Point cloud data processing method for earthwork

Abstract

The invention discloses a point cloud data processing method for earthwork, and belongs to the field of point cloud data processing. The point cloud data processing method comprises the following steps: acquiring three-dimensional scanning data; analyzing the three-dimensional scanning data to obtain a first point cloud file in an LAS format; performing point cloud cutting on the first point cloud file to obtain an initial point cloud of the target object; classifying and denoising the initial point cloud, and exporting the initial point cloud as a first intermediate point cloud; overturning the first intermediate point cloud, and exporting the first intermediate point cloud as a second intermediate point cloud; classifying and denoising the second intermediate point cloud, and exporting a third intermediate point cloud; turning over the third intermediate point cloud, and exporting a de-noising point cloud of the target object; obtaining a reference image which covers a target object and has geographical coordinate information; aligning the de-noised point cloud with the reference image; and taking the RGB value of the pixel in the image as the additional attribute of the corresponding point in the de-noising point cloud, and exporting to obtain a color point cloud. According to the invention, the accuracy of point cloud denoising is improved, and the cost of purchasing a three-dimensional laser color scanner is saved.

Description

Technical Field

[0001] The present invention belongs to the field of point cloud data processing, and in particular relates to a point cloud data processing method for earthwork engineering. Background Art

[0002] In a foundation treatment and earthwork project for an airport expansion and renovation project, 3D laser scanning technology was innovatively adopted, combined with BIM modeling, and applied to earthwork collection and measurement. 3D laser scanning technology offers advantages such as intuitiveness, authenticity, and convenience. Compared to traditional measurement methods, it can reduce management difficulties, save personnel costs, improve work efficiency, and generate favorable economic benefits. In actual project applications, the software included with handheld 3D laser scanners often fails to meet data processing requirements. Currently, the project construction process faces prominent issues such as data noise processing, the high cost of high-precision color laser scanners, and the large, slow opening and storage difficulties of point cloud data files. Summary of the Invention

[0003] The purpose of the present invention is to overcome the shortcomings of the existing technology and provide a point cloud data processing method for earthwork engineering, which can accurately remove noise in the point cloud.

[0004] The present invention is implemented through the following technical solution: A method for processing point cloud data of earthwork engineering, comprising:

[0005] Acquiring three-dimensional scanning data collected by using a three-dimensional laser colorless scanner;

[0006] Analyze the 3D scanning data to obtain the first point cloud file in LAS format;

[0007] Perform point cloud cutting on the first point cloud file in Trimble RealWorks software to obtain the initial point cloud of the target object;

[0008] The initial point cloud was classified and denoised in Trimble RealWorks software, and then exported as the first intermediate point cloud;

[0009] Flip the first intermediate point cloud in MicroStation software and then export it as the second intermediate point cloud;

[0010] The second intermediate point cloud is classified and denoised in Trimble RealWorks software, and then exported as the third intermediate point cloud;

[0011] In MicroStation software, the third intermediate point cloud is flipped and then exported as a denoised point cloud of the target object;

[0012] Obtain a reference image covering the target object and with geographic coordinate information;

[0013] Write a registration algorithm script in Cyclone3DR software to align the denoised point cloud and the reference image;

[0014] A script was written in the Cyclone3DR software to use the RGB values ​​of the pixels in the image as additional attributes of the corresponding points in the denoised point cloud, and then exported to obtain a colored point cloud.

[0015] Furthermore, the point cloud data processing method further includes:

[0016] Obtaining a measurement data file of the target object, wherein the measurement data file is in dat format;

[0017] Convert measurement data files into DWG format based on CAD software;

[0018] In Cyclone3DR software, register the measurement data file and the denoised point cloud or colored point cloud to the same coordinate system;

[0019] In Cyclone3DR software, create measurement points representing the actual measured locations on site based on the measurement data file;

[0020] In Cyclone3DR software, use the distance analysis tool to calculate the distance between each measurement point and the denoised point cloud or colored point cloud surface;

[0021] Adjusts measurement points whose distance to the surface of a denoised point cloud or colored point cloud exceeds a preset range.

[0022] Furthermore, the distance analysis tools are used to calculate the distance of each measurement point to the denoised point cloud or colored point cloud surface, including:

[0023] Perform triangulation modeling on denoised point clouds or colored point clouds, and calculate triangulation deviations on measured points.

