Point cloud-based assisted three-dimensional model construction method

Abstract

The application provides a point cloud-based auxiliary three-dimensional model construction method, which acquires point cloud distribution data of a target object surface by using a point cloud scanning device, partitions and corrects the object surface, enriches the point cloud data amount of the point cloud distribution area, maximally ensures that each point cloud distribution area includes sufficient point cloud data, realizes three-dimensional construction of the target object surface, removes abnormal point cloud data, ensures the credibility of the overall point cloud data, and improves the construction accuracy and efficiency of the three-dimensional model.

Description

Technical Field

[0001] This invention relates to the technical field of automatic 3D model construction, and particularly to a point cloud-based assisted 3D model construction method. Background Technology

[0002] Point cloud scanning mainly includes two methods: laser-based point cloud scanning and image-based point cloud scanning. By illuminating or photographing an object with a laser, the laser reflection intensity or image color at different locations on the object's surface is obtained. Then, the laser reflection intensity or image color is correlated with the corresponding 3D coordinates of these points to generate point cloud data, thereby constructing a 3D model of the object's surface. The more data points the point cloud data contains, the more accurately the constructed 3D model matches the object's true appearance. Current technologies involve manually inputting all point cloud data directly into 3D modeling software for 3D model construction. This approach does not perform pre-screening or other processing of the point cloud data and lacks automatic classification and adjustment capabilities. This can easily lead to abnormal point cloud data affecting the accuracy of the 3D model construction, thus reducing the reliability and efficiency of the 3D model construction. Summary of the Invention

[0003] To address the shortcomings of existing technologies, this invention provides a point cloud-based assisted 3D model construction method. It utilizes a point cloud scanning device to obtain point cloud distribution data of the target object's surface, thereby acquiring corresponding point cloud distribution features. The target object's surface is divided into several point cloud distribution regions. Based on the point cloud distribution features within these regions, the corresponding regions are re-scanned to obtain new point cloud distribution data, which is then used to correct these regions. All point cloud distribution regions are integrated to obtain point cloud distribution data for the entire target object's surface, and curve fitting is performed to obtain several point cloud distribution curves. After adjusting the point distribution of each curve, a 3D model of the target object's surface is constructed based on all the curves. This method uses a point cloud scanning device to acquire point cloud distribution data of the target object's surface, thereby partitioning and correcting the surface, enriching the point cloud data volume in each region, maximizing the inclusion of sufficient point cloud data in each region, achieving 3D construction of the target object's surface, and simultaneously removing abnormal point cloud data to ensure the overall reliability of the point cloud data, thus improving the accuracy and efficiency of 3D model construction.

[0004] This invention provides a point cloud-based assisted 3D model construction method, which includes the following steps:

[0005] Step S1: Scan the target object using a point cloud scanning device to obtain point cloud distribution data of the target object surface; analyze and process the point cloud distribution data to obtain point cloud distribution features of the target object surface; and then divide the target object surface into several point cloud distribution regions based on the point cloud distribution features.

[0006] Step S2: Based on the point cloud distribution characteristics of the point cloud distribution area, the corresponding area of ​​the target object surface is re-scanned using the point cloud scanning device; based on the point cloud distribution data obtained from the re-scanning, the corresponding point cloud distribution area is corrected.

[0007] Step S3: Integrate all point cloud distribution areas to obtain point cloud distribution data for the entire surface area of ​​the target object; perform curve fitting on the point cloud distribution data for the entire surface area of ​​the target object to obtain several point cloud distribution curves for the entire surface area of ​​the target object.

[0008] Step S4: After adjusting the point distribution of each point cloud distribution curve, a three-dimensional surface model of the target object is constructed based on all point cloud distribution curves.

[0009] Further, in step S1, the target object is scanned using a point cloud scanning device to obtain point cloud distribution data of the target object's surface; the point cloud distribution data is analyzed and processed to obtain the point cloud distribution features of the target object's surface, specifically including:

[0010] A laser point cloud scanning device is used to perform an all-round scan of the surface of the target object to obtain the three-dimensional coordinates and laser reflection intensity data of several points on the surface of the target object; the three-dimensional coordinates and laser reflection intensity data of all points are transformed and processed to obtain the point cloud distribution data corresponding to all points.

