Unmanned aerial vehicle surveying and mapping data management system based on cloud platform

Abstract

The invention relates to the technical field of three-dimensional modeling, in particular to an unmanned aerial vehicle surveying and mapping data management system based on a cloud platform, and the system comprises an uploading module which is used for uploading unmanned aerial vehicle surveying and mapping data to a corresponding cloud platform, and a calculation module which is used for extracting and matching feature points and forming a three-dimensional sparse point cloud. The filling module is used for filling the three-dimensional sparse point cloud into a three-dimensional dense point cloud; the construction module is used for constructing a three-dimensional grid and generating a three-dimensional live-action model; the management module is used for judging the generation of the three-dimensional live-action model based on the luminosity error and judging whether the generation of the three-dimensional live-action model meets the standard or not based on the reconstruction confidence proportion; and the optimization module is used for adjusting corresponding parameters or sending out corresponding notifications. According to the invention, accurate monitoring and intelligent adjustment of the three-dimensional real model generated according to the surveying and mapping data of the unmanned aerial vehicle are effectively realized, and the efficiency of generating the three-dimensional real model according to the surveying and mapping data of the unmanned aerial vehicle is effectively improved.

Description

Technical Field

[0001] This invention relates to the field of 3D modeling technology, and in particular to a cloud-based UAV mapping data management system. Background Technology

[0002] Unmanned aerial vehicle (UAV) mapping data is a collection of various digitized information related to geospatial data collected by UAVs equipped with mapping sensors during airborne flight. Uploading UAV data to a cloud platform addresses the limitations of local management, enabling efficient data utilization and value amplification, overcoming data processing speed bottlenecks, and ensuring data security. Generating 3D models from UAV mapping data on a cloud platform transforms fragmented mapping data into a 3D geographic information carrier. Cloud platforms possess massive computing power, capable of rapidly processing large volumes of mapping data, including imagery, overcoming the limitations of insufficient computing power and low modeling efficiency of local devices. The generated 3D models can intuitively recreate geospatial morphology, supporting precise measurement and analysis, and enabling cross-scenario applications. Comparing 3D models across different time periods reveals changes in the geographical environment or project progress, providing data support for decision-making. With the development of technology, research on generating 3D models from UAV mapping data via cloud platforms has significant practical implications.

[0003] Chinese Patent Publication No. CN114581608A discloses a cloud-based intelligent 3D model construction system and method. The system includes a cloud platform, an image acquisition device, and a local terminal. After acquiring an image, the image acquisition device copies the image to obtain two source images. A synchronous connection is established between these two source images, and one source image is sent to the local terminal, while the other is sent to the cloud platform. The local terminal obtains the 3D mapping result from the cloud image and generates a 3D model in the opposite direction to the cloud-based 3D model generation. Simultaneously, it acquires the 3D model generation result from the cloud in real time and stitches it with the 3D model generation result from the local terminal to obtain the final constructed 3D model. By simultaneously generating a portion of the 3D model on both the cloud and local terminals, the system improves generation efficiency.

[0004] Therefore, the above scheme improves generation efficiency by establishing a synchronous connection between two source images to obtain the 3D model generated in the cloud, stitching it together with the 3D model generated locally, and simultaneously generating the 3D model in both the cloud and local environments. However, this scheme cannot accurately monitor and intelligently adjust the 3D realistic model generated from UAV mapping data, thus failing to guarantee the generation efficiency of the 3D realistic model from UAV mapping data. Summary of the Invention

[0005] To address this issue, the present invention provides a cloud-based UAV mapping data management system to overcome the problem in existing technologies where the generation efficiency of 3D realistic models generated from UAV mapping data is low due to the inability to accurately monitor and intelligently adjust the data.