[0024] Furthermore, the measurement points whose distance to the denoised point cloud or the colored point cloud surface exceeds a preset range are adjusted, including:

[0025] The measurement points whose distances to the surface of the denoised point cloud or the colored point cloud exceed the preset range are determined as the measurement points to be adjusted;

[0026] Analyze the reason why the distance between the measurement point to be adjusted and the denoised point cloud or colored point cloud surface exceeds the preset range;

[0027] Based on the reasons obtained from the analysis, corresponding adjustments are made to ensure that the distances corresponding to all measurement points are within the preset range.

[0028] Furthermore, corresponding adjustments are made based on the reasons obtained through analysis, including:

[0029] For the measurement points to be adjusted whose distance exceeds the preset range due to smoothing, an adjustment area is determined;

[0030] Adjust the smoothing parameters to re-smooth the adjustment area.

[0031] Furthermore, the smoothing parameters are adjusted to re-smooth the adjustment area, including:

[0032] Adjust the smoothing parameters, then perform smoothing on the adjustment area, and determine whether the smoothed adjustment area meets the requirements;

[0033] If the smoothed adjustment area meets the requirements, the re-smoothing process of the adjustment area is terminated;

[0034] If the adjustment area after smoothing does not meet the requirements, the smoothing parameters are adjusted again to smooth the adjustment area until the adjustment area meets the requirements.

[0035] Furthermore, the point cloud data processing method further includes:

[0036] Import the denoised point cloud or colored point cloud into Trimble RealWorks software to generate a second point cloud file in DWG format;

[0037] Import the second point cloud file into OpenRoadsDesigner software;

[0038] In the OpenRoadsDesigner software, adjust the parameters according to the preset view values, and then print the second point cloud file as a JPG grayscale image.

[0039] Furthermore, the point cloud data processing method further includes:

[0040] In the OpenRoadsDesigner software, select Get Surface from Image, then select the JPG grayscale image, and import the JPG grayscale image into the OpenRoadsDesigner software;

[0041] Set parameters in the OpenRoadsDesigner software to reconstruct the triangulated network model and restore the JPG grayscale image to a point cloud model of the target object.

[0042] Furthermore, the point cloud data processing method further includes:

[0043] Perform thin sampling on the denoised point cloud or colored point cloud, build a triangulated network, smooth it and export the third intermediate point cloud in DWG format;

[0044] The third intermediate point cloud is imported into OpenRoadsDesigne software to automatically identify the mesh for modeling.

[0045] Furthermore, the reference image is a drone aerial image or a satellite map.

[0046] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0047] (1) The present invention first classifies and denoises the initial point cloud, then flips the point cloud, and then classifies and denoises the flipped point cloud for the second time, and finally flips the point cloud back to its original state. This overcomes the shortcoming of Trimble RealWorks software that easily identifies ground points as noise points when performing point cloud denoising, thereby improving the accuracy of point cloud denoising.

[0048] (2) The present invention uses the method of image inversion to give point cloud RGB attributes, making the point cloud data more realistic and achieving the same effect as a 3D laser color scanner, thus saving the cost of purchasing a 3D laser color scanner;

[0049] (3) The present invention combines point cloud with traditional measurement. Through data processing and file format adjustment, it realizes comparative analysis between point cloud data and measurement points, thus improving the accuracy and authenticity of point cloud.

[0050] (4) The present invention converts the point cloud file into a grayscale image for local storage, turning a file of tens of GB into tens of MB, saving storage resources and indirectly saving economic costs and personnel time costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] The drawings described herein are used to provide a further understanding of the embodiments of the present invention, constitute a part of this application, and do not constitute a limitation of the embodiments of the present invention. In the drawings:

[0052] Figure 1 A flow chart of a point cloud data processing method;

[0053] Figure 2 This is an original point cloud image without denoising in an example;

[0054] Figure 3 A schematic diagram of noise points on a point cloud model in an example;

[0055] Figure 4 A schematic diagram of noise points separated by automatic system classification in an example;

[0056] Figure 5 A schematic diagram of an example where a desired part is misclassified as noise;

