Geographic information system GIS modeling method and device, electronic equipment and medium

Abstract

The embodiment of the invention provides a geographic information system (GIS) modeling method and device, electronic equipment and a medium, and the method comprises the steps: collecting video image data of a target area, recording the latitude and longitude coordinates of a collection position, carrying out the extraction of a key frame, carrying out the denoising processing, extracting the image feature points of the key frame, and carrying out the feature matching, determining a relative position between the key frames; the method comprises the following steps of: calculating a space coordinate of each image feature point by utilizing matched image feature points and GPS data of acquisition equipment, generating point cloud data, performing 3D Gaussian Splitting operation on each point in the point cloud data by applying a Gaussian weighting function to obtain an integral Splitting operation result, and reconstructing a three-dimensional model through the integral Splitting operation result to obtain a three-dimensional model. Smooth processing is carried out to optimize the surface quality of the three-dimensional model; and importing the reconstructed three-dimensional model into a GIS (Geographic Information System), managing model data by using a spatial database, and placing the three-dimensional model at a corresponding position in the GIS by using latitude and longitude coordinates.

Description

Technical Field

[0001] This document relates to the field of geographic information system technology, and in particular to a geographic information system GIS modeling method, device, electronic equipment and medium. Background Art

[0002] In the traditional GIS modeling process, building a 3D model usually requires complex data collection work first, especially relying on oblique photography technology to obtain multi-angle images of objects. This process not only requires professional equipment and technical support, but also involves a lot of data processing work, such as image matching and 3D reconstruction. Subsequently, based on these processed data, GIS professionals also need to manually build and adjust the model to ensure the accuracy and realism of the model. This process is not only time-consuming and labor-intensive, but also prone to human errors, limiting the efficiency and popularity of GIS modeling.

[0003] Although oblique photography has improved the efficiency of data collection to a certain extent, the subsequent modeling process is still cumbersome and complicated. In addition, with the continuous increase in the amount of geospatial data, traditional modeling methods are unable to cope with large-scale data sets and are unable to meet the needs of real-time and efficient 3D modeling. Summary of the invention

[0004] The purpose of the present invention is to provide a geographic information system (GIS) modeling method, device, electronic equipment and medium, aiming to solve the above-mentioned problems in the prior art.

[0005] The present invention provides a GIS modeling method, comprising:

[0006] The video image data of the target area is collected by the collection device and the longitude and latitude coordinates of the collection position are recorded, key frames are extracted from the video image data, and the extracted key frames are subjected to denoising, image feature points of the key frames are extracted, and feature matching is performed to determine the relative positions between the key frames;

[0007] Using the matched image feature points and the GPS data of the acquisition device, the spatial coordinates of each image feature point are calculated by triangulation to generate point cloud data, and a Gaussian weight function is applied to each point in the point cloud data to perform a 3D Gaussian Splatting operation to obtain an integral Splatting operation result. Based on the integral Splatting operation result, a voxelization method is used to reconstruct a three-dimensional model, and smoothing is performed to optimize the surface quality of the three-dimensional model;

[0008] The reconstructed three-dimensional model is imported into the GIS system, the model data is managed using a spatial database, and the three-dimensional model is placed at a corresponding position in the GIS system using the latitude and longitude coordinates.

[0009] The present invention provides a GIS modeling device, comprising:

[0010] The acquisition, extraction and matching module is used to acquire video image data of the target area through an acquisition device and record the longitude and latitude coordinates of the acquisition position, extract key frames from the video image data, perform denoising on the extracted key frames, extract image feature points of the key frames, perform feature matching, and determine the relative positions between the key frames;

[0011] A point cloud modeling module is used to calculate the spatial coordinates of each image feature point by triangulation using the matched image feature points and the GPS data of the acquisition device, generate point cloud data, apply a Gaussian weight function to each point in the point cloud data to perform a 3D Gaussian Splatting operation to obtain an integral Splatting operation result, reconstruct a three-dimensional model using a voxelization method based on the integral Splatting operation result, and perform smoothing to optimize the surface quality of the three-dimensional model;

[0012] The import module is used to import the reconstructed three-dimensional model into the GIS system, use the spatial database to manage the model data, and use the latitude and longitude coordinates to place the three-dimensional model at a corresponding position in the GIS system.

[0013] An embodiment of the present invention further provides an electronic device, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program implements the steps of the above-mentioned GIS modeling method when executed by the processor.