[0006] To achieve the above objectives, the present invention provides a cloud platform-based UAV mapping data management system, comprising: The upload module is used to upload UAV mapping data to the corresponding cloud platform. The UAV mapping data includes mapping photos and position and attitude data. A calculation module, which is connected to the upload module, is used to batch process the survey photos to extract and match feature points; The calculation module is also used to adjust the position and orientation parameters of the survey photograph and the three-dimensional coordinates of the feature point to converge the rays from related survey photographs pointing to the same feature point at a point in three-dimensional space; The calculation module is also used to form a three-dimensional sparse point cloud based on the survey photograph and the feature points; A filling module, which is connected to the calculation module, is used to fill the three-dimensional sparse point cloud into a three-dimensional dense point cloud based on the survey photo by adjusting the density matching strictness. A construction module, connected to the filling module, is used to triangulate the 3D dense point cloud to construct a 3D mesh and optimize the 3D mesh by adjusting the mesh filtering intensity; the construction module is also used to perform texture mapping on the optimized 3D mesh based on the survey photograph to generate a 3D real-world model. The management module, connected to the construction module, is used to determine the photometric mean square error of the 3D reality model; the management module is also used to make a judgment on the generation of the 3D reality model based on the photometric mean square error; the management module is also used to determine whether the generation of the 3D reality model meets the standard based on the reconstruction confidence ratio, or to generate corresponding processing instructions, or to complete the generation of the 3D reality model and push it. An optimization module, which is connected to the calculation module, the filling module, the construction module, and the management module respectively, is used to adjust the suboptimal matching ratio based on the second-order difference of photometric data, adjust the strictness of dense matching based on the reconstruction confidence ratio difference, increase the mesh filtering intensity based on the increase in the strictness of dense matching, adjust the texture mapping mode, or issue a notification that UAV mapping data needs to be reacquired.

[0007] Furthermore, the management module is used to determine the generation of the 3D real scene model based on the photometric mean square error, and to determine whether the generation of the 3D real scene model meets the standard based on the reconstruction confidence ratio according to the determination result, or to adjust the matching suboptimal ratio based on the photometric second-order difference.

[0008] Furthermore, the management module is also used to determine whether the generation of the 3D real scene model meets the standard based on the reconstruction confidence ratio, and to complete the generation of the 3D real scene model and push it according to the determination result, or to adjust the strictness of dense matching based on the reconstruction confidence ratio difference.

[0009] Furthermore, the optimization module is used to reduce the second-best matching ratio based on the second photometric difference, and the reduction in the second-best matching ratio is proportional to the second photometric difference.

[0010] Furthermore, the management module is also used to determine the generation of the 3D real scene model based on the photometric error after the matching second-best ratio is increased, and to determine whether the generation of the 3D real scene model meets the standard based on the reconstruction confidence ratio according to the determination result, or to issue a notification that the UAV mapping data needs to be reacquired.

[0011] Furthermore, the optimization module is also used to increase the tightness of dense matching based on the reconstruction confidence ratio difference, and the increase in tightness of dense matching is proportional to the reconstruction confidence ratio difference.

[0012] Furthermore, the optimization module is also used to increase the mesh filtering intensity based on the increase in the density matching strictness, and the increase in mesh filtering intensity is proportional to the increase in density matching strictness.

[0013] Furthermore, the management module is also used to determine whether the generation of the 3D real scene model meets the standard based on the reconstruction confidence ratio after the mesh filtering intensity adjustment is completed, and to complete the generation of the 3D real scene model and push it according to the determination result, or to adjust the texture mapping mode.

[0014] Furthermore, the optimization module is also used to adjust the texture mapping mode, and to switch the texture mapping mode to a second texture mapping mode or to a third texture mapping mode according to the determination result.

[0015] Furthermore, the management module is also used to determine whether the generation of the 3D reality model meets the standard based on the reconstruction confidence ratio after the texture mapping mode adjustment is completed, and to generate and push the 3D reality model according to the determination result, or to issue a notification that the UAV mapping data needs to be reacquired. Compared with the prior art, the beneficial effects of the present invention are as follows: By setting up a management module and an optimization module, the management module judges the generation of the 3D real scene model based on the photometric mean square error and determines whether the generation of the 3D real scene model meets the standard or generates the corresponding processing instructions based on the reconstruction confidence ratio. This enables timely and accurate judgment of the 3D real model generated from UAV mapping data, effectively realizing precise monitoring of the 3D real model generated from UAV mapping data. The optimization module adjusts the corresponding parameters or issues a notification that UAV mapping data needs to be reacquired. While effectively realizing intelligent adjustment of the 3D real model generated from UAV mapping data, the generation efficiency of the 3D real model generated from UAV mapping data is effectively improved.

[0016] Furthermore, the management module of this invention is also used to judge the generation of the 3D real scene model based on the photometric mean error. It promptly determines whether the generation of the 3D real scene model meets the standard based on the reconstruction confidence ratio or whether the matching suboptimal ratio needs to be adjusted based on the photometric second difference, so as to avoid misjudgment. While further realizing the accurate monitoring of the 3D real model generated from UAV mapping data, it further improves the generation efficiency of the 3D real model generated from UAV mapping data.

