weight matrix of similar grid pixels, S ijk 、T ijk and D ijk t k The position of the similar grid pixel set in the initial screening at the moment is (x i ,y j ) is the spectral distance, temporal distance and spatial distance between the similar grid pixels and the target grid pixels, w is the row and column size of the moving window, and n is the number of remote sensing monitoring image pairs selected for calculation. Specifically, when performing the fusion of high and low spatial resolution images, it is necessary to select at least one pair of remote sensing monitoring image pairs (high spatial and low temporal remote sensing monitoring images and high temporal and low spatial remote sensing monitoring images) for calculation, and n is the number of remote sensing monitoring image pairs selected for calculation. For example, the time of the first pair of remote sensing monitoring image pairs is t1, the time of the second pair is t2, and so on. The time of the last pair of image pairs is t n .

[0086] S5. Calculate the surface reflectance of the target grid pixel based on the surface reflectance values ​​of each similar grid pixel and the weight matrix of each similar grid pixel, and complete the high spatial resolution fusion of the pre-processed low spatial resolution image to obtain a high spatiotemporal resolution image. Specifically, the surface reflectance of the target grid pixel is calculated according to the following formula:

[0087]

[0088] in, t k The surface reflectance of the target grid pixel at the moment, L(x i ,y j ,t0) is the position (x i ,y j ) of the grid pixel surface reflectance value, H(x i ,y j ,t k ) is t k The position of the pre-processed high spatial resolution image at the moment is (x i ,y j ) of the grid pixel surface reflectance value, L(x i ,y j ,t k ) is t k The position of the pre-processed low spatial resolution image at the moment is (x i ,y j ). By traversing the positions of all raster pixels in sequence, the surface reflectance of all raster pixels can be calculated, and finally a complete high-temporal and spatial resolution remote sensing image can be obtained.

[0089] The method proposed in this application is applied to a specific example. Zhukui Reservoir (39°50'-39°52'N, 122°45'-122°53'E) is located on the west tributary of Zhuanghe River in Taipingling Township, Zhuanghe City, Liaoning Province. Figure 2 As shown, the reservoir area is relatively small, about 17 km 2 . In recent years, the water quality in the reservoir has deteriorated significantly during the flood season, and the algal bloom phenomenon has become increasingly serious. The water quality inversion conditions are good, and the accuracy of the generated high-temporal and spatial resolution image Sentinel2-like dataset can be verified. The measured data was obtained through field sampling and laboratory analysis, and the samples were taken once a month on average during the non-freezing period. After the above-mentioned solution of this application was applied to the Zhukuima Reservoir, the comparison between the pre-processing results of step S1 and the original image is shown in the figure below. Figure 3 shown.

[0090] The test results of Zhukuma Reservoir are as follows Figure 4 and Figure 5 As shown in the figure, the accuracy of the water body remote sensing reflectance in the fusion results is verified using indicators such as root mean square error (RMSE), structural similarity (SSIM), peak signal-to-noise ratio (PSNR) and dimensionless global integrated relative error in spectral / spatial dimensions (ERGAS). Between 2016 and 2024, the test results showed that RMSE < 0.015sr -1 , PSNR>37dB, SSIM>0.8, dimensionless global comprehensive relative error ERGAS in spectral dimension spectral <3.8, dimensionless global comprehensive relative error ERGAS in spatial dimension spatial <3.3, indicating that the method of this embodiment has a high fusion accuracy. By comparing with the ESTARFM method, Figure 6 As shown, it is found that the high spatiotemporal resolution image Sentinel2-like generated by the improved STARFM method adopted in this embodiment is closer in color to the real Sentinel-2 image and has higher fusion accuracy.

[0091] The test results of inversion of the Sentinel2-like dataset generated by the method of this embodiment are as follows: Figure 7 As shown in the figure, the inversion model is applied to the Sentinel-2 and Sentinel2-like datasets, and the correlation coefficient R 2 The values ​​are 0.65 and 0.59 respectively, which can meet the accuracy requirements of long-term inversion of chlorophyll a (Chla) concentration in water bodies. By comparing and analyzing the inversion results of Sentinel-2, Sentinel-3 and Sentinel2-like images, we can see that Figure 8As shown in the figure, it is concluded that Sentinel2-like images can accurately capture water details, and the obtained Chla value is close to that of Sentinel-2. It can be effectively used in long-term monitoring of water bodies and ecological assessment, and has high practical value.

