A data acquisition method based on fast spelling images and three-dimensional point clouds

Abstract

The present invention discloses a data collection method based on quick-patch images and three-dimensional point clouds, the data collection method comprising: synchronously loading two-dimensional and three-dimensional data in a data collection and editing platform; acquiring auxiliary information; collecting contour information or feature point information of objects by comparing the quick-patch image data; and performing vector projection correction on the image data, thereby improving production efficiency.

Description

Technical Field

[0001] The present invention relates to the field of surveying and mapping, and in particular to a data acquisition method based on quick-matching images and three-dimensional point clouds. Background Art

[0002] In today's digital world with advanced science and technology, the data collection methods in the field of surveying and mapping geographic information have also improved. Surveying and mapping units often use a combination of multiple operation methods when collecting data, striving to ensure efficient and high-precision operation results, but most of the operation methods have their own limitations. For example, in the production of 1:500 topographic map data, in order to meet the accuracy standards, manual field collection is often used. This method has high accuracy, but consumes manpower and material resources and has low collection efficiency; in the production of 1:2000 topographic map data, aerial photography stereo collection is often used to reduce field workload, but this method has high requirements on personnel quality and requires special instruments and equipment to realize data collection in special environments. Therefore, when the production business time is tight and there is no aerial survey stereo collection equipment, we strive to explore a new data collection technology that can replace the above methods and realize the rapid and high-precision collection of geographic information data. Summary of the invention

[0003] In view of the above problems, the present invention is proposed to provide a data acquisition method based on quick-matching images and three-dimensional point clouds, which overcomes the above problems or at least partially solves the above problems.

[0004] According to one aspect of the present invention, a data collection method based on quick-matching images and three-dimensional point clouds is provided, the data collection method comprising:

[0005] Load 2D and 3D data simultaneously in the data collection and editing platform;

[0006] Get auxiliary information;

[0007] Collect the contour information or feature point information of the object by comparing the quick stitch image data;

[0008] Perform vector projection correction on image data.

[0009] Optionally, the synchronously loading two-dimensional and three-dimensional data in the data collection and editing platform specifically includes:

[0010] Load the quick-match image data and the point cloud data corresponding to the position and coordinates, and display the two-dimensional and three-dimensional data simultaneously.

[0011] Optionally, the obtaining of auxiliary information specifically includes:

[0012] The airborne positioning and orientation system POS (position and orientation system) is used to collect POS data;

[0013] When generating DOM image data, the image matching algorithm is used to process the overlapping parts of the images to generate a stitching network of the effective area;

[0014] Loads the information file required for frame projection correction.

[0015] Optionally, the POS data includes the camera station information, including the external position elements of the photograph measured by the POS system. .

[0016] Optionally, the information file required for loading the frame projection correction specifically includes: stitching line data of the quick stitching image data and POS data information file of the frame sensor.

[0017] Optionally, the collecting of contour information or feature point information of the object by comparing the quick stitch image data specifically includes:

[0018] The contour information or feature point information of the features is collected by comparing with the quick-matching image data. For the existing vector features, the elevation values ​​of the corresponding positions are manually collected on the point cloud data and linked to the vector data information.

[0019] Optionally, performing vector projection correction on the image data specifically includes:

[0020] The effective area stitching network generated during the production of the quick stitching data makes the angles and distances of the objects offset in multiple stitching planes different after the images are stitched;

[0021] Frame aerial photography refers to the aerial photography technology that images the entire frame at the same time at the moment of exposure. Its central projection principle is to image the object being photographed on the ground onto the image plane through a lens.

[0022] After in-depth analysis and research, the lodging characteristics of frame-width data have radial lodging characteristics centered on the image principal point;

[0023] The spatial relationship is analyzed based on the radial lodging characteristics, and the projection correction is performed after the correction amount is obtained.

[0024] Optionally, analyzing the spatial relationship according to the radial lodging characteristics and performing projection correction after obtaining the correction amount specifically includes:

[0025] Focus the viewing angle on the vertical section between the principal point of the image and the ground. The degree of lodging is affected by the angle α between the ground object and the principal point of the image. Using the inherent mathematical principles, the lodging deviation angle α is obtained:

[0026]

[0027] in, To collect coordinate plane values, is the image principal point coordinate;

[0028] The mosaic range of the object is determined according to the mosaic surface information of the quick mosaic image, and the corresponding unique principal point is read through the POS information file of the quick mosaic image;

[0029] According to the obtained lodging deviation angle, the lodging correction value of the object is calculated according to mathematical principles, and the contour and feature points are corrected back according to the coordinate direction;

[0030] Plane coordinates of real objects for:

[0031]

[0032] The projection correction technology based on quick-matching images and point cloud data uses the lodging principle and POS data to determine the lodging correction value.

[0033] The present invention provides a data collection method based on quick-patch images and three-dimensional point clouds, the data collection method comprising: synchronously loading two-dimensional and three-dimensional data in a data collection and editing platform; acquiring auxiliary information; collecting contour information or feature point information of objects by comparing the quick-patch image data; and performing vector projection correction on the image data, thereby improving production efficiency.