[0017] Furthermore, the management module of this invention is also used to determine whether the generation of the 3D reality model meets the standard based on the reconstruction confidence ratio, promptly determine the situation where the generation of the 3D reality model needs to be completed and pushed, and determine whether the density matching strictness needs to be adjusted based on the reconstruction confidence ratio difference, thus ensuring the accuracy of the determination. While further realizing the accurate monitoring of the 3D real model generated from UAV mapping data, it further improves the generation efficiency of the 3D real model generated from UAV mapping data.

[0018] Furthermore, the optimization module of this invention is used to reduce the second-best matching ratio based on the second-order difference of photometry, which effectively avoids the situation where the generation of the 3D real scene model does not meet the standard due to the second-best matching ratio not meeting the standard. While further realizing the intelligent adjustment of the 3D real model generated from UAV mapping data, it further improves the generation efficiency of the 3D real model generated from UAV mapping data.

[0019] Furthermore, the management module of this invention is also used to determine the generation of the 3D real scene model based on the photometric mean error after the matching suboptimal ratio is increased. It accurately determines whether the generation of the 3D real scene model meets the standard or needs to issue a notification to reacquire UAV mapping data based on the reconstruction confidence ratio, thus avoiding misjudgment. This further realizes the accurate monitoring of the 3D real model generated from UAV mapping data and further improves the generation efficiency of the 3D real model generated from UAV mapping data.

[0020] Furthermore, the optimization module of this invention is also used to increase the strictness of dense matching based on the difference in reconstruction confidence ratio, which effectively avoids the situation where the generation of the 3D real scene model does not meet the standard due to the strictness of dense matching not meeting the standard. While further realizing the intelligent adjustment of the 3D real model generated from UAV mapping data, it further improves the generation efficiency of the 3D real model generated from UAV mapping data.

[0021] Furthermore, the optimization module of this invention is also used to increase the mesh filtering intensity based on the increase in the density matching strictness, which further effectively avoids the situation where the generation of the 3D real scene model does not meet the standard due to the mesh filtering intensity not meeting the standard. While further realizing the intelligent adjustment of the 3D real model generated from UAV mapping data, it further improves the generation efficiency of the 3D real model generated from UAV mapping data.

[0022] Furthermore, the management module of this invention is also used to determine whether the generation of the 3D real scene model meets the standard based on the reconstruction confidence ratio after the mesh filtering intensity adjustment is completed. It can effectively determine whether the generation of the 3D real scene model needs to be completed and pushed or whether the texture mapping mode needs to be adjusted, so as to avoid misjudgment. While further realizing the accurate monitoring of the 3D real model generated from UAV mapping data, it further improves the generation efficiency of the 3D real model generated from UAV mapping data.

[0023] Furthermore, the optimization module of this invention is also used to adjust the texture mapping mode, which effectively avoids the situation where the generation of the 3D real scene model does not meet the standard due to the texture mapping mode not meeting the standard. While further realizing the intelligent adjustment of the 3D real model generated from UAV mapping data, it further improves the generation efficiency of the 3D real model generated from UAV mapping data.

[0024] Furthermore, the management module of this invention is also used to determine whether the generation of the 3D real scene model meets the standard based on the reconstruction confidence ratio after the texture mapping mode adjustment is completed. It can promptly and accurately determine whether the generation of the 3D real scene model needs to be completed and pushed or whether a notification needs to be issued to reacquire UAV mapping data, thus ensuring the accuracy of the judgment and avoiding misjudgment. While further realizing the accurate monitoring of the 3D real model generated from UAV mapping data, it also further improves the generation efficiency of the 3D real model generated from UAV mapping data. Attached Figure Description

[0025] Figure 1 This is a structural block diagram of the UAV mapping data management system based on a cloud platform, according to an embodiment of the present invention. Figure 2 This is a flowchart illustrating the workflow of the cloud-based UAV mapping data management system according to an embodiment of the present invention. Figure 3 This is a flowchart illustrating how to determine whether the generated 3D reality model conforms to the standard and the reasons why the generated 3D reality model does not conform to the standard, according to an embodiment of the present invention. Figure 4 This is a flowchart illustrating the reasons why the generation of a 3D reality model does not meet the standards in an embodiment of the present invention. Detailed Implementation

[0026] To make the objectives and advantages of the present invention clearer, the present invention will be further described below with reference to embodiments; it should be understood that the specific embodiments described herein are merely for explaining the present invention and are not intended to limit the present invention.