[0092] Based on the same inventive concept, embodiments of the present application also provide a system for implementing the aforementioned method for generating high spatiotemporal resolution images based on a spatiotemporal fusion algorithm. The solution provided by this system is similar to the solution described in the aforementioned method. Therefore, the specific limitations in one or more system embodiments provided below can be found in the aforementioned limitations of the method for generating high spatiotemporal resolution images based on a spatiotemporal fusion algorithm, and will not be further elaborated here.

[0093] In an exemplary embodiment, Figure 9 As shown in the figure, a high spatiotemporal resolution image generation system based on spatiotemporal fusion algorithm is provided, which includes the following functional modules:

[0094] The remote sensing monitoring image preprocessing module is used to preprocess the initial high spatial resolution images and the initial low spatial resolution images with similar dates in sequence to obtain the preprocessed high spatial resolution images and the preprocessed low spatial resolution images; the projection coordinate system and grid size of the preprocessed high spatial resolution images and the preprocessed low spatial resolution images are the same; the initial high spatial resolution images and the initial low spatial resolution images are both remote sensing monitoring images including several bands.

[0095] The similar grid pixel preliminary screening module is used to perform preliminary screening according to the reflectivity difference for any target grid pixel in the preprocessed low spatial resolution image in the moving window at the corresponding position in the preprocessed high spatial resolution image, and obtain a set of preliminary screened similar grid pixels of the target grid pixel; the moving window at the corresponding position is a moving window of a preset size centered on the position of the target grid pixel, and the moving window includes a number of grid pixels.

[0096] The similar grid pixel final screening module is used to screen the final similar grid pixel set of the target grid pixel based on the spectral distance, spatial distance, temporal distance and corresponding threshold between each similar grid pixel in the initial screening similar grid pixel set and the target grid pixel; when calculating the spectral distance between any similar grid pixel and the target grid pixel, different spectral distance calculation coefficients and spectral distance calculation constant terms are used according to the different bands of the similar grid pixels.

[0097] The pixel weight matrix technology module is used to calculate the weight matrix of each similar grid pixel based on the spectral distance, spatial distance and temporal distance between each similar grid pixel and the target grid pixel in the final similar grid pixel set.

[0098] The surface reflectance fusion calculation module is used to calculate the surface reflectance of the target grid pixel based on the surface reflectance value of each similar grid pixel and the weight matrix of each similar grid pixel, complete the high spatial resolution fusion of the preprocessed low spatial resolution image, and obtain a high spatiotemporal resolution image.

[0099] certainly, Figure 9 The architecture shown is only exemplary and can be omitted according to actual needs when implementing different functions. Figure 9 One or at least two components of the system shown.

[0100] In an exemplary embodiment, a computer device is provided. The computer device may be a server or a terminal. The internal structure diagram thereof may be as follows: Figure 10 As shown. The computer device includes a processor, a memory, an input / output interface (Input / Output, abbreviated as I / O) and a communication interface. The processor, memory and 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 used 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 computer program in the non-volatile storage medium. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a high spatiotemporal resolution image generation method based on a spatiotemporal fusion algorithm provided in the above embodiment can be implemented.

[0101] Those skilled in the art will understand that Figure 10 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than shown in the figure, or combine certain components, or have a different component arrangement.

[0102] In an exemplary embodiment, a computer device is further provided, including a memory and a processor. The memory stores a computer program, and the processor implements the steps in the above method embodiments when executing the computer program.

[0103] In an exemplary embodiment, a computer-readable storage medium is provided, storing a computer program. When the computer program is executed by a processor, the steps in the above-mentioned method embodiments are implemented.

[0104] In an exemplary embodiment, a computer program product is provided, including a computer program. When the computer program is executed by a processor, the steps in the above method embodiments are implemented.

[0105] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0106] Those skilled in the art will understand that all or part of the processes in the above-mentioned embodiment methods can be implemented by instructing the relevant hardware through a computer program, and 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 embodiments of the above-mentioned methods. Among them, any reference to memory, database or other media used in the embodiments provided in this application may include at least one of non-volatile and volatile memory. Non-volatile memory may include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory may include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM may be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM).

[0107] The databases involved in the various embodiments provided herein may include at least one of a relational database and a non-relational database. Non-relational databases may include, but are not limited to, distributed databases based on blockchains. The processors involved in the various embodiments provided herein may include, but are not limited to, general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic units, data processing logic units based on quantum computing, and the like.