[0014] An embodiment of the present invention further provides a computer-readable storage medium, on which a program for implementing information transmission is stored, and when the program is executed by a processor, the steps of the above-mentioned GIS modeling method are implemented.

[0015] By adopting the embodiment of the present invention, the 3D Gaussian Splatting technology is introduced to change the traditional process of GIS modeling. After a simple aerial video or photo collection is performed by a drone, the 3D Gaussian Splatting algorithm can be directly used for three-dimensional modeling and the longitude and latitude coordinates at the time of shooting can be automatically imported into the GIP system, which greatly simplifies the modeling process, eliminates tedious data preprocessing and manual modeling steps, and realizes automation and rapidity from data collection to model generation. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate one or more embodiments of this specification or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this specification. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0017] Figure 1 is a flow chart of a GIS modeling method according to an embodiment of the present invention;

[0018] Figure 2 is a schematic diagram of a detailed process of the GIS modeling method according to an embodiment of the present invention;

[0019] Figure 3 is a technical flow chart of 3D Gaussian Splatting according to an embodiment of the present invention;

[0020] Figure 4 is a schematic diagram of a GIS modeling device according to an embodiment of the present invention;

[0021] Figure 5 is a schematic diagram of an electronic device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0022] In order to enable those skilled in the art to better understand the technical solutions in one or more embodiments of this specification, the following will be combined with the drawings in one or more embodiments of this specification to clearly and completely describe the technical solutions in one or more embodiments of this specification. Obviously, the described embodiments are only part of the embodiments of this specification, not all of the embodiments. Based on one or more embodiments of this specification, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of this document.

[0023] Method Embodiment

[0024] According to an embodiment of the present invention, a GIS modeling method is provided. Figure 1 is a flow chart of the GIS modeling method of an embodiment of the present invention, such as Figure 1 As shown, the GIS modeling method according to the embodiment of the present invention specifically includes:

[0025] Step S101, collecting video image data of the target area through a collection device and recording the longitude and latitude coordinates of the collection position, extracting key frames from the video image data, performing denoising on the extracted key frames, extracting image feature points of the key frames, performing feature matching, and determining the relative positions between the key frames; specifically includes:

[0026] The key frames of the video image data are extracted by using a time interval frame extraction algorithm or a content-based frame extraction algorithm, wherein the time interval frame extraction algorithm is expressed as Formula 1:

[0027] Frame n =Video[(n×Interval)mod Total Frames] Formula 1;

[0028] Among them, Frame n Indicates the video frame to be obtained, Video indicates the entire video file or video stream, n indicates the serial number of the frame to be selected, Interval indicates the selected interval or step size, mod indicates the remainder, and TotalFrames indicates the total number of frames in the video.

[0029] According to Formula 2, the extracted key frame is denoised using an image denoising algorithm through a Gaussian filter:

[0030]

[0031] Among them, I is the original image, I filtered is the filtered image, Gσ is the Gaussian kernel with standard deviation σ, h is the size of the filter, x, y represent the coordinates of the pixel in the original image, and i, j represent the relative coordinates within the Gaussian kernel.

[0032] The SIFT or SURF algorithm is used to extract the image feature points of the key frames, and feature matching is performed to determine the relative positions between the key frames.

[0033] Step S102, using the matched image feature points and the GPS data of the acquisition device, calculating the spatial coordinates of each image feature point by triangulation, generating point cloud data, applying a Gaussian weight function to each point in the point cloud data to perform a 3D Gaussian Splatting operation to obtain an integral Splatting operation result, reconstructing a three-dimensional model by using a voxelization method based on the integral Splatting operation result, and performing smoothing to optimize the surface quality of the three-dimensional model; specifically comprising:

[0034] According to Formula 3, each point P in the point cloud data i Apply the Gaussian weight function to perform a 3DGaussianSplatting operation and get the result of the integral Splatting operation:

[0035]

[0036] Where S(x) is the surface contribution at position x, C is the covariance matrix, and D is the dimension of the point.

[0037] Step S103, importing the reconstructed three-dimensional model into the GIS system, using a spatial database to manage the model data, and using the latitude and longitude coordinates to place the three-dimensional model at a corresponding position in the GIS system.

[0038] The above technical solution of the embodiment of the present invention is described in detail below in conjunction with the accompanying drawings. Figure 2 As shown, the technical solution of the embodiment of the present invention specifically includes the following processing:

[0039] Data collection work

[0040] 1.1 UAV Configuration

[0041] (1) High-resolution camera: Choose a camera with high resolution (at least 4K) to ensure image clarity.