[0027] Preferred embodiments of the present invention will now be described with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are merely illustrative of the technical principles of the present invention and are not intended to limit the scope of protection of the present invention.

[0028] Furthermore, it should be noted that, in the description of this invention, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0029] Please see Figure 1 The diagram shown is a structural block diagram of a cloud-based UAV mapping data management system according to an embodiment of the present invention. The cloud-based UAV mapping data management system described in this embodiment includes an upload module, a calculation module, a data filling module, a construction module, a management module, and an optimization module; wherein, The upload module is used to upload UAV mapping data to the corresponding cloud platform. The UAV mapping data includes mapping photos and position and attitude data. The calculation module is connected to the upload module and is used to batch process the survey photos to extract and match feature points; The calculation module is also used to adjust the position and orientation parameters of the survey photograph and the three-dimensional coordinates of the feature point to converge the rays from related survey photographs pointing to the same feature point at a point in three-dimensional space; The calculation module is also used to form a three-dimensional sparse point cloud based on the survey photograph and the feature points; The filling module is connected to the calculation module and is used to fill the three-dimensional sparse point cloud into a three-dimensional dense point cloud based on the survey photo by adjusting the density matching strictness. The construction module is connected to the filling module and is used to triangulate the three-dimensional dense point cloud to construct a three-dimensional mesh and optimize the three-dimensional mesh by adjusting the mesh filtering intensity. The construction module is also used to perform texture mapping on the optimized 3D mesh based on the survey photos to generate a 3D real-scene model. The management module is connected to the construction module and is used to determine the photometric error of the 3D reality model. The management module is also used to make a judgment on the generation of the three-dimensional real scene model based on the photometric mean error; The management module is also used to determine whether the generation of the 3D reality model meets the standard based on the reconstruction confidence ratio. Alternatively, generate corresponding processing instructions. Alternatively, it can generate and push out a 3D reality model; The optimization module is connected to the calculation module, the filling module, the construction module, and the management module, respectively, and is used to adjust the suboptimal matching ratio based on the second-order difference of photometric data, adjust the strictness of dense matching based on the reconstruction confidence ratio difference, increase the mesh filtering intensity based on the increase in the strictness of dense matching, adjust the texture mapping mode, or issue a notification that UAV mapping data needs to be reacquired.

[0030] Please see Figure 2The diagram shows the workflow of the UAV mapping data management system based on a cloud platform according to an embodiment of the present invention. When the UAV mapping data management system based on a cloud platform is running, the upload module uploads the UAV mapping data to the corresponding cloud platform. The calculation module batch processes the mapping photos to extract and match feature points. The calculation module also adjusts the position and attitude parameters of the mapping photos and the three-dimensional coordinates of the feature points to converge rays from related mapping photos pointing to the same feature point at a single point in three-dimensional space. The calculation module also forms the three-dimensional sparse point cloud based on the mapping photos and feature points. The filling module fills the three-dimensional sparse point cloud into the three-dimensional dense point cloud by adjusting the density matching strictness based on the mapping photos. The construction module triangulates the three-dimensional dense point cloud to construct the three-dimensional mesh and optimizes it by adjusting the mesh filtering intensity. The 3D mesh construction module generates the 3D reality model by performing texture mapping on the optimized 3D mesh based on the survey photos. The management module determines the photometric mean square error of the 3D reality model. The management module judges the generation of the 3D reality model based on the photometric mean square error. The management module determines whether the generation of the 3D reality model meets the standard based on the reconstruction confidence ratio, or generates corresponding processing instructions, or completes the generation of the 3D reality model and pushes it out. The optimization module adjusts the suboptimal matching ratio based on the photometric second-order difference, adjusts the dense matching strictness based on the reconstruction confidence ratio difference, increases the mesh filtering intensity based on the increase in dense matching strictness, adjusts the texture mapping mode, or issues a notification that the UAV survey data needs to be reacquired.