[0011] The point cloud distribution data is processed by adjacent point distance analysis to obtain the point cloud data distribution spacing characteristics of the target object surface.

[0012] Furthermore, in step S1, dividing the surface of the target object into several point cloud distribution regions based on the point cloud distribution characteristics specifically includes:

[0013] Based on the point cloud data distribution spacing characteristics, the surface of the target object is divided into several point cloud distribution regions with corresponding average point cloud spacing; wherein, the average point cloud spacing of each point cloud distribution region is different; the average point cloud spacing refers to the average distance between each point cloud contained in the point cloud distribution region and its adjacent other point clouds.

[0014] Furthermore, in step S2, rescanning the corresponding area of ​​the target object surface using the point cloud scanning device based on the point cloud distribution characteristics of the point cloud distribution area specifically includes:

[0015] If the average point cloud spacing in the point cloud distribution area is less than or equal to a preset spacing threshold, then it is not necessary to rescan the area corresponding to the target object surface and the point cloud distribution area.

[0016] If the average point cloud density of the point cloud distribution area is greater than a preset spacing threshold, then the boundary line between the target object surface and the corresponding area of ​​the point cloud distribution area is determined; and the boundary line is used as the scanning boundary to instruct the laser point cloud scanning device to rescan the corresponding area of ​​the target object surface; wherein the scanning frequency of the rescanning is greater than the scanning frequency of the previous scan.

[0017] Furthermore, in step S2, the correction process for the corresponding point cloud distribution area based on the rescanned point cloud distribution data specifically includes:

[0018] The point cloud distribution data obtained from the rescan is superimposed onto the point cloud distribution data obtained from the previous scan of the corresponding point cloud distribution area to obtain new point cloud distribution data for the corresponding point cloud distribution area, thereby achieving the correction processing of the corresponding point cloud distribution area.

[0019] Furthermore, in step S3, the integration processing of all point cloud distribution areas to obtain point cloud distribution data for the entire surface area of ​​the target object specifically includes:

[0020] Based on the boundary distribution information of each point cloud distribution area, determine all point cloud distribution areas adjacent to each point cloud distribution area; then, based on the relative positional relationship between all two adjacent point cloud distribution areas, stitch together and integrate all point cloud distribution areas to obtain point cloud distribution data for the entire surface area of ​​the target object.

[0021] Furthermore, in step S3, curve fitting processing is performed on the point cloud distribution data of the entire surface area of ​​the target object to obtain several point cloud distribution curves for the entire surface area of ​​the target object, specifically including:

[0022] The point cloud height value of each point cloud data is obtained from the point cloud distribution data of the entire surface area of ​​the target object; the point cloud distribution data is divided into several point cloud data clusters according to the point cloud height value; wherein the point cloud height values ​​of all point cloud data included in each point cloud data cluster are located within the same height range, and the height ranges corresponding to different point cloud data clusters do not overlap.

[0023] Curve fitting is performed on all point cloud data in each point cloud data cluster to obtain a smooth point cloud distribution curve for each point cloud data cluster.

[0024] Furthermore, in step S4, after adjusting the point distribution of each point cloud distribution curve, constructing a surface 3D model of the target object based on all point cloud distribution curves specifically includes:

[0025] Each point cloud distribution smooth curve is adjusted by removing outliers; then, based on all the adjusted point cloud distribution smooth curves, a surface three-dimensional model of the target object is constructed.

[0026] Compared to existing technologies, this point cloud-based assisted 3D model construction method utilizes a point cloud scanning device to obtain point cloud distribution data of the target object's surface, thereby obtaining corresponding point cloud distribution features. The target object's surface is divided into several point cloud distribution regions. Based on the point cloud distribution features within these regions, the corresponding regions are re-scanned to obtain new point cloud distribution data, which is then used to correct these regions. All point cloud distribution regions are integrated to obtain point cloud distribution data for the entire target object's surface, and curve fitting is performed to obtain several point cloud distribution curves. After adjusting the point distribution of each curve, a 3D model of the target object's surface is constructed based on all the curves. This method uses a point cloud scanning device to acquire point cloud distribution data of the target object's surface, thereby partitioning and correcting the surface, enriching the point cloud data volume in each region, maximizing the inclusion of sufficient point cloud data in each region, achieving 3D construction of the target object's surface, and simultaneously removing abnormal point cloud data to ensure the overall reliability of the point cloud data, thus improving the accuracy and efficiency of 3D model construction.