[0042] (2) GPS module: Ensure that the drone is equipped with a high-precision GPS module to record the exact location when collecting data.

[0043] 1.2 Aerial photography planning

[0044] (1) Flight route design: Design the flight route of the drone based on the terrain and topographic features of the target area to ensure comprehensive coverage.

[0045] (2) Flight parameter setting: Set the UAV’s flight altitude, speed, and camera shooting angle to adapt to different environmental conditions.

[0046] 1.3 Data Collection

[0047] (1) Video image acquisition: The UAV flies along a predetermined route, and the camera captures video image data of the target area in real time.

[0048] (2) Longitude and latitude recording: The GPS module of the drone synchronously records the longitude and latitude coordinates of the acquisition location of each image frame.

[0049] Video data processing

[0050] 2.1 Frame extraction algorithm

[0051] (1) Fixed time interval frame extraction: Set a fixed time interval (for example, extract a frame every 1 second) to reduce the amount of data and retain key information.

[0052] (2) Content-based frame extraction: For areas with large dynamic changes, a content-based frame extraction algorithm is used to ensure that key changes are captured.

[0053] 2.2 Image Denoising

[0054] (1) Gaussian filter parameter selection: According to the noise level of the image, select the appropriate Gaussian kernel standard deviation σ and filter size h.

[0055] Feature point extraction and matching

[0056] 3.1 SIFT or SURF Algorithm

[0057] (1) Feature point extraction: Apply SIFT or SURF algorithm to each frame of image to extract key feature points.

[0058] (2) Feature descriptor generation: Generate a descriptor for each feature point for subsequent feature matching.

[0059] 3.2 Feature Matching

[0060] (1) Feature point matching: Use feature descriptors to match feature points of consecutive frames or adjacent frames to determine the relative positions between images.

[0061] Calculate spatial coordinates using triangulation

[0062] 4.1 Combining feature points with GPS data

[0063] (1) Triangulation: Using the matched feature points and the corresponding GPS data, the spatial coordinates of each feature point are calculated by triangulation.

[0064] Gaussian weight function Splatting operation

[0065] 5.1 Application of Gaussian weight function

[0066] (1) Splatting operation: Apply a Gaussian weight function to each feature point and map the point cloud data into three-dimensional space based on its spatial coordinates and weight value.

[0067] Surface reconstruction and optimization, Figure 3 3D Gaussian Splatting is a technical flow chart of an embodiment of the present invention. Figure 3 As shown, the following processing is included:

[0068] 6.1 Voxelization Method

[0069] (1) 3D surface reconstruction: The 3D surface is reconstructed using the voxelization method by integrating the results of the splatting operation.

[0070] 6.2 Smoothing

[0071] (1) Surface optimization: Smoothing the reconstructed 3D surface to improve the accuracy and visual effect of the model.

[0072] GIS system integration

[0073] 7.1 3D model import

[0074] (1) GIS system connection: Import the reconstructed 3D model into the GIS system and use the spatial database to manage the model data.

[0075] 7.2 Importing longitude and latitude coordinates

[0076] (1) Coordinate matching: Since the UAV has recorded the latitude and longitude coordinates of each frame of image when collecting data, these coordinates can be directly imported into the GIS system and associated with the corresponding 3D model data.

[0077] (2) Spatial positioning: In the GIS system, the three-dimensional model is accurately placed at the corresponding position in the geographic space based on the imported longitude and latitude coordinates.

[0078] In summary, the embodiment of the present invention simplifies the GIS modeling process and improves the modeling efficiency and accuracy by introducing the 3D Gaussian Splatting technology, while reducing the technical threshold and labor cost. It has wide applicability and brings significant technological progress and practical application value to the field of GIS modeling.

[0079] Compared with traditional methods, the GIS modeling method of the present invention has the following significant advantages:

[0080] Simplified process: No need for complicated oblique photography data processing and manual modeling process, which reduces the technical threshold and labor cost of modeling.

[0081] Efficiency: The 3D Gaussian Splatting algorithm can efficiently process drone aerial photography data, quickly generate three-dimensional models, and improve modeling efficiency.

[0082] Accuracy: Automatically processing data through algorithms reduces human intervention and improves the accuracy and consistency of the model.

[0083] Wide applicability: It is suitable for modeling needs in various geographical environments and provides strong technical support for smart cities, environmental monitoring, geological exploration and other fields.

