Water conservancy engineering earthwork measurement method and system based on ai

Abstract

The application discloses an AI-based water conservancy engineering earthwork measurement method and system, and the method comprises the following steps: setting a scanning path according to the topographic features and position range of a to-be-measured region, starting a UAV to carry a laser radar device to scan the to-be-measured region, and obtaining original point cloud data; setting a plurality of feature regions according to the topographic features of the to-be-measured region, and scanning each feature region to obtain local point cloud data; converting and constructing the original point cloud data and the plurality of local point cloud data into the same coordinate system, complementing the local point cloud data to the original point cloud data according to the position of each feature region in the to-be-measured region to obtain overall point cloud data of the to-be-measured region, and then constructing a point cloud model according to the overall point cloud data of the to-be-measured region; and calculating the earthwork amount required by the to-be-measured region according to the point cloud model of the to-be-measured region and the fill-dig standard of the to-be-measured region. The application can solve the problems of complex operation and high operation threshold in the existing earthwork measurement.

Description

Technical Field

[0001] This invention relates to water conservancy engineering, and more particularly to an AI-based method and system for measuring earthwork in water conservancy projects. Background Technology

[0002] For water conservancy projects, accurately calculating project costs and improving project quality is crucial, and efficiently and accurately calculating the volume of earthwork and cutaway is a vital process. Earthwork surveying is an important component of surveying engineering, and its results directly impact project costs and progress. Currently, earthwork surveying in water conservancy projects typically uses drones based on oblique photogrammetry to measure the elevation information of the cutaway and fill areas, and then estimates the earthwork volume based on cutaway and fill standards. However, this elevation-based surveying method is labor-intensive, complex to operate, inconvenient for engineers, and inefficient. Summary of the Invention

[0003] In order to overcome the shortcomings of the existing technology, one of the objectives of this invention is to provide an AI-based earthwork measurement method for water conservancy projects, which can solve the problems of complex earthwork measurement operations and high operating thresholds in the existing technology.

[0004] The second objective of this invention is to provide an AI-based earthwork measurement system for water conservancy projects, which can solve the problems of complex earthwork measurement operations and high operational thresholds in the existing technology.

[0005] One of the objectives of this invention is achieved through the following technical solution:

[0006] An AI-based method for measuring earthwork in water conservancy projects, comprising:

[0007] Overall scanning steps: Set a scanning path based on the terrain features and location range of the area to be tested, and then scan the area to be tested by launching a drone carrying a lidar device according to the scanning path to obtain the raw point cloud data of the area to be tested;

[0008] Local scanning steps: Based on the terrain features of the area to be measured, multiple feature regions are set up, and point cloud scanning is performed on each feature region to obtain local point cloud data for each feature region;

[0009] Point cloud fusion steps: After converting the original point cloud data of the area to be tested and the local point cloud data of each feature area into the same coordinate system, the local point cloud data of each feature area is supplemented into the original point cloud data of the area to be tested according to the coordinate position of each feature area in the area to be tested to obtain the overall point cloud data of the area to be tested, and then a point cloud model is constructed based on the overall point cloud data of the area to be tested.

[0010] Calculation steps: Calculate the required earthwork volume for the area to be measured based on the point cloud model of the area to be measured and the cut and fill standards of the area to be measured.

[0011] Further, the point cloud fusion step includes: acquiring a 3D image of the area to be tested using an UAV based on oblique photogrammetry technology, dividing the area into grids based on the 3D image, and combining it with a GIS map to obtain a 3D grid map of the area to be tested; matching the original point cloud data of the area to be tested with the 3D grid map of the area to be tested to obtain a point cloud set within each grid, and combining it with the grid corresponding to each feature region to obtain the point cloud data of each feature region; deleting the point cloud data of each feature region from the original point cloud data of the area to be tested, and then supplementing the local point cloud data of each feature region into the original point cloud data of the area to be tested to obtain the point cloud data of the area to be tested.