[0057] Figure 6 A diagram showing an example of noise that the automatic system classification failed to remove;

[0058] Figure 7 This is a schematic diagram of an example where a point cloud is flipped upside down;

[0059] Figure 8 This is a schematic diagram of the point cloud after denoising in an example;

[0060] Figure 9 Schematic diagram of the third intermediate point cloud obtained based on the denoised point cloud in an example;

[0061] Figure 10 A schematic diagram of the automatic mesh recognition of the third intermediate point cloud imported into OpenRoadsDesigne software in an example;

[0062] Figure 11 is a schematic diagram of a reference image in an example;

[0063] Figure 12 A schematic diagram of aligning a point cloud with a reference image in one example;

[0064] Figure 13 A schematic diagram of a colored point cloud in an example;

[0065] Figure 14 This is a schematic diagram of the elevation points collected in an example after being imported into CAD software;

[0066] Figure 15 This is a schematic diagram of an example after the denoised point cloud is imported into Cyclone3DR software;

[0067] Figure 16 A schematic diagram of triangulation deviation calculation for survey points in an example;

[0068] Figure 17 A schematic diagram of a measurement point where the distance exceeds the limit in an example;

[0069] Figure 18 A schematic diagram showing an example where the observation point is located too many times above the triangulation network;

[0070] Figure 19 A schematic diagram of local adjustment of smoothing parameters when modeling a triangulated network in an example;

[0071] Figure 20 A diagram showing how to readjust the smoothing parameter after local adjustment of the smoothing parameter in an example;

[0072] Figure 21 This is a diagram of an example after smoothing parameter adjustment is completed;

[0073] Figure 22 This is a schematic diagram of an example after the noise-removed point cloud is imported into Trimble RealWorks software;

[0074] Figure 23 A schematic diagram of triangulated network modeling of point clouds in an example;

[0075] Figure 24 This is a schematic diagram of converting a triangulated mesh model into a grayscale image and saving it in JPG format;

[0076] Figure 25 This is a diagram of setting parameters to reconstruct a triangulated network model from an image in an example;

[0077] Figure 26 A schematic diagram of the reconstructed triangulated network model in an example. DETAILED DESCRIPTION

[0078] The technical solution of the present invention will be further described below with reference to the accompanying drawings and through specific implementation methods.

[0079] Among them, the accompanying drawings are only for illustrative purposes and represent only schematic diagrams rather than actual pictures, and should not be understood as limiting the present invention. In order to better illustrate the embodiments of the present invention, some parts of the accompanying drawings may be omitted, enlarged or reduced, and do not represent the dimensions of actual products. For those skilled in the art, it is understandable that some well-known structures and their descriptions may be omitted in the accompanying drawings.

[0080] The same or similar numbers in the drawings of the embodiments of the present invention correspond to the same or similar parts; in the description of the present invention, it should be understood that if the terms "upper", "lower", "left", "right", "inside", "outside" and the like indicate directions or positional relationships, they are based on the directions or positional relationships shown in the drawings, which are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific direction, be constructed and operated in a specific direction. Therefore, the terms describing the positional relationship in the drawings are only used for illustrative purposes and cannot be understood as limiting the present invention. For ordinary technicians in this field, the specific meanings of the above terms can be understood according to specific circumstances.

[0081] In the description of the present invention, unless otherwise expressly specified or limited, when the term "connection" or the like appears to indicate a connection relationship between components, such term should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium; it can be internal communication between two components or an interaction between two components. For those skilled in the art, the specific meanings of the above terms in the present invention can be understood in specific circumstances.

[0082] like Figures 1 to 26 As shown, this embodiment discloses a point cloud data processing method for earthwork engineering.

[0083] like Figure 1 As shown, a point cloud data processing method for earthwork engineering includes steps S001 to S010.

[0084] Step S001: Acquire three-dimensional scanning data collected by a three-dimensional laser colorless scanner.

[0085] The 3D laser scanner uses the principle of laser ranging. By recording the 3D coordinates, reflectivity, texture and other information of a large number of dense points on the surface of the object being measured, it can quickly restore the 3D model of the object being measured and various drawing data such as lines, surfaces and bodies.