[0084] Device Example 1

[0085] According to an embodiment of the present invention, a GIS modeling device is provided. Figure 4 Schematic diagram of a GIS modeling device according to an embodiment of the present invention. Figure 4 As shown, the GIS modeling device according to the embodiment of the present invention specifically includes:

[0086] The acquisition, extraction and matching module 40 is used to acquire video image data of the target area through an acquisition device and record the longitude and latitude coordinates of the acquisition position, extract key frames from the video image data, perform denoising on the extracted key frames, extract image feature points of the key frames, perform feature matching, and determine the relative positions between the key frames; specifically, it is used to:

[0087] The key frames of the video image data are extracted by using a time interval frame extraction algorithm or a content-based frame extraction algorithm, wherein the time interval frame extraction algorithm is expressed as Formula 1:

[0088] Frame n =Video[(n×Interval)mod Total Frames] Formula 1;

[0089] Among them, Frame n Indicates the video frame to be obtained, Video indicates the entire video file or video stream, n indicates the serial number of the frame to be selected, Interval indicates the selected interval or step size, mod indicates the remainder, and TotalFrames indicates the total number of frames in the video.

[0090] According to Formula 2, the extracted key frame is denoised using an image denoising algorithm through a Gaussian filter:

[0091]

[0092] Among them, I is the original image, I filtered is the filtered image, Gσ is the Gaussian kernel with standard deviation σ, h is the size of the filter, x, y represent the coordinates of the pixel in the original image, and i, j represent the relative coordinates within the Gaussian kernel.

[0093] The SIFT or SURF algorithm is used to extract the image feature points of the key frames, and feature matching is performed to determine the relative positions between the key frames.

[0094] The point cloud modeling module 42 is used to calculate the spatial coordinates of each image feature point by triangulation using the matched image feature points and the GPS data of the acquisition device, generate point cloud data, apply a Gaussian weight function to each point in the point cloud data to perform a 3D Gaussian Splatting operation to obtain an integral Splatting operation result, reconstruct a three-dimensional model using a voxelization method based on the integral Splatting operation result, and perform smoothing to optimize the surface quality of the three-dimensional model; specifically, it is used to:

[0095] According to Formula 3, each point P in the point cloud data iApply the Gaussian weight function to perform a 3DGaussianSplatting operation and get the result of the integral Splatting operation:

[0096]

[0097] Where S(x) is the surface contribution at position x, C is the covariance matrix, and D is the dimension of the point.

[0098] The import module 44 is used to import the reconstructed three-dimensional model into the GIS system, use the spatial database to manage the model data, and place the three-dimensional model at a corresponding position in the GIS system using the latitude and longitude coordinates.

[0099] The embodiment of the present invention is a device embodiment corresponding to the above method embodiment. The specific operations of each module can be understood by referring to the description of the method embodiment, which will not be repeated here.

[0100] Device Example 2

[0101] An embodiment of the present invention provides an electronic device, such as Figure 5 As shown, it includes: a memory 50, a processor 52, and a computer program stored in the memory 50 and executable on the processor 52. When the computer program is executed by the processor 52, the steps described in the method embodiment are implemented.

[0102] Device Example 3

[0103] An embodiment of the present invention provides a computer-readable storage medium, on which a program for implementing information transmission is stored. When the program is executed by the processor 52, the steps described in the method embodiment are implemented.

[0104] The computer-readable storage medium in this embodiment includes, but is not limited to, ROM, RAM, magnetic disk or optical disk, etc.

[0105] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A geographic information system (GIS) modeling method, characterized in that: include: The video image data of the target area is collected by the collection device and the longitude and latitude coordinates of the collection position are recorded, key frames are extracted from the video image data, and the extracted key frames are subjected to denoising, image feature points of the key frames are extracted, and feature matching is performed to determine the relative positions between the key frames; Using the matched image feature points and the GPS data of the acquisition device, the spatial coordinates of each image feature point are calculated by triangulation to generate point cloud data, and a Gaussian weight function is applied to each point in the point cloud data to perform a 3D Gaussian Splatting operation to obtain an integral Splatting operation result. Based on the integral Splatting operation result, a voxelization method is used to reconstruct a three-dimensional model, and smoothing is performed to optimize the surface quality of the three-dimensional model; The reconstructed three-dimensional model is imported into the GIS system, the model data is managed using a spatial database, and the three-dimensional model is placed at a corresponding position in the GIS system using the latitude and longitude coordinates.