[0012] Furthermore, the calculation steps specifically include: obtaining a standard model of the area to be measured based on the coordinate position range of the area to be measured and the design parameters of the area to be measured; matching the standard model of the area to be measured with the point cloud model to obtain a difference model; and then obtaining the fill area, the cut area, the fill volume of the fill area, and the cut volume of the cut area based on the difference model.

[0013] It also includes: report generation steps: based on the calculation steps, the fill area, cut area, and the earthwork volume of the fill area and the earthwork volume of the cut area are obtained from the area to be tested. The GIS map of the area to be tested generates a distribution map of the fill and cut areas of the area to be tested and distinguishes the fill area and cut area with different colors in the distribution map of the fill and cut areas.

[0014] Furthermore, the feature region includes at least one of the following regions: vegetation cover region, building cover region, water body cover region, complex slope region; the local scanning step specifically includes obtaining local point cloud data of each feature region by setting up a corresponding scanning device in each feature region; the scanning device is a total station or a handheld mobile scanner.

[0015] Furthermore, the point cloud fusion step also includes: preprocessing the original point cloud data of the area to be tested and the local point cloud data of each feature region; the preprocessing includes removing invalid point clouds, abnormal point clouds, and point clouds that do not belong to the area to be tested.

[0016] Furthermore, it also includes: a dynamic demonstration step: by periodically acquiring the completed filling and excavation area of ​​the area to be tested and combining it with the 3D image of the area to be tested to generate a 3D animation model, which is then displayed to supervisors through a background display terminal and a mobile terminal.

[0017] Furthermore, the overall scanning step of setting the scanning path based on the terrain features and location range of the area to be tested specifically includes: acquiring multiple images of the area to be tested through a network or camera, extracting the terrain features and location range of the area to be tested from the multiple images of the area to be tested based on image processing technology, simulating the area to be tested based on the terrain features and location range of the area to be tested to obtain a two-dimensional planar image of the area to be tested, then designing multiple different scanning paths for the two-dimensional planar image of the area to be tested using an AI large model, simulating point cloud scanning for each scanning path, and finally selecting the optimal scanning path as the final scanning path based on the obtained point cloud data results.

[0018] Furthermore, it also includes: a task generation step: based on the coordinate range of the area to be measured, the required earthwork volume of the area to be measured, the number of excavation and filling vehicles available in the system, and combined with big data mining and models, the number of days required for excavation and filling is determined; and based on the start date of excavation and filling, the required number of days for excavation and filling, and future weather forecasts, the end date of excavation and filling and filling schedule is determined; and a daily excavation and filling task schedule is allocated based on the excavation and filling schedule, the start date of excavation and filling, the end date of excavation and filling, and the number of excavation and filling vehicles dispatched each day, and the daily excavation and filling task schedule is pushed to the vehicle dispatch center, which then dispatches the excavation and filling vehicles; the daily excavation and filling task schedule includes multiple excavation and filling tasks, each of which includes the excavation and filling coordinate range and the earthwork volume of excavation and filling; when the daily excavation and filling task schedule is completed, the required earthwork volume of the area to be measured and the start date of excavation and filling are updated, and the process returns to the task generation step to re-estimate the required number of days for excavation and filling and generate a new excavation and filling schedule; until the required earthwork volume of the area to be measured is excavated and filled.

[0019] Based on the coordinate range of the area to be measured and the required earthwork volume, combined with big data mining and modeling, the number of days required for filling and excavation is determined. Specifically, this involves: firstly, collecting data on the earthwork volume, number of days, and vehicles used for filling and excavation of similar historical water conservancy projects, and then using model training to derive a correlation model between the number of days, vehicles, and earthwork volume; and then estimating the number of days based on the correlation model and the coordinate range of the area to be measured, the required earthwork volume, and the vehicles used for filling and excavation.

[0020] Task detection steps: When each cutting and filling vehicle completes its corresponding cutting and filling task, it reports the completion time of the cutting and filling task and judges whether there are any abnormalities in the work of each cutting and filling vehicle based on the completion time of the cutting and filling task of each vehicle, so as to maintain the cutting and filling vehicles with abnormalities in a timely manner and dispatch new cutting and filling vehicles to work.

