A method for designing a spatiotemporal database of land value

Abstract

This invention discloses a method and system for designing a spatiotemporal database of land value, applied in the fields of spatial geographic information data processing and image correction technology. The method involves acquiring spectral remote sensing image data; performing geographic coordinate matching on the spectral remote sensing image data to generate coordinate spectral remote sensing image data; digitally rasterizing the spectral remote sensing image data using a digital elevation model to generate point cloud data; performing 3D reconstruction based on the point cloud data to obtain a 3D terrain model; performing boundary region reconstruction processing on the 3D terrain model to obtain a reconstructed model; extracting ground features from the reconstructed model to obtain ground feature data; and constructing a spatiotemporal database by spatiotemporally indexing the ground feature data. This method can correct the brightness of edge defects of interval-type and linear ground features based on predicted brightness changes, significantly reducing grayscale distortion in 3D images and eliminating blurring caused by interference.

Description

Technical Field

[0001] This invention relates to the fields of spatial geographic information processing technology and image correction technology, and specifically to a method for designing a spatiotemporal database of land value. Background Technology

[0002] To achieve accurate and timely acquisition and analysis of land resource value information, establishing a land resource mapping database is an effective approach. However, traditional methods of constructing mapping databases typically rely on field surveys and manual data collection, resulting in slow data updates and an inability to reflect changes in land resources in a timely manner. Furthermore, conventional databases often struggle to deeply explore the spatial relationships and development trends within geographic information when storing mapping data.

[0003] Therefore, for efficient storage and retrieval of temporal geographic information, a method and system for constructing a land resource mapping database, as provided in Chinese invention patent CN117611978B, can integrate image information from different time points and generate multi-temporal image data by performing temporal alignment and overlay processing on coordinate multispectral image data. By performing brightness correction on the multi-temporal images, brightness differences between images at different time points can be eliminated, ensuring data consistency. Three-dimensional modeling of the brightness-corrected image data provides three-dimensional information about the land surface, which can be used for terrain analysis, geomorphic feature extraction, and environmental change monitoring. By performing residual reconstruction on point cloud terrain models to repair potential data gaps, noise, or errors, a more accurate and complete terrain model can be generated. However, during the registration, temporal alignment, and overlay processing of the loaded multi-temporal spectral data, flares formed by strong light spots generated by the refraction and scattering of solar radiation in the atmosphere can cause unclear information in some areas of the satellite remote sensing image. This is because the location of the radiation from non-target pixels to target pixels or the interference phenomenon between various flares (where the light intensity weakens and dark fringes appear) causes the edges of ground features in the image (including the boundaries of ground features, terrain contours, geological linear structures, and ring structures, etc.) to become blurred. In particular, a large amount of data gaps, noise, or errors inevitably exist at the image boundaries and cannot be corrected. Summary of the Invention

[0004] The purpose of this invention is to propose a method and system for designing a spatiotemporal database of land value, so as to solve one or more technical problems existing in the prior art, and at least provide a beneficial option or create conditions.

[0005] To achieve the above objectives, according to one aspect of the present invention, a method for designing a spatiotemporal database of land value is provided, the method comprising the following steps:

[0006] S100, acquire spectral remote sensing image data;

[0007] S200, performs geographic coordinate matching on spectral remote sensing image data to generate coordinate spectral remote sensing image data;

[0008] S300 uses a digital elevation model to digitally rasterize spectral remote sensing image data to generate point cloud data, and then performs 3D reconstruction based on the point cloud data to obtain a 3D terrain model.

[0009] S400, performs boundary region reconstruction processing on a 3D terrain model to obtain a reconstructed model;

[0010] S500 extracts ground features from the reconstructed model to obtain ground feature data;

[0011] S600 constructs a spatiotemporal database by spatiotemporally indexing ground feature data.

[0012] Furthermore, in S100, the spectral remote sensing image data is either multispectral image data or hyperspectral image data of a preset region of interest (ROI).

[0013] Furthermore, in S200, the method for generating coordinate spectral remote sensing image data by matching geographic coordinates of spectral remote sensing image data is as follows: by using the open-source raster spatial data transformation library GDAL (Geospatial Data Abstraction Library), the conversion between pixel coordinates of spectral remote sensing image data and geographic coordinates is realized to generate coordinate spectral remote sensing image data, i.e., inverse geographic transformation.