[0027] Other features and advantages of the invention will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures particularly pointed out in the written description, claims, and drawings.

[0028] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0029] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0030] Figure 1 This is a flowchart illustrating the point cloud-based assisted 3D model construction method provided by the present invention. Detailed Implementation

[0031] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0032] See Figure 1 This is a flowchart illustrating the point cloud-based assisted 3D model construction method provided in an embodiment of the present invention. The point cloud-based assisted 3D model construction method includes the following steps:

[0033] Step S1: Scan the target object using a point cloud scanning device to obtain point cloud distribution data on the surface of the target object; analyze and process the point cloud distribution data to obtain the point cloud distribution characteristics on the surface of the target object; and then, based on the point cloud distribution characteristics, divide the surface of the target object into several point cloud distribution regions.

[0034] Step S2: Based on the point cloud distribution characteristics of the point cloud distribution area, the corresponding area of ​​the target object surface is re-scanned using the point cloud scanning device; based on the point cloud distribution data obtained from the re-scanning, the corresponding point cloud distribution area is corrected.

[0035] Step S3: Integrate all point cloud distribution areas to obtain point cloud distribution data for the entire surface area of ​​the target object; perform curve fitting on the point cloud distribution data for the entire surface area of ​​the target object to obtain several point cloud distribution curves for the entire surface area of ​​the target object.

[0036] Step S4: After adjusting the point distribution of each point cloud distribution curve, a three-dimensional surface model of the target object is constructed based on all point cloud distribution curves.

[0037] The beneficial effects of the above technical solution are as follows: This point cloud-based assisted 3D model construction method uses a point cloud scanning device to obtain point cloud distribution data of the target object surface, thereby obtaining the corresponding point cloud distribution features; the target object surface is divided into several point cloud distribution regions, and according to the point cloud distribution features, the corresponding regions of the target object surface are re-scanned to obtain new point cloud distribution data, thereby correcting the corresponding point cloud distribution regions; all point cloud distribution regions are integrated to obtain point cloud distribution data for the entire target object surface, and curve fitting is performed to obtain several point cloud distribution curves; after adjusting the point distribution of each point cloud distribution curve, a 3D model of the target object surface is constructed based on all point cloud distribution curves. It uses a point cloud scanning device to obtain point cloud distribution data of the target object surface, thereby partitioning and correcting the object surface, enriching the point cloud data volume of the point cloud distribution regions, maximizing the inclusion of sufficient point cloud data in each point cloud distribution region, realizing the 3D construction of the target object surface, and simultaneously removing abnormal point cloud data to ensure the overall reliability of the point cloud data, thereby improving the accuracy and efficiency of 3D model construction.

[0038] Preferably, in step S1, the target object is scanned using a point cloud scanning device to obtain point cloud distribution data of the target object's surface; the point cloud distribution data is then analyzed and processed to obtain the point cloud distribution features of the target object's surface, specifically including:

[0039] A laser point cloud scanning device is used to perform an all-round scan of the surface of the target object to obtain the three-dimensional coordinates and laser reflection intensity data of several points on the surface of the target object; the three-dimensional coordinates and laser reflection intensity data of all points are transformed and processed to obtain the point cloud distribution data corresponding to all points.

[0040] The distance analysis of adjacent points is performed on the point cloud distribution data to obtain the distribution spacing characteristics of the point cloud data on the surface of the target object.

[0041] The beneficial effects of the above technical solution are as follows: In actual operation, a laser point cloud scanning device is first used to perform a full-range scan of the target object's surface. This involves scanning and irradiating several points on the target object's surface with laser beams, obtaining the laser reflection intensity data at each point. Then, the spatial three-dimensional coordinates and laser reflection intensity of each point are correlated and transformed one-to-one to obtain the point cloud distribution data corresponding to all points. Next, the distance between each point in the point cloud distribution data and all its adjacent points is determined. This distance serves as the point cloud data distribution spacing feature of the target object's surface, allowing for the quantitative calibration of the relative positional relationships between each point in the point cloud distribution data.