[0021] The second objective of this invention is achieved by the following technical solution:

[0022] An AI-based earthwork measurement system for water conservancy projects includes:

[0023] The overall scanning module is used to set a scanning path according to the terrain features and location range of the area to be tested, and then scan the area to be tested by launching a drone carrying a lidar device according to the scanning path to obtain the original point cloud data of the area to be tested.

[0024] The local scanning module is used to set multiple feature regions according to the terrain features of the area to be measured and to perform point cloud scanning on each feature region to obtain local point cloud data of each feature region.

[0025] The point cloud fusion module is used to convert the original point cloud data of the area to be tested and the local point cloud data of each feature area into the same coordinate system, and then fill in the local point cloud data of each feature area into the original point cloud data of the area to be tested according to the coordinate position of each feature area in the area to be tested to obtain the overall point cloud data of the area to be tested, and then construct a point cloud model based on the overall point cloud data of the area to be tested.

[0026] The calculation module is used to calculate the required earthwork volume for the area to be measured based on the point cloud model of the area to be measured and the cut and fill standards of the area to be measured.

[0027] Furthermore, the point cloud fusion module is also used to acquire 3D images of the area to be tested using oblique photography technology via a UAV, divide the area into grids based on the 3D images, and combine them with a GIS map to obtain a 3D grid map of the area to be tested; match the original point cloud data of the area to be tested with the 3D grid map of the area to be tested to obtain the point cloud set within each grid, and combine the grid corresponding to each feature region to obtain the point cloud data of each feature region; after deleting the point cloud data of each feature region from the original point cloud data of the area to be tested, the local point cloud data of each feature region is added to the original point cloud data of the area to be tested to obtain the point cloud data of the area to be tested.

[0028] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0029] This invention utilizes a drone to perform a large-scale scan of the cut and fill area, generating point cloud data. It then combines this with local area scanning to compensate for point cloud data in complex areas within the large-scale scan, resulting in a more complete point cloud model of the area to be measured. Based on the cut and fill standards, the required earthwork volume for the measured area is then estimated. This invention uses point clouds to estimate the earthwork volume, avoiding complex calculation methods, improving system performance, and simplifying operation. Attached Figure Description

[0030] Figure 1 The flowchart of the AI-based earthwork measurement method for water conservancy projects provided by this invention is shown. Detailed Implementation

[0031] The present invention will now be further described in conjunction with the accompanying drawings and specific embodiments. It should be noted that, without conflict, the various embodiments or technical features described below can be arbitrarily combined to form new embodiments.

[0032] Example 1

[0033] Addressing the shortcomings of existing methods for measuring earthwork volume in water conservancy projects, this invention utilizes unmanned aerial vehicles (UAVs) to scan the point cloud of the area to be measured. By combining this UAV-based method with local point cloud data from complex areas to construct a point cloud model of the entire region, the earthwork volume can be measured, improving operational convenience and efficiency. Specifically, this invention provides a preferred embodiment, such as... Figure 1 As shown, an AI-based earthwork measurement method for water conservancy projects includes:

[0034] Step S1: Set a scanning path based on the terrain features and location range of the area to be measured, and then scan the area to be measured by starting a drone carrying a lidar device according to the scanning path to obtain the original point cloud data of the area to be measured.

[0035] This invention utilizes a lidar device carried by a drone to scan the point cloud data of the area to be measured, thereby scanning the cut and fill area to obtain a point cloud model of the cut and fill area, and calculating the required earthwork volume of the cut and fill area.

[0036] When scanning the area to be tested, it is also necessary to set the scanning path of the UAV according to the size and location range of the area to be tested, so as to scan the area to be tested as comprehensively as possible, so as to obtain every position of the area to be tested, thereby ensuring the integrity of the scanning data.

[0037] Step S2: Set up multiple feature regions based on the terrain features of the area to be measured, and perform point cloud scanning on each feature region to obtain local point cloud data for each feature region.