[0014] Furthermore, in S300, the method for 3D reconstruction based on point cloud data is any one of the following: Poisson surface reconstruction algorithm, 3D point cloud surface reconstruction algorithm based on Delaunay growth method, greedy projection triangulation algorithm, or surface reconstruction algorithm based on B-spline curve.

[0015] In practical applications, the radiation from non-target pixels to target pixels leads to significant data loss, noise, or errors in the boundaries, terrain contours, linear geological structures, and ring structures of ground features in spectral remote sensing images. Furthermore, interference phenomena (where light intensity weakens and dark fringes appear) between flares occur at these locations. This results in the 3D models obtained through existing 3D terrain model reconstruction techniques inevitably becoming blurred at the boundaries of ground features or producing numerous holes (corresponding to missing viewpoints) at the edges of ground features (where interference phenomena are common). This significantly affects the accuracy of 3D reconstruction, leading to inaccurate reconstruction results and severely distorted 3D models. Therefore, this application employs the following method to reconstruct the boundaries of ground features in the 3D model to eliminate these problems:

[0016] Furthermore, in S400, the boundary region reconstruction process of the 3D terrain model to obtain the reconstructed model includes the following steps:

[0017] The 3D terrain model is converted to grayscale to obtain a grayscale model. This means that the grayscale value of each pixel in the grayscale model is the grayscale value of the corresponding pixel in the spectral remote sensing image and the grayscale model. The grayscale model is then segmented using a watershed algorithm to obtain reconstructed regions formed by watershed lines. The average grayscale value of all pixels in each reconstructed region is recorded as the grayscale mean of that region. The grayscale mean of the reconstructed region with the smallest grayscale mean is taken as the reference grayscale. The average grayscale value of all pixels on the boundary (i.e., the watershed line) of each reconstructed region is recorded as the boundary grayscale. Reconstructed regions with boundary grayscale values ​​less than the reference grayscale are selected and recorded as interference regions. The light interference intensity of each interference region is calculated, and interference regions with light interference intensity greater than the average light interference intensity of all interference regions are marked as affected regions. The affected regions on the 3D terrain model are then classified and reconstructed to obtain the reconstructed model.

[0018] The method for calculating the optical interference intensity of the interference interval involves: sequentially acquiring the average gray values ​​of all pixels on the boundary of the interference interval over the most recent N time resolutions, and marking each acquired average gray value as follows: if the average gray value acquired at the previous time resolution is less than the average gray value acquired at the next time resolution, then calculate the difference between the average gray value acquired at the next time resolution and the average gray value acquired at the previous time resolution, and mark all these differences as the brightening gray amplitude; if the average gray value acquired at the previous time resolution is greater than the average gray value acquired at the next time resolution, then calculate the difference between the average gray value acquired at the previous time resolution and the average gray value acquired at the next time resolution, and mark all these differences as the darkening gray amplitude; calculate the sum of all the darkening gray amplitudes as the darkening amplitude; calculate the sum of all the brightening gray amplitudes as the brightening amplitude; use the sum of the darkening amplitude and the brightening amplitude as the total amplitude; and use the ratio of the darkening amplitude to the total amplitude as the optical interference intensity.

[0019] The principle of light interference intensity is to infer the strength of interference phenomena or noise on the edges of ground objects in a spectral image by calculating the amplitude of gray-scale brightness changes within a preset time period. If the light interference intensity value is large, it indicates that there is a strong interference state, and the area near this boundary position will also be greatly affected in the future, and reconstruction and repair must be carried out; if the value is small, it indicates that the interference is weak, and the area near this boundary position will be affected in the future, and reconstruction and repair are not necessary.

[0020] Where N is the preset data, and the value of N is in the range of [20, 100].

[0021] Furthermore, the classification and reconstruction of the affected areas on the 3D terrain model to obtain the reconstructed model includes the following steps: Each affected area on the 3D terrain model is classified and reconstructed sequentially. Specifically, the interference region with the smallest absolute value of the difference between the mean gray value and the mean gray value of the current affected area among the interference regions adjacent to the boundary of the affected area is selected as the template region. (The template region is selected to choose a region whose brightness is closest to that of the affected area before the affected area, and whose brightness change is relatively stable based on the light interference intensity prediction, as a reference region for restoration. This ensures that there is almost no visual difference after reconstruction and restoration, and the distortion after reconstruction is small compared to reality.)