[0031] Please see Figure 3 The diagram shows a flowchart illustrating how an embodiment of the present invention determines whether the generated 3D reality model conforms to a standard and the reasons why the generated 3D reality model does not conform to the standard. The management module described in this embodiment of the present invention is used to determine the generation of the 3D reality model based on the photometric error. If the photometric error is less than or equal to the preset photometric error G set in the management module, the management module determines whether the generation of the three-dimensional real scene model meets the standard based on the reconstruction confidence ratio. In this embodiment, the preset photometric error G = 0.8 pixels. If the photometric error is greater than the preset photometric error G, the management module determines that the generation of the 3D real scene model does not meet the standard, and adjusts the second-best matching ratio based on the photometric second difference. Specifically, the photometric error values ​​of several points in the three-dimensional dense point cloud are determined, the average of the squares of each photometric error value is calculated, the square root of the average value is taken, and the obtained square root is recorded as the photometric error. The total number of triangular patches formed after triangulation is counted. For each triangular patch, the reconstruction confidence level is determined. The number of triangular patches with a reconstruction confidence level greater than or equal to the preset reconstruction confidence level A is counted. The ratio of the number of triangular patches to the total number of triangular patches is calculated. The obtained ratio is recorded as the reconstruction confidence ratio. The preset reconstruction confidence level A = 0.9. In this embodiment, the preset photometric error is set to 0.8 pixels, which is an empirical threshold. The second photometric difference is the difference between the photometric mean error and the preset photometric mean error.

[0032] Please continue reading. Figure 3 As shown, the management module of this invention is also used to determine whether the generation of the 3D reality model meets the standard based on the reconstruction confidence ratio: If the reconstruction confidence ratio is greater than or equal to the preset reconstruction confidence ratio R set in the management module, the management module determines that the generation of the 3D reality model meets the standard, completes the generation of the 3D reality model and pushes it out. In this embodiment, the preset reconstruction confidence ratio R = 95%; If the reconstruction confidence ratio is less than the preset reconstruction confidence ratio R, the management module determines that the generation of the 3D reality model does not meet the standard, and adjusts the density matching strictness based on the reconstruction confidence ratio difference. Specifically, in this embodiment, the preset reconstruction confidence ratio is set to 95%, where 95% is an empirical threshold. The reconstruction confidence ratio difference is the difference between the preset reconstruction confidence ratio and the reconstruction confidence ratio.

[0033] Please continue reading. Figure 3 As shown, the optimization module described in this embodiment of the invention is used to reduce the suboptimal matching ratio based on the second-order photometric difference: If the photometric second-order difference is greater than the second preset photometric second-order difference △K2 set in the management module, the optimization module will reduce the second-best matching ratio to 0.800 times the initial second-best matching ratio, wherein, in this embodiment, the second preset photometric second-order difference △K2 = 0.8 pixels; If the second-order photometric difference is less than or equal to the second preset second-order photometric difference △K2 and greater than the first preset second-order photometric difference △K1 set in the management module, the optimization module reduces the second-best matching ratio to 0.853 times the initial second-best matching ratio. In this embodiment, the first preset second-order photometric difference △K1 = 0.4 pixels. If the photometric second-order difference is less than or equal to the first preset photometric second-order difference ΔK1, the optimization module reduces the second-best matching ratio to 0.907 times the initial second-best matching ratio; Specifically, as mentioned above, the preset photometric error is 0.8 pixels, and the value range of the photometric second-order difference is greater than 0 pixels and less than 1.2 pixels. If the photometric second-order difference is greater than or equal to 1.2 pixels, i.e., the photometric error is greater than 2.0 pixels, a notification is issued that the UAV mapping data needs to be reacquired. The larger the photometric second-order difference, the greater the reduction factor of the matching second-best ratio. The values ​​of the photometric second-order difference and the matching second-best ratio are both derived from the actual debugging results.

[0034] Please continue reading. Figure 3 As shown, the management module of this invention is used to determine the generation of the 3D real-scene model based on the photometric error after the second-best matching ratio has increased: If the photometric error is less than or equal to the preset photometric error G, the management module determines whether the generation of the 3D real scene model meets the standard based on the reconstruction confidence ratio. If the photometric error is greater than the preset photometric error G, the management module determines that the generation of the 3D real-scene model does not meet the standard and issues a notification that the UAV mapping data needs to be reacquired.