[0042] Preferably, in step S1, dividing the surface of the target object into several point cloud distribution regions based on the point cloud distribution characteristics specifically includes:

[0043] Based on the distribution spacing characteristics of the point cloud data, the surface of the target object is divided into several point cloud distribution regions with corresponding average point cloud spacing; wherein, the average point cloud spacing of each point cloud distribution region is different; the average point cloud spacing refers to the average distance between each point cloud contained in the point cloud distribution region and its adjacent other point clouds.

[0044] The beneficial effects of the above technical solution are as follows: By using the point cloud data distribution spacing characteristics as a benchmark, the entire surface of the target object is divided into several point cloud distribution areas with corresponding average point cloud spacing. This division of the target object surface facilitates the subsequent rescanning of areas with insufficient point cloud data, thereby ensuring that each area has enough point cloud data for detailed 3D model construction and facilitating targeted processing of different areas.

[0045] Preferably, in step S2, rescanning the corresponding area of ​​the target object surface using the point cloud scanning device based on the point cloud distribution characteristics of the point cloud distribution area specifically includes:

[0046] If the average point cloud spacing in the point cloud distribution area is less than or equal to the preset spacing threshold, then it is not necessary to rescan the area corresponding to the point cloud distribution area on the surface of the target object.

[0047] If the average point cloud density of the point cloud distribution area is greater than the preset spacing threshold, the boundary line between the target object surface and the corresponding area of ​​the point cloud distribution area is determined; and the boundary line is used as the scanning boundary to instruct the laser point cloud scanning device to rescan the corresponding area of ​​the target object surface; wherein the scanning frequency of the rescan is greater than the scanning frequency of the previous scan.

[0048] The beneficial effects of the above technical solution are as follows: When the average point cloud spacing in the point cloud distribution area is less than or equal to a preset spacing threshold, it indicates that the point cloud data in the distribution area is relatively dense, meaning there is sufficient point cloud data. In this case, it is not necessary to rescan the corresponding area of ​​the target object's surface to ensure sufficient point cloud data for high-precision 3D model construction. When the average point cloud spacing in the point cloud distribution area is greater than the preset spacing threshold, it indicates that the point cloud data in the distribution area is relatively sparse, meaning there is insufficient point cloud data. In this case, it is necessary to rescan the corresponding area of ​​the target object's surface to ensure sufficient point cloud data for high-precision 3D model construction. Setting the rescanning frequency to be greater than the previous scan frequency maximizes the amount of point cloud data during the rescanning process.

[0049] Preferably, in step S2, the correction process for the corresponding point cloud distribution area based on the point cloud distribution data obtained from the rescan specifically includes:

[0050] The re-scanned point cloud distribution data is overlaid into the point cloud distribution data of the previous scan of the corresponding point cloud distribution area to obtain new point cloud distribution data for the corresponding point cloud distribution area, thereby achieving the correction processing of the corresponding point cloud distribution area.

[0051] The beneficial effects of the above technical solution are as follows: In actual operation, the point cloud distribution data obtained by rescanning is fused with the point cloud distribution data obtained by the previous scan in an overlay manner, which can greatly increase the number of point cloud data in the corresponding point cloud distribution area and ensure the sufficiency of the number of point cloud data in the point cloud distribution area.

[0052] Preferably, in step S3, the integration processing of all point cloud distribution areas to obtain point cloud distribution data for the entire surface area of ​​the target object specifically includes:

[0053] Based on the boundary distribution information of each point cloud distribution area, determine all point cloud distribution areas adjacent to each point cloud distribution area; then, based on the relative positional relationship between all two adjacent point cloud distribution areas, stitch together and integrate all point cloud distribution areas to obtain point cloud distribution data for the entire surface area of ​​the target object.

[0054] The beneficial effects of the above technical solution are as follows: by using the above method, the boundary shape and boundary position of each point cloud distribution area are determined. Then, based on the boundary shape and boundary position, all point cloud distribution areas adjacent to each point cloud distribution area are determined. Furthermore, the relative positional relationship between all two adjacent point cloud distribution areas is determined. This allows for the accurate splicing and integration of all point cloud distribution areas, thereby representing the point cloud distribution of the entire surface area of ​​the target object.