[0034] The above description is only an overview of the technical solution of the present invention. In order to more clearly understand the technical means of the present invention, it can be implemented according to the contents of the specification. In order to make the above and other purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0036] Figure 1 A flow chart of a data acquisition method based on quick-matching images and three-dimensional point clouds provided by an embodiment of the present invention;

[0037] Figure 2 A schematic diagram of the central projection principle provided by an embodiment of the present invention;

[0038] Figure 3 A frame-type central projection lodging characteristic map provided by an embodiment of the present invention;

[0039] Figure 4 A frame-type central projection profile characteristic diagram provided for an embodiment of the present invention. DETAILED DESCRIPTION

[0040] The exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

[0041] The terms "comprises" and "having" and any variations thereof in the description embodiments, claims and drawings of the present invention are intended to cover non-exclusive inclusions, for example, including a series of steps or units.

[0042] The technical solution of the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments.

[0043] like Figure 1 As shown, a data acquisition method based on quick-matching images and three-dimensional point clouds includes:

[0044] Data loading, including:

[0045] The Data Acquisition Platform (DAP) is a surveying and mapping data production software that integrates data acquisition and editing. Among them, the loading and display of two-dimensional vector data and three-dimensional data are its mature functions. Based on the basic functions of two-dimensional and three-dimensional data loading of the data acquisition and editing platform, the two-dimensional and three-dimensional synchronous loading function has been expanded and developed.

[0046] The two-dimensional and three-dimensional data are loaded synchronously in the data collection and editing platform, that is, the quick-matching image data and the point cloud data corresponding to its position and coordinates are loaded, and the two-dimensional and three-dimensional data are displayed at the same time to facilitate subsequent data collection.

[0047] Auxiliary information acquisition, including:

[0048] POS (position and orientation system) is a modern aerial photography navigation system based on the global positioning system and inertial measurement unit (IMU) to directly measure the external orientation elements of the image.

[0049] POS data refers to the camera station information, which is the external position elements of the image (Xp, Yp, Zp, ψ, ω, к) measured by the POS system.

[0050] The rapid stitching method of aerial survey images refers to the process of generating DOM image data. Since a single image cannot completely cover the entire study area, it is necessary to use an image matching algorithm to process the overlapping parts of the images. In the process of image mosaicking, in order to avoid areas with poor projection of houses, buildings, tree crowns, etc. on the orthophoto and areas with large color differences, the contribution of single images is selected to generate a stitching line network of the effective area.

[0051] Load the information files required for frame projection correction, including the stitching line data of the quick stitching image data and the POS data information files of the frame sensor, read and save. Generally, in the production process, the area of ​​surveying and mapping data collection is large, and the same area may be collected in multiple time periods. In order to facilitate the direct use of the projection correction function after collection, it is placed before data collection, so that the subsequent project can be closed and opened automatically. The operator can directly continue the unfinished work last time without having to repeatedly load auxiliary information.

[0052] The data collection process includes:

[0053] Since point cloud data has not been classified, its accuracy is high, but some objects in the data cannot be distinguished from their specific types, and it is not easy to collect the accurate plane position of the objects. The quick-scratching image data is spliced ​​after automatic processing and has not been manually edited. Although there are objects that fall down, its plane accuracy is high and the edges of the objects are clear. It is suitable for collecting plane coordinate information or feature point coordinate information of objects. After the two-dimensional and three-dimensional linkage, the two are combined for data collection, and the contour information or feature point information of the objects is collected by comparing the quick-scratching image data. For existing vector objects, the elevation values ​​of the corresponding positions are manually collected on the point cloud data and linked to the vector data information.

[0054] Vector projection correction, including:

[0055] DOM image data will have a plane offset due to the projection difference of the part of the object above the ground, resulting in a certain error in the spatial position accuracy, and the quick stitch data also has errors. During the data collection process, an algorithm specially designed for projection is used to solve the problem of correcting the spatial position of the object on the quick stitch image, so that the plane coordinates can be corrected to the correct position.

[0056] The network of stitching lines in the effective area generated during the production process of the quick stitching data makes the angles and distances of the objects offset in different stitching planes different after the images are stitched.

[0057] Digital aerial cameras are instruments and equipment mounted on aircraft platforms to photograph ground targets and obtain digital images. Digital aerial cameras can be divided into two types: frame-type (area array CCD) and push-broom type (line array CCD). Frame-type aerial photography refers to the aerial photography technology that simultaneously images the entire frame at the moment of exposure. Its central projection principle is to image the ground object being photographed onto the image plane through a lens, such as Figure 2 shown.

[0058] The lodging characteristics of frame data have radial lodging characteristics centered on the image principal point.