[0022] If the light interference intensity of the template area is less than the average light interference intensity of all interference intervals, the repair type of the affected interval is classified as enhanced repair; if the light interference intensity of the template area is greater than the average light interference intensity of all interference intervals, the repair type of the affected interval is classified as weakened repair.

[0023] If the repair type is enhanced repair, the average of all brightening grayscale amplitudes in the template area is taken as the template brightening mean. The grayscale values ​​of all pixels in the current affected area are increased by the template brightening mean to obtain the reconstructed model.

[0024] If the repair type is weakening repair, the average of all darkening grayscale amplitudes in the template area is taken as the template darkening mean. The grayscale values ​​of all pixels in the currently affected area are reduced by the template darkening mean to obtain the reconstructed model.

[0025] The brightening grayscale amplitude of the template area is as follows: obtain the average grayscale value of all pixels on the boundary of the template area in the most recent N time resolutions and mark each of the collected average grayscale values ​​as follows: if the average grayscale value collected in the previous time resolution is less than the average grayscale value collected in the next time resolution, calculate the difference between the average grayscale value collected in the next time resolution and the average grayscale value collected in the previous time resolution, and mark all these differences as the brightening grayscale amplitude;

[0026] The darkening grayscale amplitude of the template area is as follows: obtain the average grayscale value of all pixels on the boundary of the template area in the most recent N time resolutions and mark each of the collected average grayscale values ​​as follows: if the average grayscale value collected in the previous time resolution is greater than the average grayscale value collected in the next time resolution, calculate the difference between the average grayscale value collected in the previous time resolution and the average grayscale value collected in the next time resolution, and mark all these differences as the darkening grayscale amplitude.

[0027] The above methods can correct the brightness of edge defects of interval-type features with a certain area based on the predicted brightness changes, which can significantly reduce grayscale distortion of 3D images and eliminate remote sensing image blurring caused by strong interference. However, for some linear terrain contours and geological linear structures, edge blurring is still easy to occur. Therefore, to solve the above problems, this application reconstructs the boundary through the following method:

[0028] Preferably, the process of classifying and reconstructing the affected regions on the 3D terrain model to obtain the reconstructed model includes the following steps:

[0029] The interference region with the smallest absolute value of the difference between the mean gray value and the mean gray value of the current affected region among all interference regions adjacent to the boundary of the affected region is selected as the template region. (The template region is selected to select a region whose brightness is closest to that of the affected region before the affected region and whose brightness change is relatively stable according to the light interference intensity prediction, so as a reference region for restoration, so that there is almost no visual difference after reconstruction and restoration, and the distortion after reconstruction is small compared with the actual situation.)

[0030] The maximum gray value point on the boundary of the template region is designated as the maximum point; the minimum gray value point on the boundary of the template region is designated as the minimum point; the shortest curve segment on the boundary of the template region passing between the maximum and minimum points is designated as the reconstruction curve; the point with the smallest absolute difference in gray value between the affected interval and the maximum point is designated as the large projection point; the point with the smallest absolute difference in gray value between the affected interval and the minimum point is designated as the small projection point; the shortest curve segment on the boundary of the affected interval passing between the large and small projection points is deleted; the reconstruction curve is scaled until the length between the maximum and minimum points is equal to the length between the large and small projection points, and is designated as the scaling curve; the scaling curve is moved to coincide between the maximum and large projection points and between the minimum and small projection points, thus reconstructing the boundary of the affected interval; the reconstructed model is obtained.

[0031] Furthermore, in S500, the method for extracting ground features from the reconstructed model to obtain ground feature data is any one of the following methods: ENVI thresholding, deep learning algorithm, mathematical morphology-based analysis method, texture feature analysis method, statistical analysis method, and region growing method to obtain ground feature data such as buildings, vegetation, and water bodies.

[0032] Furthermore, in S600, the methods for constructing a spatiotemporal database by spatiotemporally indexing ground feature data include:

[0033] Three-dimensional occlusion detection is performed on the ground feature data and object feature segmentation is performed to obtain land data and terrain data respectively;

[0034] Temporal data is obtained by temporal processing of land and terrain data; spatiotemporal indexing is performed on the temporal data to construct a spatiotemporal database.

[0035] Temporal processing involves adding timestamps or time period attributes.

[0036] Land data can be divided into three types: spatial characteristic data (location data), temporal attribute data (scale data), and thematic attribute data (non-location data).

[0037] The terrain data includes terrain features such as elevation models and contour lines.