[0055] Preferably, in step S3, the point cloud distribution data of the entire surface area of ​​the target object is subjected to curve fitting processing to obtain several point cloud distribution curves for the entire surface area of ​​the target object, specifically including:

[0056] Obtain the point cloud height value of each point cloud data in the point cloud distribution data of the entire surface area of ​​the target object; based on the point cloud height value, divide the point cloud distribution data into several point cloud data clusters; wherein, the point cloud height values ​​of all point cloud data included in each point cloud data cluster are located within the same height range, and the height ranges corresponding to different point cloud data clusters do not overlap.

[0057] Curve fitting is performed on all point cloud data in each point cloud data cluster to obtain a smooth point cloud distribution curve for each point cloud data cluster.

[0058] The beneficial effects of the above technical solution are as follows: By using the point cloud height value contained in each point cloud data as a benchmark, all point cloud distribution data are divided into clusters. This allows all point cloud data whose point cloud height values ​​are all within the same height range to be grouped into a point cloud data cluster. Thus, each point cloud data cluster corresponds to the three-dimensional shape within the same height range. When curve fitting is performed on all point cloud data in each point cloud data cluster, the smooth curve of the fitted point cloud distribution corresponds to the three-dimensional shape curve within a height range.

[0059] Preferably, in step S4, after adjusting the point distribution of each point cloud distribution curve, constructing a surface 3D model of the target object based on all point cloud distribution curves specifically includes:

[0060] Each point cloud distribution smoothing curve undergoes outlier removal and adjustment; then, based on all the adjusted point cloud distribution smoothing curves, a surface 3D model of the target object is constructed.

[0061] The beneficial effects of the above technical solution are as follows: By performing outlier removal and adjustment processing on each point cloud distribution smooth curve using the above method, points that deviate significantly from the smooth curve can be removed, avoiding the impact of outliers on the accuracy of subsequent surface 3D model construction. Inputting all the adjusted point cloud distribution smooth curves into the corresponding 3D model construction software for integration and conversion to obtain the surface 3D model of the target object is a conventional technique in this field and will not be described in detail here.

[0062] As can be seen from the above embodiments, this point cloud-based assisted 3D model construction method utilizes a point cloud scanning device to obtain point cloud distribution data of the target object surface, thereby obtaining corresponding point cloud distribution features; the target object surface is divided into several point cloud distribution regions, and based on the point cloud distribution features, the corresponding regions of the target object surface are re-scanned to obtain new point cloud distribution data, thereby correcting the corresponding point cloud distribution regions; all point cloud distribution regions are integrated to obtain point cloud distribution data for the entire target object surface, and curve fitting processing is performed to obtain several point cloud distribution curves; after point distribution adjustment processing for each point cloud distribution curve, a 3D model of the target object surface is constructed based on all point cloud distribution curves. It utilizes a point cloud scanning device to obtain point cloud distribution data of the target object surface, thereby partitioning and correcting the object surface, enriching the point cloud data volume of the point cloud distribution regions, maximizing the inclusion of sufficient point cloud data in each point cloud distribution region, realizing the 3D construction of the target object surface, and simultaneously removing abnormal point cloud data to ensure the overall reliability of the point cloud data, thereby improving the accuracy and efficiency of 3D model construction.

[0063] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A point cloud-based assisted 3D model construction method, characterized in that, It includes the following steps: Step S1: Scan the target object using a point cloud scanning device to obtain point cloud distribution data of the target object surface; analyze and process the point cloud distribution data to obtain point cloud distribution features of the target object surface; and then divide the target object surface into several point cloud distribution regions based on the point cloud distribution features. Step S2: Based on the point cloud distribution characteristics of the point cloud distribution area, the corresponding area of ​​the target object surface is re-scanned using the point cloud scanning device; based on the point cloud distribution data obtained from the re-scanning, the corresponding point cloud distribution area is corrected. Step S3: Integrate all point cloud distribution areas to obtain point cloud distribution data for the entire surface area of ​​the target object. Curve fitting is performed on the point cloud distribution data of the entire surface area of ​​the target object to obtain several point cloud distribution curves for the entire surface area of ​​the target object. Step S4: After adjusting the point distribution of each point cloud distribution curve, a three-dimensional surface model of the target object is constructed based on all point cloud distribution curves. Specifically, in step S2, rescanning the corresponding area of ​​the target object surface using the point cloud scanning device based on the point cloud distribution characteristics of the point cloud distribution area includes: If the average point cloud spacing in the point cloud distribution area is less than or equal to a preset spacing threshold, then it is not necessary to rescan the area corresponding to the target object surface and the point cloud distribution area. If the average point cloud density of the point cloud distribution area is greater than a preset spacing threshold, then the boundary line between the target object surface and the corresponding area of ​​the point cloud distribution area is determined; and the boundary line is used as the scanning boundary to instruct the laser point cloud scanning device to rescan the corresponding area of ​​the target object surface; wherein the scanning frequency of the rescanning is greater than the scanning frequency of the previous scan.