[0108] The technical features of the above embodiments can be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0109] This document uses specific examples to illustrate the principles and implementation methods of this application. The description of the above examples is only intended to help understand the method and core concept of this application. At the same time, for those skilled in the art, based on the concept of this application, there may be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting this application.

Claims

1. A method for generating high spatiotemporal resolution images based on a spatiotemporal fusion algorithm, characterized in that: include: Preprocessing the initial high spatial resolution image and the initial low spatial resolution image with similar dates in sequence to obtain a preprocessed high spatial resolution image and a preprocessed low spatial resolution image; the projection coordinate system and grid size of the preprocessed high spatial resolution image and the preprocessed low spatial resolution image are the same; and both the initial high spatial resolution image and the initial low spatial resolution image are remote sensing monitoring images including multiple bands; For any target grid pixel of the preprocessed low spatial resolution image, preliminary screening is performed according to reflectivity differences in a moving window at a corresponding position in the preprocessed high spatial resolution image to obtain a set of pre-screened similar grid pixels of the target grid pixel; the moving window at the corresponding position is a moving window of a preset size centered at the position of the target grid pixel, and the moving window includes a plurality of grid pixels; A final set of similar raster pixels of the target raster pixel is screened and obtained based on the spectral distance, spatial distance, temporal distance and corresponding threshold between each similar raster pixel in the initially screened set of similar raster pixels and the target raster pixel; when calculating the spectral distance between any similar raster pixel and the target raster pixel, different spectral distance calculation coefficients and spectral distance calculation constant terms are used according to different wavebands of the similar raster pixels; Calculating a weight matrix of each similar grid pixel according to the spectral distance, spatial distance and temporal distance between each similar grid pixel in the final similar grid pixel set and the target grid pixel; According to the surface reflectance value of each similar grid pixel and the weight matrix of each similar grid pixel, the surface reflectance of the target grid pixel is calculated, and the high spatial resolution fusion of the preprocessed low spatial resolution image is completed to obtain a high spatiotemporal resolution image.

2. The method for generating high spatiotemporal resolution images based on a spatiotemporal fusion algorithm according to claim 1, characterized in that: The preprocessing of the initial high spatial resolution images and the initial low spatial resolution images with similar dates respectively includes: atmospheric correction using the ACOLITE atmospheric correction preprocessing software provided by the DSF atmospheric correction algorithm, reprojection using the WGS_1984_UTM_Zone_51N projection system, and resampling using the cubic convolution interpolation method.

3. The method for generating high spatiotemporal resolution images based on a spatiotemporal fusion algorithm according to claim 2, characterized in that: When atmospheric correction is performed on initial high spatial resolution images and initial low spatial resolution images with similar dates, the surface reflectance is calculated according to the following formula: Among them, ρ s is the surface reflectivity, ρ t is the reflectivity of the top of the atmosphere, t g is the atmospheric gas transmittance, ρ path is the atmospheric path reflectivity, ρ sky is the sky diffuse reflectivity, t du is the bidirectional diffuse atmospheric transmittance, s a is the spherical albedo of the atmosphere.

4. The method for generating high spatiotemporal resolution images based on a spatiotemporal fusion algorithm according to claim 1, characterized in that: A preliminary screening is performed based on the reflectivity difference in the moving window at the corresponding position of the pre-processed high spatial resolution image according to the following formula: Among them, H(x i ,y j ,t k ) is t k The position of the pre-processed high spatial resolution image at the moment is (x i ,y j ), t k The surface reflectance value of the central grid pixel of the preprocessed high spatial resolution image at the moment, w is the row and column size of the moving window, σ is the standard deviation of the surface reflectance within the moving window at the corresponding position of the preprocessed high spatial resolution image, and m is the number of ground object categories.