[0086] Three-dimensional laser scanners can be divided into three-dimensional laser colorless scanners and three-dimensional laser color scanners. High-precision three-dimensional laser color scanners are too expensive, so a three-dimensional laser colorless scanner is used in this embodiment.

[0087] Three-dimensional laser scanners can be divided into various types, such as airborne and handheld types. In this embodiment, a three-dimensional laser colorless scanner equipped with RTK is used.

[0088] Step S002: Analyze the three-dimensional scanning data to obtain a first point cloud file in LAS format.

[0089] Specifically, the three-dimensional scanning data is parsed to obtain a first point cloud file, which is in LAS format.

[0090] Step S003: Perform point cloud cutting on the first point cloud file in Trimble RealWorks software to obtain an initial point cloud of the target object.

[0091] Specifically, the first point cloud file in LAS format is imported into Trimble RealWorks software, and point cloud cutting is performed on the first point cloud file in Trimble RealWorks to remove areas irrelevant to the target object, thereby obtaining a point cloud file of the target object, which is recorded as the initial point cloud of the target object.

[0092] Step S004: Classify and denoise the initial point cloud in Trimble RealWorks software, and then export it as a first intermediate point cloud.

[0093] For example, the point cloud can be divided into noise, ground, dust, machinery, buildings, etc., and then the noise can be removed, as well as unnecessary machinery, buildings, etc.

[0094] Specifically, it is first determined whether there is noise information that needs to be processed in the initial point cloud. If there is noise information in the initial point cloud, the initial point cloud is classified and denoised in the Trimble RealWorks software. After denoising is completed, the first intermediate point cloud in ASC format is exported.

[0095] In this embodiment, the denoising function in the Trimble RealWorks software can be used.

[0096] Step S005: Flip the first intermediate point cloud in MicroStation software, and then export it as a second intermediate point cloud.

[0097] Specifically, the first intermediate point cloud in the ASC format is imported into the MicroStation software, and then the first intermediate point cloud is flipped in the MicroStation software, and then exported as the second intermediate point cloud.

[0098] For example, the first intermediate point cloud is flipped upside down in MicroStation software.

[0099] Step S006: Classify and denoise the second intermediate point cloud in Trimble RealWorks software, and then export it as a third intermediate point cloud.

[0100] Specifically, the second intermediate point cloud is imported into the Trimble RealWorks software, and then the second intermediate point cloud is classified and denoised in the Trimble RealWorks software, and then exported to the third intermediate point cloud in the ASC format.

[0101] Step S007: Flip the third intermediate point cloud in MicroStation software, and then export it as a denoised point cloud of the target object.

[0102] Specifically, the third intermediate point cloud in ASC format is imported into MicroStation software for flipping and restoration, and then exported to obtain the denoised point cloud corresponding to the initial point cloud denoising, and the denoised point cloud is in LAS format.

[0103] In this embodiment, by first performing the first classification and denoising on the initial point cloud, then flipping the point cloud, and then performing the second classification and denoising on the flipped point cloud, and finally flipping the point cloud back to restore, the shortcomings of the Trimble RealWorks software in point cloud denoising, such as easily identifying ground points as noise points and missing noise points, are overcome, thereby improving the accuracy of point cloud denoising.

[0104] Step S008: Acquire a reference image covering the target object and including geographic coordinate information.

[0105] For example, the reference image can be an image taken by a drone or a satellite map.

[0106] The reference image must contain the complete target object.

[0107] The reference image is a high-definition image with no obstructions to the target object. The higher the definition of the reference image, the better.

[0108] Step S009: Write a registration algorithm script in the Cyclone3DR software to align the denoised point cloud and the reference image.

[0109] Specifically, the scale parameters of the reference image were adjusted in the Cyclone3DR software, and then the reference image was aligned with the denoised point cloud.

[0110] For example, first write a script for the registration algorithm in the Cyclone3DR software, then import the denoised point cloud and the reference image into the Cyclone3DR software, and use the registration algorithm to correspond the points in the denoised point cloud with the pixels in the reference image, thereby achieving registration of the denoised point cloud and the reference image.

[0111] The registration algorithm in this embodiment can adopt the existing related algorithm.

[0112] Step S010. Write a script in the Cyclone3DR software to use the RGB values ​​of the pixels in the image as additional attributes of the corresponding points in the denoised point cloud, and then export it to obtain a colored point cloud.