[0035] Please continue reading. Figure 3 As shown, the optimization module of this invention is further used to increase the strictness of the dense matching based on the reconstruction confidence ratio difference: If the reconstruction confidence ratio difference is greater than the second preset reconstruction confidence ratio difference △N2 set in the management module, the optimization module increases the dense matching strictness to 1.29 times the initial dense matching strictness, wherein, in this embodiment, the second preset reconstruction confidence ratio difference △N2 = 10%; If the reconstruction confidence ratio difference is less than or equal to the second preset reconstruction confidence ratio difference △N2 and greater than the first preset reconstruction confidence ratio difference △N1 set in the management module, the optimization module increases the dense matching strictness to 1.21 times the initial dense matching strictness, wherein, in this embodiment, the first preset reconstruction confidence ratio difference △N1 = 5%; If the reconstruction confidence ratio difference is less than or equal to the first preset reconstruction confidence ratio difference △N1, the optimization module increases the dense matching strictness to 1.14 times the initial dense matching strictness; Specifically, as mentioned above, the preset reconstruction confidence ratio is 95%, and the reconstruction confidence ratio difference is greater than 0% and less than 15%. If the reconstruction confidence ratio difference is greater than or equal to 15%, i.e., the reconstruction confidence ratio is less than or equal to 80%, a notification is issued that the UAV mapping data needs to be reacquired. The larger the reconstruction confidence ratio difference, the greater the increase in the density matching strictness. The values ​​of the reconstruction confidence ratio difference and the density matching strictness are both derived from the actual debugging results.

[0036] Please continue reading. Figure 3 As shown, the optimization module in this embodiment of the invention is further used to increase the mesh filtering intensity based on the increase in the density matching strictness: If the dense matching strictness is increased to 1.29 times the initial dense matching strictness, the optimization module increases the mesh filtering strength to 1.8 times the initial mesh filtering strength; If the dense matching strictness is increased to 1.21 times the initial dense matching strictness, the optimization module increases the mesh filtering strength to 1.6 times the initial mesh filtering strength; If the dense matching strictness is increased to 1.14 times the initial dense matching strictness, the optimization module will increase the mesh filtering strength to 1.4 times the initial mesh filtering strength. Specifically, the value of the increase factor for the mesh filtering intensity is derived from the actual debugging results.

[0037] Please see Figure 4 The diagram shows a flowchart illustrating the reasons why the generated 3D reality model does not meet the standards in an embodiment of the present invention. The management module in this embodiment is further configured to determine whether the generated 3D reality model meets the standards based on the reconstruction confidence ratio after the mesh filtering intensity adjustment is completed. If the reconstruction confidence ratio is greater than or equal to the preset reconstruction confidence ratio R, the management module determines that the generation of the 3D reality model meets the standard, completes the generation of the 3D reality model, and pushes it. If the reconstruction confidence ratio is less than the preset reconstruction confidence ratio R, the management module determines that the generation of the 3D reality model does not meet the standard and adjusts the texture mapping mode.

[0038] Please continue reading. Figure 4 As shown, the optimization module in this embodiment of the invention is further used to adjust the texture mapping mode: If the texture mapping mode is the first texture mapping mode, the optimization module will switch the texture mapping mode to the second texture mapping mode; If the texture mapping mode is the second texture mapping mode, the optimization module will switch the texture mapping mode to the third texture mapping mode; Specifically, in this embodiment, the first texture mapping mode is global optimization, the second texture mapping mode is multi-view blending, and the third texture mapping mode is view-dependent; Additionally, if the texture mapping mode is the third texture mapping mode, a notification will be issued that the UAV mapping data needs to be reacquired.

[0039] Please continue reading. Figure 4As shown in the embodiment of the present invention, the management module is further used to determine whether the generation of the 3D real-scene model meets the standard based on the reconstruction confidence ratio after the texture mapping mode adjustment is completed: If the reconstruction confidence ratio is greater than or equal to the preset reconstruction confidence ratio R, the management module determines that the generation of the 3D reality model meets the standard, completes the generation of the 3D reality model, and pushes it. If the reconstruction confidence ratio is less than the preset reconstruction confidence ratio R, the management module determines that the generation of the 3D reality model does not meet the standard and issues a notification that the UAV mapping data needs to be reacquired.

[0040] The technical solution of the present invention has been described above with reference to the preferred embodiments shown in the accompanying drawings. However, it will be readily understood by those skilled in the art that the scope of protection of the present invention is obviously not limited to these specific embodiments. Without departing from the principles of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will all fall within the scope of protection of the present invention.