5. The method for generating high spatiotemporal resolution images based on a spatiotemporal fusion algorithm according to claim 1, characterized in that: The spectral distance between any similar grid pixel in the initial screening similar grid pixel set and the target grid pixel is calculated according to the following formula: S ijk =|aH(x i ,y j ,t k )+b-L(x i ,y j ,t k )|; Among them, S ijk t k The position of the similar grid pixel set in the initial screening at the moment is (x i ,y j ) is the spectral distance between the similar grid pixel and the target grid pixel, a is the spectral distance calculation coefficient, b is the spectral distance calculation constant, H(x i ,y j ,t k ) is t k The position of the pre-processed high spatial resolution image at the moment is (x i ,y j ) of the grid pixel surface reflectance value, L(x i ,y j ,t k ) is t k The position of the pre-processed low spatial resolution image at the moment is (x i ,y j ) of the raster pixels; The time distance between any similar grid pixel in the initial screening similar grid pixel set and the target grid pixel is calculated according to the following formula: T ijk =|L(x i ,y j ,t k )-L(x i ,y j ,t0)|; Among them, T ijk t k The position of the similar grid pixel set in the initial screening at the moment is (x i ,y j ) of the similar grid pixel and the time distance of the target grid pixel, L(x i ,y j ,t0) is the position (x i ,y j ) of the raster pixels; The spatial distance between any similar grid pixel in the initial screening similar grid pixel set and the target grid pixel is calculated according to the following formula: Among them, D ijk t k The position of the similar grid pixel set in the initial screening at the moment is (x i ,y j ) is the spatial distance between the similar grid pixel and the target grid pixel, and w is the row and column size of the moving window.

6. The method for generating high spatiotemporal resolution images based on a spatiotemporal fusion algorithm according to claim 1, characterized in that: The weight matrix of any similar grid pixel is calculated according to the following formula: Among them, W ijk t k The position of the similar grid pixel set in the initial screening at the moment is (x i ,y j ) is the weight matrix of similar grid pixels, S ijk 、T ijk and D ijk t k The position of the similar grid pixel set in the initial screening at the moment is (x i ,y j ) is the spectral distance, temporal distance and spatial distance between the similar grid pixels and the target grid pixels, w is the row and column size of the moving window, and n is the number of remote sensing monitoring image pairs selected for calculation; The surface reflectance of the target grid pixel is calculated according to the following formula: in, t k The surface reflectance of the target grid pixel at the moment, L(x i ,y j ,t0) is the position (x i ,y j ) of the grid pixel surface reflectance value, H(x i ,y j ,t k ) is t k The position of the pre-processed high spatial resolution image at the moment is (x i ,y j ) of the grid pixel surface reflectance value, L(x i ,y j ,t k ) is t k The position of the pre-processed low spatial resolution image at the moment is (x i ,y j ) raster pixels.

7. A high spatiotemporal resolution image generation system based on spatiotemporal fusion algorithm, characterized in that: include: A remote sensing monitoring image preprocessing module is used to preprocess the initial high spatial resolution image and the initial low spatial resolution image of similar dates in sequence to obtain a preprocessed high spatial resolution image and a preprocessed low spatial resolution image; the projection coordinate system and grid size of the preprocessed high spatial resolution image and the preprocessed low spatial resolution image are the same; and the initial high spatial resolution image and the initial low spatial resolution image are both remote sensing monitoring images including multiple bands; A similar grid pixel preliminary screening module is configured to perform preliminary screening of any target grid pixel of the preprocessed low spatial resolution image according to reflectivity differences in a moving window at a corresponding position in the preprocessed high spatial resolution image, thereby obtaining a set of pre-screened similar grid pixels of the target grid pixel; the moving window at the corresponding position is a moving window of a preset size centered on the position of the target grid pixel, and the moving window includes a plurality of grid pixels; A similar grid pixel final screening module is used to screen and obtain a final similar grid pixel set of the target grid pixel based on the spectral distance, spatial distance, temporal distance and corresponding threshold value between each similar grid pixel in the preliminary screening similar grid pixel set and the target grid pixel; when calculating the spectral distance between any similar grid pixel and the target grid pixel, different spectral distance calculation coefficients and spectral distance calculation constant terms are used according to different wavebands of the similar grid pixels; A pixel weight matrix technology module is used to calculate the weight matrix of each similar grid pixel according to the spectral distance, spatial distance and temporal distance between each similar grid pixel in the final similar grid pixel set and the target grid pixel; The surface reflectance fusion calculation module is used to calculate the surface reflectance of the target grid pixel based on the surface reflectance value of each similar grid pixel and the weight matrix of each similar grid pixel, complete the high spatial resolution fusion of the preprocessed low spatial resolution image, and obtain a high spatiotemporal resolution image.

8. A computer device comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the high spatiotemporal resolution image generation method based on the spatiotemporal fusion algorithm according to any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method for generating high spatiotemporal resolution images based on a spatiotemporal fusion algorithm according to any one of claims 1 to 6 is implemented.

10. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the method for generating high spatiotemporal resolution images based on a spatiotemporal fusion algorithm according to any one of claims 1 to 6 is implemented.

Images