[0113] In this embodiment, the point cloud is given RGB attributes by using an image inverse point cloud method, making the point cloud data more realistic and achieving the same effect as a three-dimensional laser color scanner, thereby saving the cost of purchasing a three-dimensional laser color scanner.

[0114] In one example, a 3D laser colorless scanner is used to perform a 3D scan of the ground of an airport flight area, and the 3D scan data is parsed to obtain a first point cloud file in LAS format. The first point cloud file is then imported into Trimble RealWorks software, such as Figure 2 shown; from Figure 3 It can be seen from the figure that there are noise points other than the soil surface in the point cloud model that need to be processed, such as Figure 3 The position indicated by the arrow; Figure 4 As shown in Figure 1, most noise points can be classified through automatic system classification in TrimbleRealWorks software. The positions indicated by arrows in the figure are the separated noise points. Figure 5 As shown in the figure, the automatic system classification may mistakenly remove the required point cloud as noise. For example, the pink part of the bottom point in the figure is the part that is automatically removed, but it is actually the part that is needed; Figure 6 As shown in the figure, the circled part is the part that the automatic system classification denoising fails to remove successfully. For this part, the existing method can only remove it manually; Figure 7As shown, after the automatic denoising is reversed using the method of this embodiment, the system can successfully identify the parts ignored by the forward denoising, and the denoising effect is relatively ideal; Figure 8 The denoised point cloud obtained after denoising is shown. It can be seen that the denoising effect is ideal; Figure 9 The third intermediate point cloud obtained based on the denoised point cloud is shown, for example, the denoised point cloud or the colored point cloud is subjected to sparse sampling, a triangulated network is established, and the third intermediate point cloud in DWG format is exported after smoothing; the third intermediate point cloud is imported into the OpenRoadsDesigne software to automatically identify the mesh for modeling, such as Figure 10 As shown;

[0115] Figure 11 A reference image covering the target object and with geographic coordinate information is shown; Figure 12 As shown in , adjust the scale parameters of the reference image in Cyclone3DR software, and then align the reference image with the denoised point cloud; Figure 13 As shown in Figure 1, after the denoised point cloud is given RGB attributes through calculation, a colored point cloud is obtained; as shown in Table 1, it can be seen that the obtained colored point cloud already has RGB attributes.

[0116] Table 1 Additional attributes of a point in a color point cloud:

[0117]

[0118] In some implementations of this embodiment, the point cloud data processing method further includes steps S011 to S016.

[0119] Step S011: Obtain a measurement data file of the target object, wherein the measurement data file is in a dat format.

[0120] For example, RTK is used to collect measurement data of a target object.

[0121] Step S012: Convert the measurement data file into DWG format based on CAD software.

[0122] Specifically, the measurement data file in dat format is imported into the CAD software, and then the measurement data file is exported to DWG format.

[0123] The measurement data file includes feature point data of the target object.

[0124] Step S013: In the Cyclone3DR software, the measurement data file and the denoised point cloud or the colored point cloud are registered to the same coordinate system.

[0125] Step S014: In the Cyclone3DR software, create measurement points representing the actual measured positions on site based on the measurement data file.

[0126] The number of measuring points is determined according to actual needs.

[0127] Step S015: In the Cyclone3DR software, use the distance analysis tool to calculate the distance between each measurement point and the denoised point cloud or the colored point cloud surface.

[0128] Step S016: Adjust the measurement points whose distances to the denoised point cloud or the color point cloud surface exceed a preset range.

[0129] In some implementations of this embodiment, adjusting measurement points whose distances to the denoised point cloud or the colored point cloud surface exceed a preset range includes: determining the measurement points whose distances to the denoised point cloud or the colored point cloud surface exceed the preset range as measurement points to be adjusted; analyzing the reason why the distances to the measurement points to be adjusted from the denoised point cloud or the colored point cloud surface exceed the preset range; and making corresponding adjustments based on the analyzed reasons to ensure that the distances corresponding to all measurement points are within the preset range.

[0130] The reason why the distance between the measurement point to be adjusted and the surface of the denoised point cloud or the colored point cloud exceeds the preset range includes errors caused by mesh smoothing and errors during data acquisition.