2. The method according to claim 1, characterized in that: Extracting key frames from the video image data specifically includes: The key frames of the video image data are extracted by using a time interval frame extraction algorithm or a content-based frame extraction algorithm, wherein the time interval frame extraction algorithm is expressed as Formula 1: Frame n = Video[(n × Interval) mod TotalFrames] Formula 1; Among them, Frame n Indicates the video frame to be obtained, Video indicates the entire video file or video stream, n indicates the serial number of the frame to be selected, Interval indicates the selected interval or step size, mod indicates the remainder, and TotalFrames indicates the total number of frames in the video.

3. The method according to claim 1, characterized in that The denoising process for the extracted key frame specifically includes: According to Formula 2, the extracted key frame is denoised using an image denoising algorithm through a Gaussian filter: Among them, I is the original image, I filtered is the filtered image, Gσ is the Gaussian kernel with standard deviation σ, h is the size of the filter, x, y represent the coordinates of the pixel in the original image, and i, j represent the relative coordinates within the Gaussian kernel.

4. The method according to claim 1, characterized in that Extracting the image feature points of the key frames, performing feature matching, and determining the relative positions between the key frames specifically include: The SIFT or SURF algorithm is used to extract the image feature points of the key frames, and feature matching is performed to determine the relative positions between the key frames.

5. The method according to claim 1, characterized in that Applying a Gaussian weight function to each point in the point cloud data to perform a 3D Gaussian Splatting operation, and obtaining an integral Splatting operation result specifically includes: According to Formula 3, each point P in the point cloud data i Apply the Gaussian weight function to perform a 3DGaussianSplatting operation and get the result of the integral Splatting operation: Where S(x) is the surface contribution at position x, C is the covariance matrix, and D is the dimension of the point.

6. A geographic information system (GIS) modeling device, characterized in that: include: The acquisition, extraction and matching module is used to acquire video image data of the target area through an acquisition device and record the longitude and latitude coordinates of the acquisition position, extract key frames from the video image data, perform denoising on the extracted key frames, extract image feature points of the key frames, perform feature matching, and determine the relative positions between the key frames; A point cloud modeling module is used to calculate the spatial coordinates of each image feature point by triangulation using the matched image feature points and the GPS data of the acquisition device, generate point cloud data, apply a Gaussian weight function to each point in the point cloud data to perform a 3D Gaussian Splatting operation to obtain an integral Splatting operation result, reconstruct a three-dimensional model using a voxelization method based on the integral Splatting operation result, and perform smoothing to optimize the surface quality of the three-dimensional model; The import module is used to import the reconstructed three-dimensional model into the GIS system, use the spatial database to manage the model data, and use the latitude and longitude coordinates to place the three-dimensional model at a corresponding position in the GIS system.

7. The device according to claim 6, characterized in that The acquisition, extraction and matching module is specifically used for: The key frames of the video image data are extracted by using a time interval frame extraction algorithm or a content-based frame extraction algorithm, wherein the time interval frame extraction algorithm is expressed as Formula 1: Frame n = Video[(n × Interval) mod TotalFrames]; Equation 1 Among them, Frame n Indicates the video frame to be obtained, Video indicates the entire video file or video stream, n indicates the serial number of the frame to be selected, Interval indicates the selected interval or step size, mod indicates the remainder, and TotalFrames indicates the total number of frames in the video. According to Formula 2, the extracted key frame is denoised using an image denoising algorithm through a Gaussian filter: Among them, I is the original image, I filtered is the filtered image, Gσ is the Gaussian kernel with standard deviation σ, h is the size of the filter, x, y represent the coordinates of the pixel in the original image, and i, j represent the relative coordinates within the Gaussian kernel. The SIFT or SURF algorithm is used to extract the image feature points of the key frames, and feature matching is performed to determine the relative positions between the key frames.

8. The device according to claim 6, characterized in that The point cloud modeling module is specifically used for: According to Formula 3, each point P in the point cloud data i Apply the Gaussian weight function to perform a 3DGaussianSplatting operation and get the result of the integral Splatting operation: Where S(x) is the surface contribution at position x, C is the covariance matrix, and D is the dimension of the point.

9. An electronic device, characterized in that: include: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program, when executed by the processor, implements the steps of the GIS modeling method as described in any one of claims 1 to 5.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores an implementation program for information transmission, and when the program is executed by a processor, the steps of the GIS modeling method according to any one of claims 1 to 5 are implemented.

Images