[0038] Furthermore, in S600, the method for constructing a spatiotemporal database by performing spatiotemporal indexing on temporal data specifically includes:

[0039] Map the attribute values ​​of land data and terrain data to standardized codes and unify the units to form geographic meta-attribute data;

[0040] The temporal and spatial index is designed based on the quadtree index principle, forming a temporal and spatial database by combining the temporal data with various geographic meta-attribute data.

[0041] Preferably, the spatiotemporal database is deployed by uploading the constructed spatiotemporal database to the server through data migration tools, database backup and restoration, etc., and the spatiotemporal database is used for land value assessment.

[0042] Preferably, the three-dimensional occlusion detection is implemented by the method in the following literature: Wu Di, Zhang Xudong, Zhang Jun, et al. Depth acquisition of light field image guided by 3D occlusion model [J]. Journal of Image and Graphics, 2021, 26(4):15.

[0043] This invention also provides a land value spatiotemporal database design system, the system 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 in a unit of the following system:

[0044] Remote sensing acquisition unit, used to acquire spectral remote sensing image data;

[0045] The coordinate matching unit is used to perform geographic coordinate matching on spectral remote sensing image data to generate coordinate spectral remote sensing image data.

[0046] The 3D reconstruction unit is used to digitally rasterize spectral remote sensing image data using a digital elevation model to generate point cloud data, and then perform 3D reconstruction based on the point cloud data to obtain a 3D terrain model.

[0047] The region reconstruction unit is used to reconstruct the boundary region of a 3D terrain model to obtain a reconstructed model.

[0048] The feature extraction unit is used to extract features from the reconstructed model to obtain feature data.

[0049] The database construction unit is used to construct a spatiotemporal database by spatiotemporally indexing ground feature data.

[0050] The beneficial effects of the present invention are as follows: The present invention provides a land value spatiotemporal database design method and system, which can correct the brightness of edge defects of interval type and linear land features according to the predicted brightness changes, which can significantly reduce the grayscale distortion of three-dimensional images and eliminate the blur caused by interference. Attached Figure Description

[0051] The above and other features of the present invention will become more apparent from the detailed description of the embodiments shown in conjunction with the accompanying drawings. In the accompanying drawings, the same reference numerals denote the same or similar elements. Obviously, the drawings described below are merely some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without any creative effort. In the drawings:

[0052] Figure 1 The diagram shows a flowchart of a land value spatiotemporal database design method.

[0053] Figure 2 The image shown is a 3D model image where dark fringes appear at the boundary due to interference caused by the reconstructed light intensity.

[0054] Figure 3 The image shown is a 3D model image with numerous holes and missing viewpoint images after reconstruction.

[0055] Figure 4 The image shown is a 3D model image of Example 1 with dark stripes removed;

[0056] Figure 5 The image shown is a 3D model image with the missing viewpoint image removed in Example 2;

[0057] Figure 6 The diagram shown is a system architecture diagram for a spatiotemporal database design of land value. Detailed Implementation

[0058] The following will provide a clear and complete description of the concept, specific structure, and technical effects of the present invention in conjunction with the embodiments and accompanying drawings, so as to fully understand the purpose, solution, and effects of the present invention. It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other.

[0059] Example 1

[0060] like Figure 1 The diagram shown is a flowchart of a land value spatiotemporal database design method according to the present invention. The following is a combination of... Figure 1 This paper describes a method for designing a spatiotemporal database of land value according to an embodiment of the present invention.

[0061] This invention proposes a method for designing a spatiotemporal database of land value, which specifically includes the following steps:

[0062] S100, acquire spectral remote sensing image data;

[0063] S200, performs geographic coordinate matching on spectral remote sensing image data to generate coordinate spectral remote sensing image data;

[0064] S300 uses a digital elevation model to digitally rasterize spectral remote sensing image data to generate point cloud data, and then performs 3D reconstruction based on the point cloud data to obtain a 3D terrain model.

[0065] S400, performs boundary region reconstruction processing on a 3D terrain model to obtain a reconstructed model;

[0066] S500 extracts ground features from the reconstructed model to obtain ground feature data;

[0067] S600 constructs a spatiotemporal database by spatiotemporally indexing ground feature data.

[0068] Furthermore, in S100, the spectral remote sensing image data is the spectral remote sensing image in the Landsat-8-OLI_TIRS raw remote sensing image dataset with a preset region of interest (ROI).