[0131] In some implementations of this embodiment, corresponding adjustments are made based on the reasons obtained through analysis, including: determining an adjustment area for measurement points to be adjusted whose distance exceeds a preset range due to smoothing; and adjusting smoothing parameters to re-smooth the adjustment area.

[0132] In these implementations, point clouds are combined with traditional measurements. Through data processing and file format adjustment, comparative analysis between point cloud data and measurement points is achieved, thereby improving the accuracy and authenticity of the point cloud.

[0133] In one example, a measurement data file of a target object (such as elevation point data collected on site) is obtained, and the measurement data file is converted into DWG format using CAD software, such as Figure 14 As shown; import the measurement data file and the denoised point cloud or color point cloud into the Cyclone3DR software, as shown Figure 15 As shown in the figure; create measurement points based on the measurement data file; use the distance analysis tool to perform triangulation modeling on the denoised point cloud or colored point cloud, and calculate the triangulation deviation of the measurement points, as shown in the figure. Figure 16 As shown; find the measurement point that exceeds the limit, such as Figure 17 As shown in the figure, we start to investigate the cause by observing that the measurement points are located too many times above the triangulation network, such as Figure 18 As shown in the figure, the smoothing parameters of the triangulated network modeling are locally adjusted, such as Figure 19 As shown; after local restoration, it was found that the smoothing parameter was too large, resulting in the loss of elevation information. The smoothing parameter was readjusted, as shown in Figure 20As shown in the figure, the deviation between the measured point elevation and the 3D scanning data meets the requirements, and the adjustment is completed. Figure 21 shown.

[0134] In some implementations of this embodiment, the point cloud data processing method further includes steps S017 to S019.

[0135] Step S017: Import the denoised point cloud or the colored point cloud into the Trimble RealWorks software to generate a second point cloud file in DWG format.

[0136] Step S018: Import the second point cloud file into the OpenRoadsDesigner software.

[0137] Step S019. In the OpenRoadsDesigner software, adjust the parameters according to the preset view values, and then print out the second point cloud file as a JPG grayscale image.

[0138] Adjust parameters according to preset view values, including background extension, grayscale, image ratio, etc.

[0139] Point cloud data files for earthwork projects are often very large, slow to open, and difficult to store. These implementations, by defining basic standard parameters, convert point cloud files into grayscale images for local storage, reducing files of tens of gigabytes to tens of megabytes. This saves storage resources and indirectly reduces both financial and personnel time.

[0140] In one example, a denoised point cloud or a colored point cloud is imported into Trimble RealWorks software, such as Figure 22 As shown; generate a second point cloud file in DWG format, import the second point cloud file into OpenRoadsDesigner software, and perform triangulation modeling on the point cloud, as shown Figure 23 As shown; by setting the parameters, the triangulated network format is converted to a grayscale image, and the grayscale image is saved as a JPG image format, as shown Figure 24 shown.

[0141] In some implementations of this embodiment, the point cloud data processing method further includes: in the OpenRoadsDesigner software, selecting to obtain a surface from an image, and then selecting the JPG grayscale image to obtain a restored point cloud model of the target object.

[0142] In one example, in OpenRoadsDesigner software, select Get Surface from Image and import the grayscale image; set parameters to reconstruct the triangulated network model, such as Figure 25 As shown; restore the data saved in the grayscale image in JPG format into a triangulated network model, as shown Figure 26 shown.

[0143] In this embodiment Figure 16 、 Figure 17 、 Figure 21 The Chinese characters represent the distance measurement results and the measurement results on the dZ (Z axis).

[0144] The specific implementation methods described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific implementation method of the present invention and is 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 method for processing point cloud data of earthwork engineering, characterized in that: include: Acquiring three-dimensional scanning data collected by using a three-dimensional laser colorless scanner; Analyze the 3D scanning data to obtain the first point cloud file in LAS format; Perform point cloud cutting on the first point cloud file in Trimble RealWorks software to obtain the initial point cloud of the target object; The initial point cloud was classified and denoised in Trimble RealWorks software, and then exported as the first intermediate point cloud; Flip the first intermediate point cloud in MicroStation software and then export it as the second intermediate point cloud; The second intermediate point cloud is classified and denoised in Trimble RealWorks software, and then exported as the third intermediate point cloud; In MicroStation software, the third intermediate point cloud is flipped and then exported as a denoised point cloud of the target object; Obtain a reference image covering the target object and with geographic coordinate information; Write a registration algorithm script in Cyclone3DR software to align the denoised point cloud and the reference image; A script was written in the Cyclone3DR software to use the RGB values ​​of the pixels in the image as additional attributes of the corresponding points in the denoised point cloud, and then exported to obtain a colored point cloud.