[0038] Specifically, the characteristic region refers to some complex areas within the area to be measured, such as vegetation-covered areas, water-covered areas, complex slope areas, building-covered areas, and areas covered by piles of materials. When a drone carrying a lidar device scans these areas, there may be issues with the accuracy of the point cloud data. Therefore, this invention also uses corresponding scanning devices set up in these specific areas to scan them and obtain local point cloud data. This local point cloud data is then fused with the original point cloud data obtained by the drone to obtain the overall point cloud data of the area to be measured. This local scanning method compensates for the point cloud data obtained after the overall scan, avoiding the loss of point cloud data due to obstructions, which could lead to errors in subsequent earthwork calculations. Specifically, the scanning device can be a total station or a handheld mobile scanner, and can also be used in conjunction with RTK devices to scan data from local characteristic regions.

[0039] More specifically, the feature region can be appropriately set according to the size of the covering and the scanning range of the scanning device. In addition, the range of the feature region can be larger than the size of the covering, so that there are more points between the local point cloud data and the original point cloud data, ensuring the accuracy of subsequent point cloud data fusion.

[0040] Step S3: After converting the original point cloud data of the area to be tested and the local point cloud data of each feature area into the same coordinate system, the local point cloud data of each feature area is supplemented into the original point cloud data of the area to be tested according to the coordinate position of each feature area in the area to be tested to obtain the overall point cloud data of the area to be tested, and then a point cloud model is constructed based on the overall point cloud data of the area to be tested.

[0041] Specifically, the local point cloud data of each feature region is matched with the original point cloud data of the region to be tested to complete the missing point cloud data in the original point cloud data of the region to be tested, so as to obtain the overall point cloud data of the region to be tested, and then construct a point cloud model. The ICP algorithm can be used to match the point cloud data. At the same time, before matching the point cloud data, it is necessary to transform the point cloud data into the same coordinate system. This can be done by setting reference standards in the area to be tested on-site to calibrate each scanning device, thereby facilitating the matching of point cloud data.

[0042] Preferably, when fusing point clouds in step S3, the method further includes: first, acquiring a 3D image of the area to be tested using an UAV based on oblique photography technology, dividing the area into grids based on the 3D image of the area to be tested, and combining the GIS map to obtain a 3D grid map of the area to be tested.

[0043] Secondly, the original point cloud data of the area to be tested is matched with the 3D grid map of the area to be tested to obtain the point cloud set in each grid. The point cloud data of each feature region is obtained by combining the grid corresponding to each feature region. After deleting the point cloud data of each feature region from the original point cloud data of the area to be tested, the local point cloud data of each feature region is added to the original point cloud data of the area to be tested to obtain the point cloud data of the area to be tested.

[0044] This invention addresses the problem of inaccurate subsequent calculations caused by missing point cloud data in the original point cloud data of the entire test area due to obstacles or other objects. It improves the accuracy of subsequent model construction by supplementing the entire point cloud data with local point cloud data.

[0045] More preferably, the original point cloud data of the area to be measured and the local point cloud data of each feature region are preprocessed. Preprocessing includes removing invalid point clouds, abnormal point clouds, and point clouds that do not belong to the area to be measured. This preprocessing method ensures the accuracy of the scanned point cloud data.

[0046] Step S4: Calculate the required earthwork volume for the area to be measured based on the point cloud model of the area to be measured and the cut and fill standards of the area to be measured.

[0047] Specifically, this embodiment is applied to earthwork excavation and filling over a large area. First, a drone carrying a lidar device is used to scan a large area to be measured. Then, local scanning is performed on some specific areas, followed by point cloud fusion to obtain point cloud data for the entire area to be measured and construct a point cloud model. Finally, the required earthwork volume is calculated by combining the excavation and filling standards.

[0048] In this embodiment, the cut and fill standard generally refers to the cut and fill depth specified in the engineering project. A standard model is constructed based on the cut and fill depth standard specified in the engineering project. The cut and fill depth is derived from the design parameters of the area to be measured. That is, step S4 also includes: deriving the standard model of the area to be measured based on the coordinate position range and the design parameters of the area to be measured, matching the standard model of the area to be measured with the point cloud model to obtain a difference model, and then deriving the fill area, the cut area, the fill volume of the fill area, and the cut volume of the cut area based on the difference model.