[0069] Furthermore, in S200, the method for generating coordinate spectral remote sensing image data by matching geographic coordinates of spectral remote sensing image data is as follows: by using the open-source raster spatial data transformation library GDAL (Geospatial Data Abstraction Library), the conversion between pixel coordinates of spectral remote sensing image data and geographic coordinates is realized to generate coordinate spectral remote sensing image data, i.e., inverse geographic transformation.

[0070] Furthermore, in S300, the method for 3D reconstruction based on point cloud data is the Poisson surface reconstruction algorithm.

[0071] In practical applications, such as Figure 2As shown, because the radiation from non-target pixels to target pixels causes a large amount of missing, noisy, or erroneous data at locations such as the boundaries of ground features, terrain contours, linear geological structures, and ring structures in spectral remote sensing images, as well as locations where interference phenomena occur between various flares (where light intensity weakens and dark fringes appear), the 3D models obtained through existing 3D terrain model reconstruction techniques inevitably become blurred at the boundaries of ground features or at the edges of ground features (where interference phenomena are common). Figure 3 As shown, this results in numerous holes (corresponding missing viewpoint images), which significantly affects the accuracy of 3D reconstruction, leading to inaccurate reconstruction results and severe distortion of the reconstructed 3D model. Therefore, this application employs the following method to reconstruct the feature boundaries of the 3D model to eliminate the aforementioned problems:

[0072] Further, in S400, the boundary region reconstruction processing of the 3D terrain model to obtain the reconstructed model includes the following steps: The 3D terrain model is converted to grayscale to obtain a grayscale model, that is, the grayscale value of each pixel in the grayscale model is the grayscale value of the pixel at the corresponding position in the spectral remote sensing image and the grayscale model. The grayscale model is segmented using a watershed algorithm to obtain each reconstructed interval formed by watershed lines. The average grayscale value of all pixels in the reconstructed interval is recorded as the grayscale mean of the reconstructed interval. The grayscale mean of the reconstructed interval with the smallest grayscale mean is recorded as the reference grayscale. The grayscale mean of all pixels on the boundary of the reconstructed interval is recorded as the boundary grayscale. The reconstructed intervals with boundary grayscale values ​​less than the reference grayscale are selected and recorded as interference intervals. The light interference intensity of the interference intervals is calculated, and the interference intervals with light interference intensity greater than the average light interference intensity of all interference intervals are marked as affected intervals. The affected intervals on the 3D terrain model are classified and reconstructed to obtain the reconstructed model.

[0073] The method for calculating the optical interference intensity of the interference interval involves: sequentially acquiring the average gray values ​​of all pixels on the boundary of the interference interval over the most recent N time resolutions, and marking each acquired average gray value as follows: if the average gray value acquired at the previous time resolution is less than the average gray value acquired at the next time resolution, then calculate the difference between the average gray value acquired at the next time resolution and the average gray value acquired at the previous time resolution, and mark all these differences as the brightening gray amplitude; if the average gray value acquired at the previous time resolution is greater than the average gray value acquired at the next time resolution, then calculate the difference between the average gray value acquired at the previous time resolution and the average gray value acquired at the next time resolution, and mark all these differences as the darkening gray amplitude; calculate the sum of all the darkening gray amplitudes as the darkening amplitude; calculate the sum of all the brightening gray amplitudes as the brightening amplitude; use the sum of the darkening amplitude and the brightening amplitude as the total amplitude; and use the ratio of the darkening amplitude to the total amplitude as the optical interference intensity.

[0074] Furthermore, the classification and reconstruction of the affected regions on the 3D terrain model to obtain the reconstructed model includes the following steps: classifying and reconstructing each affected region on the 3D terrain model in sequence, specifically: selecting the interference region with the smallest absolute value of the difference between the mean gray value and the mean gray value of the current affected region among the interference regions adjacent to the boundary of the affected region as the template region;

[0075] If the light interference intensity of the template area is less than the average light interference intensity of all interference intervals, the repair type of the affected interval is classified as enhanced repair; if the light interference intensity of the template area is greater than the average light interference intensity of all interference intervals, the repair type of the affected interval is classified as weakened repair.

[0076] If the repair type is enhanced repair, the average of all brightening grayscale amplitudes in the template area is taken as the template brightening mean. The grayscale values ​​of all pixels in the current affected area are increased by the template brightening mean to obtain the reconstructed model.