2. The point cloud-based assisted 3D model construction method as described in claim 1, characterized in that: In step S1, the target object is scanned using a point cloud scanning device to obtain point cloud distribution data of the target object's surface; the point cloud distribution data is analyzed and processed to obtain the point cloud distribution features of the target object's surface, specifically including: A laser point cloud scanning device is used to perform an all-round scan of the surface of the target object to obtain the three-dimensional coordinates and laser reflection intensity data of several points on the surface of the target object; the three-dimensional coordinates and laser reflection intensity data of all points are transformed and processed to obtain the point cloud distribution data corresponding to all points. The point cloud distribution data is processed by adjacent point distance analysis to obtain the point cloud data distribution spacing characteristics of the target object surface.

3. The point cloud-based assisted 3D model construction method as described in claim 2, characterized in that: In step S1, dividing the surface of the target object into several point cloud distribution regions based on the point cloud distribution characteristics specifically includes: Based on the point cloud data distribution spacing characteristics, the surface of the target object is divided into several point cloud distribution regions with corresponding average point cloud spacing; wherein, the average point cloud spacing of each point cloud distribution region is different; the average point cloud spacing refers to the average distance between each point cloud contained in the point cloud distribution region and its adjacent other point clouds.

4. The point cloud-based assisted 3D model construction method as described in claim 1, characterized in that: In step S2, the correction process for the corresponding point cloud distribution area based on the rescanned point cloud distribution data specifically includes: The point cloud distribution data obtained from the rescan is superimposed onto the point cloud distribution data obtained from the previous scan of the corresponding point cloud distribution area to obtain new point cloud distribution data for the corresponding point cloud distribution area, thereby achieving the correction processing of the corresponding point cloud distribution area.

5. The point cloud-based assisted 3D model construction method as described in claim 4, characterized in that: In step S3, the integration of all point cloud distribution areas to obtain point cloud distribution data for the entire surface area of ​​the target object specifically includes: Based on the boundary distribution information of each point cloud distribution area, determine all point cloud distribution areas adjacent to each point cloud distribution area; then, based on the relative positional relationship between all two adjacent point cloud distribution areas, stitch together and integrate all point cloud distribution areas to obtain point cloud distribution data for the entire surface area of ​​the target object.

6. The point cloud-based assisted 3D model construction method as described in claim 5, characterized in that: In step S3, curve fitting is performed on the point cloud distribution data of the entire surface area of ​​the target object to obtain several point cloud distribution curves for the entire surface area of ​​the target object. Specifically, this includes: The point cloud height value of each point cloud data is obtained from the point cloud distribution data of the entire surface area of ​​the target object; the point cloud distribution data is divided into several point cloud data clusters according to the point cloud height value; wherein the point cloud height values ​​of all point cloud data included in each point cloud data cluster are located within the same height range, and the height ranges corresponding to different point cloud data clusters do not overlap. Curve fitting is performed on all point cloud data in each point cloud data cluster to obtain a smooth point cloud distribution curve for each point cloud data cluster.

7. The point cloud-based assisted 3D model construction method as described in claim 6, characterized in that: In step S4, after adjusting the point distribution of each point cloud distribution curve, constructing a surface 3D model of the target object based on all point cloud distribution curves specifically includes: Each point cloud distribution smooth curve is adjusted by removing outliers; then, based on all the adjusted point cloud distribution smooth curves, a surface three-dimensional model of the target object is constructed.

Images