[0049] More specifically, step S4 also includes: generating a cut-fill area distribution map of the test area based on the calculated fill and cut areas of the test area, as well as the fill volume of the fill area and the cut volume of the cut area; and distinguishing the fill and cut areas in the cut-fill area distribution map using different colors. Displaying the cut-fill areas as a GIS map is more convenient for users to view.

[0050] Furthermore, the present invention can also generate cut and fill tasks; that is, the present invention also includes:

[0051] Task generation steps: Based on the coordinate range of the area to be measured, the required earthwork volume of the area to be measured, the number of excavation and filling vehicles available in the system, and combined with big data mining and models, the number of days required for excavation and filling is determined. Based on the start date of excavation and filling, the number of days required for excavation and filling, and future weather forecasts, the end date of excavation and filling and filling schedule is determined.

[0052] Specifically, the task generation steps also include: allocating a daily cut and fill task plan based on the cut and fill schedule, the start date of cut and fill, the end date of cut and fill, and the number of cut and fill vehicles dispatched each day; and pushing the daily cut and fill task plan to the vehicle dispatch center, which then dispatches the cut and fill vehicles. The daily cut and fill task plan includes multiple cut and fill tasks, each including the cut and fill coordinate range and the volume of earth and rock excavated.

[0053] Based on the coordinate range of the area to be measured and the required earthwork volume, the number of days required for filling and excavation is determined using big data mining and AI models. Specifically, this involves: firstly, collecting data on the earthwork volume, number of days, and vehicles used in historical water conservancy projects of the same type for excavation; and then training an AI model to establish a correlation model between these factors. Based on this correlation model and the coordinate range of the area to be measured, the required earthwork volume, and the number of vehicles used, the number of days required for filling and excavation is estimated. This invention utilizes big data mining and AI models to estimate the number of days required for filling and excavation, improving the accuracy of the estimate. Simultaneously, when the daily filling and excavation task plan is completed, the required earthwork volume and the start date of filling and excavation for the area to be measured are updated, and the process returns to the task generation step to re-estimate the number of days required for filling and excavation and generate a new filling and excavation plan; this continues until the required earthwork volume for the area to be measured is completed. Real-time feedback on the completion of filling and excavation tasks allows for timely updates to the new filling and excavation plan, ensuring timely updates to the plan.

[0054] Furthermore, the present invention also includes a task detection step: when each excavation and filling vehicle completes its corresponding excavation and filling task, it reports the completion time of the task and determines whether there are any abnormalities in the work of each excavation and filling vehicle based on the completion time of its task. This allows for timely maintenance of excavation and filling vehicles with abnormalities and the dispatch of new vehicles to perform the work. By monitoring the excavation and filling vehicles in real time, abnormal vehicles can be detected in a timely manner, so that new vehicles can be dispatched to perform the work, thereby improving the efficiency of excavation and filling. Of course, the drivers of the excavation and filling vehicles can also be equipped with corresponding mobile terminals, allowing drivers to upload the completion status of the excavation and filling tasks in a timely manner and report any abnormalities in the excavation and filling vehicles, ensuring timely communication with the system backend and making the task completion efficiency even higher.

[0055] Similarly, this invention also equips operators with corresponding mobile terminals so that on-site operators can view the progress of excavation and filling in a timely manner. Specifically, the system periodically acquires the completed excavation and filling areas of the test area and combines them with 3D images of the test area to dynamically demonstrate and generate a 3D animation model. This 3D animation model is then displayed on the mobile terminal, allowing engineers to understand the excavation and filling progress in a timely manner. It is also displayed on the backend display terminal so that backend supervisors can understand the excavation and filling progress in a timely manner and dispatch excavation and filling vehicles and issue excavation and filling warnings based on the progress. Displaying the 3D animation to staff allows them to more intuitively see the completion amount of the corresponding area, providing data support for subsequent decision-making. In addition, this invention also includes an acceptance step: by accepting the completed area and generating an acceptance sub-report, the acceptance sub-report is pushed to the acceptance department to realize the acceptance of the completed excavation and filling area. Specifically, the acceptance sub-report includes information such as the completed excavation and filling area, the volume of earth and rock excavated, the start and end times of the excavation and filling, the time taken to complete the excavation and filling, the person in charge of the excavation and filling and the excavation vehicles, and images of the completed excavation and filling.