[0077] If the repair type is weakening repair, the average of all darkening grayscale amplitudes in the template area is taken as the template darkening mean. The grayscale values ​​of all pixels in the currently affected area are reduced by the template darkening mean to obtain the reconstructed model.

[0078] like Figure 4 The image shown is a 3D model image of Example 1 with dark stripes removed.

[0079] The brightening grayscale amplitude of the template area is as follows: obtain the average grayscale value of all pixels on the boundary of the template area in the most recent N time resolutions and mark each of the collected average grayscale values ​​as follows: if the average grayscale value collected in the previous time resolution is less than the average grayscale value collected in the next time resolution, calculate the difference between the average grayscale value collected in the next time resolution and the average grayscale value collected in the previous time resolution, and mark all these differences as the brightening grayscale amplitude;

[0080] The darkening grayscale amplitude of the template area is as follows: obtain the average grayscale value of all pixels on the boundary of the template area in the most recent N time resolutions and mark each of the collected average grayscale values ​​as follows: if the average grayscale value collected in the previous time resolution is greater than the average grayscale value collected in the next time resolution, calculate the difference between the average grayscale value collected in the previous time resolution and the average grayscale value collected in the next time resolution, and mark all these differences as the darkening grayscale amplitude.

[0081] Where N takes the value 20.

[0082] Furthermore, in S500, the method for extracting land features from the reconstructed model to obtain land feature data is: using the ENVI threshold method to obtain data on buildings, vegetation, and water bodies.

[0083] Furthermore, in S600, the methods for constructing a spatiotemporal database by spatiotemporally indexing ground feature data include:

[0084] Three-dimensional occlusion detection is performed on the ground feature data and object feature segmentation is performed to obtain land data and terrain data respectively;

[0085] Temporal data is obtained by temporal processing of land and terrain data; spatiotemporal indexing is performed on the temporal data to construct a spatiotemporal database.

[0086] Temporal processing involves adding a timestamp attribute.

[0087] Land data can be divided into three types: spatial characteristic data (location data), temporal attribute data (scale data), and thematic attribute data (non-location data).

[0088] The terrain data includes elevation models and contour lines.

[0089] Furthermore, in S600, the method for constructing a spatiotemporal database by performing spatiotemporal indexing on temporal data specifically includes: mapping the attribute values ​​of land data and terrain data to standardized codes and unifying the units to become geographic meta-attribute data; designing spatiotemporal indexes on temporal data using the quadtree index principle, and forming a spatiotemporal database by combining temporal data with various geographic meta-attribute data.

[0090] Preferably, the spatiotemporal database is deployed by uploading the constructed spatiotemporal database to the server using the MySQLWorkbench data migration tool. This spatiotemporal database is used for land value assessment.

[0091] Preferably, the three-dimensional occlusion detection is implemented by the method in the following literature: Wu Di, Zhang Xudong, Zhang Jun, et al. Depth acquisition of light field image guided by 3D occlusion model [J]. Journal of Image and Graphics, 2021, 26(4):15.

[0092] Example 2

[0093] This embodiment 2 replaces the method of classifying and reconstructing the affected areas on the 3D terrain model to obtain the reconstructed model, which is based on embodiment 1. Specifically:

[0094] Preferably, the process of classifying and reconstructing the affected regions on the 3D terrain model to obtain the reconstructed model includes the following steps:

[0095] Select the interference region with the smallest absolute value of the difference between the mean gray value and the mean gray value of the current affected region among all interference regions adjacent to the boundary of the affected region as the template region;

[0096] The maximum gray value point on the boundary of the template region is designated as the maximum point; the minimum gray value point on the boundary of the template region is designated as the minimum point; the shortest curve segment on the boundary of the template region passing between the maximum and minimum points is designated as the reconstruction curve; the point with the smallest absolute difference in gray value between the affected interval and the maximum point is designated as the large projection point; the point with the smallest absolute difference in gray value between the affected interval and the minimum point is designated as the small projection point; the shortest curve segment on the boundary of the affected interval passing between the large and small projection points is deleted; the reconstruction curve is scaled until the length between the maximum and minimum points is equal to the length between the large and small projection points, and is designated as the scaling curve; the scaling curve is moved to coincide between the maximum and large projection points and between the minimum and small projection points, thus reconstructing the boundary of the affected interval; the reconstructed model is obtained.

[0097] like Figure 5 The image shown is a 3D model image with the missing viewpoint image removed in Example 2.