2. The method for processing point cloud data of earthwork engineering according to claim 1, characterized in that: The point cloud data processing method further includes: Obtaining a measurement data file of the target object, wherein the measurement data file is in dat format; Convert measurement data files into DWG format based on CAD software; In Cyclone3DR software, register the measurement data file and the denoised point cloud or colored point cloud to the same coordinate system; In Cyclone3DR software, create measurement points representing the actual measured locations on site based on the measurement data file; In Cyclone3DR software, use the distance analysis tool to calculate the distance between each measurement point and the denoised point cloud or colored point cloud surface; Adjusts measurement points whose distance to the surface of a denoised point cloud or colored point cloud exceeds a preset range.

3. The method for processing point cloud data of earthwork engineering according to claim 2, characterized in that: Calculate the distance of each measurement point to the surface of a denoised or colored point cloud using distance analysis tools, including: Perform triangulation modeling on denoised point clouds or colored point clouds, and calculate triangulation deviations on measured points.

4. The method for processing point cloud data of earthwork engineering according to claim 2, characterized in that: Adjustments are made to measurement points whose distance to the denoised point cloud or colored point cloud surface exceeds a preset range, including: The measurement points whose distances to the surface of the denoised point cloud or the colored point cloud exceed the preset range are determined as the measurement points to be adjusted; Analyze the reason why the distance between the measurement point to be adjusted and the denoised point cloud or colored point cloud surface exceeds the preset range; Based on the reasons obtained from the analysis, corresponding adjustments are made to ensure that the distances corresponding to all measurement points are within the preset range.

5. The method for processing point cloud data of earthwork engineering according to claim 4, characterized in that: Make corresponding adjustments based on the reasons obtained through analysis, including: For the measurement points to be adjusted whose distance exceeds the preset range due to smoothing, an adjustment area is determined; Adjust the smoothing parameters to re-smooth the adjustment area.

6. The method for processing point cloud data of earthwork engineering according to claim 5, characterized in that: Adjust the smoothing parameters to re-smooth the adjustment area, including: Adjust the smoothing parameters, then perform smoothing on the adjustment area, and determine whether the smoothed adjustment area meets the requirements; If the smoothed adjustment area meets the requirements, the re-smoothing process of the adjustment area is terminated; If the adjustment area after smoothing does not meet the requirements, the smoothing parameters are adjusted again to smooth the adjustment area until the adjustment area meets the requirements.

7. The method for processing point cloud data of earthwork engineering according to claim 1, characterized in that: The point cloud data processing method further includes: Import the denoised point cloud or colored point cloud into Trimble RealWorks software to generate a second point cloud file in DWG format; Import the second point cloud file into OpenRoadsDesigner software; In the OpenRoadsDesigner software, adjust the parameters according to the preset view values, and then print the second point cloud file as a JPG grayscale image.

8. The method for processing point cloud data of earthwork engineering according to claim 7, characterized in that: The point cloud data processing method further includes: In the OpenRoadsDesigner software, select Get Surface from Image, then select the JPG grayscale image, and import the JPG grayscale image into the OpenRoadsDesigner software; Set parameters in the OpenRoadsDesigner software to reconstruct the triangulated network model and restore the JPG grayscale image to a point cloud model of the target object.

9. The method for processing point cloud data of earthwork engineering according to claim 1, characterized in that: The point cloud data processing method further includes: Perform thin sampling on the denoised point cloud or colored point cloud, build a triangulated network, smooth it and export the third intermediate point cloud in DWG format; The third intermediate point cloud is imported into OpenRoadsDesigne software to automatically identify the mesh for modeling.

10. The method for processing point cloud data of earthwork engineering according to claim 1, characterized in that: The reference image is an unmanned aerial image or a satellite map.

Images