[0041] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A cloud-based UAV mapping data management system, characterized in that, include: The upload module is used to upload UAV mapping data to the corresponding cloud platform. The UAV mapping data includes mapping photos and position and attitude data. A calculation module, which is connected to the upload module, is used to batch process the survey photos to extract and match feature points; The calculation module is also used to adjust the position and orientation parameters of the survey photograph and the three-dimensional coordinates of the feature point to converge the rays from related survey photographs pointing to the same feature point at a point in three-dimensional space; The calculation module is also used to form a three-dimensional sparse point cloud based on the survey photograph and the feature points; A filling module, which is connected to the calculation module, is used to fill the three-dimensional sparse point cloud into a three-dimensional dense point cloud based on the survey photo by adjusting the density matching strictness. A construction module, connected to the filling module, is used to triangulate the 3D dense point cloud to construct a 3D mesh and optimize the 3D mesh by adjusting the mesh filtering intensity; the construction module is also used to perform texture mapping on the optimized 3D mesh based on the survey photograph to generate a 3D real-world model. The management module, connected to the construction module, is used to determine the photometric mean square error of the 3D reality model; the management module is also used to make a judgment on the generation of the 3D reality model based on the photometric mean square error; the management module is also used to determine whether the generation of the 3D reality model meets the standard based on the reconstruction confidence ratio, or to generate corresponding processing instructions, or to complete the generation of the 3D reality model and push it. An optimization module, which is connected to the calculation module, the filling module, the construction module, and the management module respectively, is used to adjust the suboptimal matching ratio based on the second-order difference of photometric data, adjust the strictness of dense matching based on the reconstruction confidence ratio difference, increase the mesh filtering intensity based on the increase in the strictness of dense matching, adjust the texture mapping mode, or issue a notification that UAV mapping data needs to be reacquired.

2. The cloud-based UAV mapping data management system according to claim 1, characterized in that, The management module is used to determine the generation of the 3D real scene model based on the photometric mean error, and to determine whether the generation of the 3D real scene model meets the standard based on the reconstruction confidence ratio according to the determination result, or to adjust the matching suboptimal ratio based on the photometric second difference.

3. The cloud-based UAV mapping data management system according to claim 2, characterized in that, The management module is also used to determine whether the generation of the 3D real scene model meets the standard based on the reconstruction confidence ratio, and to complete the generation of the 3D real scene model and push it according to the determination result, or to adjust the strictness of dense matching based on the reconstruction confidence ratio difference.

4. The cloud-based UAV mapping data management system according to claim 3, characterized in that, The optimization module is used to reduce the second-best matching ratio based on the second photometric difference, and the reduction in the second-best matching ratio is proportional to the second photometric difference.

5. The cloud-based UAV mapping data management system according to claim 4, characterized in that, The management module is also used to determine the generation of the 3D real scene model based on the photometric error after the matching second-best ratio is increased, and to determine whether the generation of the 3D real scene model meets the standard based on the reconstruction confidence ratio according to the determination result, or to issue a notification that the UAV mapping data needs to be reacquired.

6. The cloud-based UAV mapping data management system according to claim 3, characterized in that, The optimization module is also used to increase the tightness of dense matching based on the reconstruction confidence ratio difference, and the increase in tightness of dense matching is proportional to the reconstruction confidence ratio difference.

7. The cloud-based UAV mapping data management system according to claim 6, characterized in that, The optimization module is also used to increase the mesh filtering intensity based on the increase in the density matching strictness, and the increase in mesh filtering intensity is proportional to the increase in density matching strictness.

8. The cloud-based UAV mapping data management system according to claim 7, characterized in that, The management module is also used to determine whether the generation of the 3D real scene model meets the standard based on the reconstruction confidence ratio after the mesh filtering intensity adjustment is completed, and to complete the generation of the 3D real scene model and push it according to the determination result, or to adjust the texture mapping mode.

9. The cloud-based UAV mapping data management system according to claim 8, characterized in that, The optimization module is also used to adjust the texture mapping mode, and to switch the texture mapping mode to a second texture mapping mode or to a third texture mapping mode according to the determination result.

10. The cloud-based UAV mapping data management system according to claim 9, characterized in that, The management module is also used to determine whether the generation of the 3D real scene model meets the standard based on the reconstruction confidence ratio after the texture mapping mode adjustment is completed, and to complete the generation of the 3D real scene model and push it according to the determination result, or to issue a notification that the UAV mapping data needs to be reacquired.

Images