[0059] The image lodging is centered on the principal point of the image, and the building lodging is in all directions. The farther the distance between the object and the principal point of the image, the greater the lodging error; the higher the height of the object, the greater the lodging error. Through research, it is found that there are several obvious characteristics of image lodging: the lodging direction radiates from the principal point of the image to the surrounding areas, and the lodging offset is proportional to the distance from the object to the principal point of the image and the height of the object. The lodging characteristics of the frame-type central projection are as follows: Figure 3 shown.

[0060] Based on this feature, the spatial relationship is analyzed and the projection correction is performed after the accurate correction amount is obtained through the inherent mathematical principles. Figure 4 shown.

[0061] Focusing the viewing angle on the vertical section between the principal point of the image and the ground, it is found that the degree of lodging is affected by the angle α between the ground object and the principal point of the image. The larger the angle, the greater the lodging offset. Using the inherent mathematical principles, the lodging offset angle α is obtained:

[0062]

[0063] in, To collect coordinate plane values, are the image principal point coordinates.

[0064] The mosaic range where the object is located is determined according to the mosaic surface information of the quick mosaic image, and the corresponding unique image principal point is read through the POS information file of the quick mosaic image.

[0065] The lodging correction value of the object is calculated based on the obtained lodging offset angle according to mathematical principles, and the contour and feature points are corrected back according to the coordinate direction.

[0066] Plane coordinates of real objects for:

[0067] The projection correction technology based on quick-matching images and point cloud data uses the lodging principle and POS data to determine the lodging correction value.

[0068] Beneficial effects: The feature contours collected by the data collection method are corrected by projection, and the present invention realizes efficient and fast vector data collection.

[0069] The above specific implementation methods further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above are only specific implementation methods of the present invention and are not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A data acquisition method based on quick-matching images and three-dimensional point clouds, characterized in that: The data collection method comprises: Load 2D and 3D data simultaneously in the data collection and editing platform; Get auxiliary information; Collect the contour information or feature point information of the object by comparing the quick stitch image data; Perform vector projection correction on image data, including: The effective area stitching network generated during the production of the quick stitching data makes the angles and distances of the objects offset in multiple stitching planes different after the images are stitched; Frame aerial photography refers to the aerial photography technology that images the entire frame at the same time at the moment of exposure. Its central projection principle is to image the object being photographed on the ground onto the image plane through a lens. After in-depth analysis and research, the lodging characteristics of frame-width data have radial lodging characteristics centered on the image principal point; According to the radial lodging characteristics, the spatial relationship is analyzed and the correction amount is obtained to perform projection correction, including: Focus the viewing angle on the vertical section between the principal point of the image and the ground. The degree of lodging is affected by the angle α between the ground object and the principal point of the image. Using the inherent mathematical principles, the lodging deviation angle α is obtained: Among them, P(Xg, Yg, Zg) is the acquisition coordinate plane value, and P(Xp, Yp, Zp) is the image principal point coordinate; The mosaic range of the object is determined according to the mosaic surface information of the quick mosaic image, and the corresponding unique principal point is read through the POS information file of the quick mosaic image; According to the obtained lodging deviation angle, the lodging correction value of the object is calculated according to mathematical principles, and the contour and feature points are corrected back according to the coordinate direction; The plane coordinates R(Xr, Yr) of the real object are: The projection correction technology based on the quick stitch image and point cloud data uses the lodging principle and POS data to determine the lodging correction value. The height of the image from the ground under the lodging offset angle α is H is the vertical distance from the collection point to the ground.

2. The data acquisition method based on quick-matching images and three-dimensional point clouds according to claim 1 is characterized in that: The synchronous loading of two-dimensional and three-dimensional data in the data collection and editing platform specifically includes: Load the quick-match image data and the point cloud data corresponding to the position and coordinates, and display the two-dimensional and three-dimensional data simultaneously.

3. The data acquisition method based on quick-matching images and three-dimensional point clouds according to claim 1 is characterized in that: The obtaining of auxiliary information specifically includes: Use the airborne positioning and orientation system POS to collect POS data; When generating DOM image data, the image matching algorithm is used to process the overlapping parts of the images to generate a stitching network of the effective area; Loads the information file required for frame projection correction.

4. The data acquisition method based on quick-matching images and three-dimensional point clouds according to claim 3 is characterized in that: The POS data includes the camera station information, including the external position elements of the image (Xp, Yp, Zp, ψ, ω, K) measured by the POS system.

5. The data acquisition method based on quick-matching images and three-dimensional point clouds according to claim 3 is characterized in that: The information files required for loading frame projection correction specifically include: stitching line data of quick stitching image data and POS data information files of the frame sensor.

6. The data acquisition method based on quick-matching images and three-dimensional point clouds according to claim 1 is characterized in that: The contour information or feature point information of the object collected by comparing the quick stitch image data specifically includes: The contour information or feature point information of the features is collected by comparing with the quick-matching image data. For the existing vector features, the elevation values ​​of the corresponding positions are manually collected on the point cloud data and linked to the vector data information.

Images