[0056] Example 2

[0057] Based on Embodiment 1, the present invention also provides an AI-based earthwork measurement system for water conservancy projects, comprising:

[0058] The overall scanning module is used to set the scanning path according to the terrain features and location range of the area to be measured, and then scan the area to be measured by launching a drone carrying a lidar device according to the scanning path to obtain the original point cloud data of the area to be measured.

[0059] The local scanning module is used to set multiple feature regions based on the terrain features of the area to be measured and to perform point cloud scanning on each feature region to obtain local point cloud data for each feature region.

[0060] The point cloud fusion module is used to convert the original point cloud data of the area under test and the local point cloud data of each feature area into the same coordinate system. Then, based on the coordinate position of each feature area in the area under test, the local point cloud data of each feature area is added to the original point cloud data of the area under test to obtain the overall point cloud data of the area under test. Finally, a point cloud model is constructed based on the overall point cloud data of the area under test.

[0061] The calculation module is used to calculate the required earthwork volume for the area under test based on the point cloud model of the area under test and the cut and fill standards of the area under test.

[0062] Furthermore, the point cloud fusion module is also used to acquire 3D images of the area to be tested using oblique photography technology via UAV, divide the area into grids based on the 3D images, and combine them with a GIS map to obtain a 3D grid map of the area to be tested; match the original point cloud data of the area to be tested with the 3D grid map of the area to be tested to obtain the point cloud set within each grid, and combine it with the grid corresponding to each feature region to obtain the point cloud data of each feature region; after deleting the point cloud data of each feature region from the original point cloud data of the area to be tested, the local point cloud data of each feature region is added to the original point cloud data of the area to be tested to obtain the point cloud data of the area to be tested.

[0063] Furthermore, the calculation module is also used to: derive a standard model of the area to be measured based on the coordinate position range of the area to be measured and the design parameters of the area to be measured, and match the standard model of the area to be measured with the point cloud model to obtain a difference model, and then derive the fill area, the cut area, the fill volume of the fill area, and the cut volume of the cut area based on the difference model.

[0064] Furthermore, the feature region includes at least one of the following regions: vegetation cover region, building cover region, water body cover region, complex slope region; the local scanning module specifically includes obtaining local point cloud data of each feature region by setting up a corresponding scanning device in each feature region; the scanning device is a total station or a handheld mobile scanner.

[0065] Furthermore, the point cloud fusion module is also used to: preprocess the original point cloud data of the area to be tested and the local point cloud data of each feature region; the preprocessing includes removing invalid point clouds, abnormal point clouds, and point clouds that do not belong to the area to be tested;

[0066] It also includes: a report generation module, which is used to generate a GIS map of the area to be tested, including the fill area, the cut area, the fill volume of the fill area, and the cut volume of the cut area, based on the calculation steps; and to generate a distribution map of the fill and cut areas of the area to be tested, and to distinguish the fill area and the cut area with different colors in the distribution map of the fill and cut areas.

[0067] Furthermore, it also includes: a task generation module, used to derive the number of days required for filling and excavation based on the coordinate range of the area to be measured, the required earthwork volume of the area, the number of available filling and excavation vehicles in the system, and big data mining and models; and to derive the filling and excavation end date and filling and excavation schedule based on the filling and excavation start date, the required number of days, and future weather forecasts; and to allocate a daily filling and excavation task schedule based on the filling and excavation schedule, the filling and excavation start date, the filling and excavation end date, and the number of filling and excavation vehicles dispatched daily, and push the daily filling and excavation task schedule to the vehicle dispatch center for scheduling of filling and excavation vehicles; the daily filling and excavation task schedule includes multiple filling and excavation tasks, each including the filling and excavation coordinate range and the earthwork volume; when the daily filling and excavation task schedule is completed, the required earthwork volume and the filling and excavation start date of the area to be measured are updated, and the system returns to the task generation module to re-estimate the required number of days for filling and excavation and generate a new filling and excavation schedule; until the required earthwork volume of the area to be measured is filled and excavated;