[0098] An embodiment of the present invention provides a land value spatiotemporal database design system, such as... Figure 6 The diagram shows a structural diagram of a land value spatiotemporal database design system according to the present invention. This embodiment of the land value spatiotemporal database design system includes: a processor, a memory, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the steps in the above embodiment of the land value spatiotemporal database design system.

[0099] The system includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program in units of the following system:

[0100] Remote sensing acquisition unit, used to acquire spectral remote sensing image data;

[0101] The coordinate matching unit is used to perform geographic coordinate matching on spectral remote sensing image data to generate coordinate spectral remote sensing image data.

[0102] The 3D reconstruction unit is used to digitally rasterize spectral remote sensing image data using a digital elevation model to generate point cloud data, and then perform 3D reconstruction based on the point cloud data to obtain a 3D terrain model.

[0103] The region reconstruction unit is used to reconstruct the boundary region of a 3D terrain model to obtain a reconstructed model.

[0104] The feature extraction unit is used to extract features from the reconstructed model to obtain feature data.

[0105] The database construction unit is used to construct a spatiotemporal database by spatiotemporally indexing ground feature data.

[0106] The aforementioned land value spatiotemporal database design system can run on computing devices such as desktop computers, laptops, handheld computers, and cloud servers. The system that can run on this land value spatiotemporal database design system may include, but is not limited to, processors and memory. Those skilled in the art will understand that the example described is merely an illustration of a land value spatiotemporal database design system and does not constitute a limitation on such a system. It may include more or fewer components, or a combination of certain components, or different components. For example, the land value spatiotemporal database design system may also include input / output devices, network access devices, buses, etc.

[0107] The processor referred to can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor. This processor is the control center of the operating system of the land value spatiotemporal database design system, connecting various parts of the operating system through various interfaces and lines.

[0108] The memory can be used to store the computer programs and / or modules. The processor implements various functions of the land value spatiotemporal database design system by running or executing the computer programs and / or modules stored in the memory and calling the data stored in the memory. The memory may mainly include a program storage area and a data storage area. The program storage area may store the operating system, at least one application program required for a function (such as sound playback function, image playback function, etc.), etc.; the data storage area may store data created based on the use of the mobile phone (such as audio data, phonebook, etc.). In addition, the memory may include high-speed random access memory, and may also include non-volatile memory, such as hard disk, memory, plug-in hard disk, smart media card (SMC), secure digital card (SD card), flash card, at least one disk storage device, flash memory device, or other volatile solid-state storage device.

[0109] Although the invention has been described in considerable detail and particularly with regard to several of the described embodiments, it is not intended to limit itself to any of these details or embodiments or any particular embodiment, thereby effectively covering the intended scope of the invention. Furthermore, the invention has been described above with respect to embodiments foreseeable by the inventors in order to provide a useful description, and non-substantial modifications to the invention that have not yet been foreseen may still represent equivalent modifications.

Claims

1. A method for designing a spatiotemporal database of land value, characterized in that, The method includes the following steps: S100, acquire spectral remote sensing image data; S200, performs geographic coordinate matching on spectral remote sensing image data to generate coordinate spectral remote sensing image data; S300 uses a digital elevation model to digitally rasterize spectral remote sensing image data to generate point cloud data, and then performs 3D reconstruction based on the point cloud data to obtain a 3D terrain model. S400, performs boundary region reconstruction processing on a 3D terrain model to obtain a reconstructed model; S500 extracts ground features from the reconstructed model to obtain ground feature data; S600 constructs a spatiotemporal database by performing spatiotemporal indexing on ground feature data. The process involves converting a 3D terrain model to grayscale to obtain a grayscale model. This means setting the grayscale value of each pixel in the grayscale model to the grayscale value of the corresponding pixel in the spectral remote sensing image. The grayscale model is then segmented using a watershed algorithm to obtain reconstructed regions formed by watershed lines. The average grayscale value of all pixels within a reconstructed region is recorded as the grayscale mean of that region. The grayscale mean of the region with the smallest grayscale mean is set as the reference grayscale. The average grayscale value of all pixels on the boundary of a reconstructed region is set as the boundary grayscale. Regions with boundary grayscale values ​​less than the reference grayscale are selected as interference regions. The light interference intensity of these interference regions is calculated, and regions with light interference intensity greater than the average light interference intensity of all interference regions are marked as affected regions. Finally, the affected regions on the 3D terrain model are classified and reconstructed to obtain the reconstructed model.

2. The method for designing a spatiotemporal database of land value according to claim 1, characterized in that, In S400, the method for calculating the light interference intensity of the interference interval is as follows: The average gray values ​​of all pixels on the boundary of the interference interval at the most recent N time resolutions are sequentially acquired, and each acquired average gray value is marked as follows: If the average gray value acquired at the previous time resolution is less than the average gray value acquired at the next time resolution, the difference between the average gray value acquired at the next time resolution and the average gray value acquired at the previous time resolution is calculated, and all these differences are marked as the brightening gray amplitude; If the average gray value acquired at the previous time resolution is greater than the average gray value acquired at the next time resolution, the difference between the average gray value acquired at the previous time resolution and the average gray value acquired at the next time resolution is calculated, and all these differences are marked as the darkening gray amplitude; The sum of all darkening gray amplitudes is calculated as the darkening amplitude; The sum of all brightening gray amplitudes is calculated as the brightening amplitude; The sum of the darkening amplitude and the brightening amplitude is taken as the total amplitude; The ratio of the darkening amplitude to the total amplitude is taken as the light interference intensity, where N ranges from [20, 100].

3. The method for designing a spatiotemporal database of land value according to claim 1, characterized in that, In S400, the process of classifying and reconstructing the affected regions on the 3D terrain model to obtain the reconstructed model includes the following steps: The affected regions on the 3D terrain model are classified and reconstructed sequentially. Specifically, the interference region with the smallest absolute value of the difference between the mean gray value and the mean gray value of the current affected region among the interference regions adjacent to the boundary of the affected region is selected as the template region.

4. The method for designing a spatiotemporal database of land value according to claim 3, characterized in that, In S400, if the light interference intensity of the template area is less than the average light interference intensity of all interference intervals, the repair type of the affected interval is classified as enhanced repair; if the light interference intensity of the template area is greater than the average light interference intensity of all interference intervals, the repair type of the affected interval is classified as weakened repair. If the repair type is enhanced repair, the average of all brightening grayscale amplitudes in the template area is taken as the template brightening mean. The grayscale values ​​of all pixels in the current affected area are increased by the template brightening mean to obtain the reconstructed model. If the repair type is weakening repair, the average of all darkening grayscale amplitudes in the template area is taken as the template darkening mean. The grayscale values ​​of all pixels in the currently affected area are reduced by the template darkening mean to obtain the reconstructed model.

5. The method for designing a spatiotemporal database of land value according to claim 4, characterized in that, In S400, the brightening grayscale amplitude of the template area is as follows: obtain the average grayscale value of all pixels on the boundary of the template area in the most recent N time resolutions and mark each of the collected average grayscale values ​​as follows: if the average grayscale value collected in the previous time resolution is less than the average grayscale value collected in the next time resolution, calculate the difference between the average grayscale value collected in the next time resolution and the average grayscale value collected in the previous time resolution, and mark all these differences as the brightening grayscale amplitude.

6. The method for designing a spatiotemporal database of land value according to claim 3, characterized in that, In S400, the method of classifying and reconstructing affected areas on a 3D terrain model to obtain a reconstructed model is replaced by the following steps: The template region is selected from the interference regions adjacent to the boundary of the affected region, with the smallest absolute value of the difference between the mean gray value and the mean gray value of the current affected region. The point with the largest gray value on the boundary of the template region is designated as the maximum point; the point with the smallest gray value on the boundary of the template region is designated as the minimum point; the curve segment with the shortest length between the maximum and minimum points on the boundary of the template region is designated as the reconstruction curve; the point with the smallest absolute value of the difference between the gray values ​​of the affected region and the maximum point is designated as the large projection point; the point with the smallest absolute value of the difference between the gray values ​​of the affected region and the minimum point is designated as the small projection point; the curve segment with the shortest length between the large and small projection points on the boundary of the affected region is deleted; the reconstruction curve is scaled until the length between the maximum and minimum points is equal to the length between the large and small projection points, and is designated as the scaling curve; the scaling curve is moved to coincide between the maximum and large projection points and between the minimum and small projection points, thereby reconstructing the boundary of the affected region; the reconstruction model is obtained.

7. A spatiotemporal database design system for land value, characterized in that, The land value spatiotemporal database design system includes: a processor, a memory, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the steps of any one of the land value spatiotemporal database design methods according to claims 1-6.

Images