[0068] Based on the coordinate location range of the area to be measured and the required earthwork volume, combined with big data mining and models, the number of days required for filling and excavation is determined. Specifically, this involves: firstly, collecting the earthwork volume, filling and excavation days, and filling and excavation vehicles from historical water conservancy projects of the same type for excavation, and combining this with model training to obtain a correlation model between filling and excavation days, filling and excavation vehicles, and earthwork volume; then, based on the correlation model and the coordinate location range of the area to be measured, the required earthwork volume, and the filling and excavation vehicles, the number of filling and excavation days is estimated.

[0069] The task detection module is used to report the completion time of each cutting and filling task when it is completed, and to determine whether there are any abnormalities in the work of each cutting and filling vehicle based on the completion time of each cutting and filling task, so as to maintain the cutting and filling vehicles with abnormalities in a timely manner and dispatch new cutting and filling vehicles to work.

[0070] Furthermore, it also includes a dynamic demonstration module, which is used to periodically acquire the completed filling and excavation area of ​​the test area and combine it with the 3D image of the test area to generate a 3D animation model, which is then displayed to supervisors through a background display terminal and a mobile terminal.

[0071] The above embodiments are merely preferred embodiments of the present invention and should not be construed as limiting the scope of protection of the present invention. Any non-substantial changes and substitutions made by those skilled in the art based on the present invention shall fall within the scope of protection claimed by the present invention.

Claims

1. An AI-based method for measuring earthwork in water conservancy projects, characterized in that, The earthwork measurement methods for water conservancy projects include: Overall scanning steps: Set a scanning path based on the terrain features and location range of the area to be tested, and then scan the area to be tested by launching a drone carrying a lidar device according to the scanning path to obtain the raw point cloud data of the area to be tested; Local scanning steps: Based on the terrain features of the area to be measured, multiple feature regions are set up, and point cloud scanning is performed on each feature region to obtain local point cloud data for each feature region; Point cloud fusion steps: After converting the original point cloud data of the area to be tested and the local point cloud data of each feature region into the same coordinate system, the local point cloud data of each feature region is supplemented into the original point cloud data of the area to be tested according to the coordinate position of each feature region in the area to be tested to obtain the overall point cloud data of the area to be tested. Then, a point cloud model is constructed based on the overall point cloud data of the area to be tested. The point cloud fusion steps include: acquiring a 3D image of the area to be tested using UAV based on oblique photogrammetry technology and dividing the 3D image of the area to be tested into a grid, and obtaining a 3D grid map of the area to be tested by combining it with a GIS map; matching the original point cloud data of the area to be tested with the 3D grid map of the area to be tested to obtain the point cloud set in each grid, and obtaining the point cloud data of each feature region by combining it with the grid corresponding to each feature region; deleting the point cloud data of each feature region from the original point cloud data of the area to be tested, and supplementing the local point cloud data of each feature region into the original point cloud data of the area to be tested to obtain the point cloud data of the area to be tested. Calculation steps: Calculate the required earthwork volume for the area to be measured based on the point cloud model of the area to be measured and the cut and fill standards of the area to be measured.

2. The AI-based earthwork measurement method for water conservancy projects according to claim 1, characterized in that, The calculation steps specifically include: obtaining a standard model of the area to be measured based on the coordinate position range of the area to be measured and the design parameters of the area to be measured; matching the standard model of the area to be measured with the point cloud model to obtain a difference model; and then obtaining the fill area, the cut area, the fill volume of the fill area, and the cut volume of the cut area based on the difference model. It also includes: report generation steps: based on the calculation steps, the fill area, cut area, and the earthwork volume of the fill area and the earthwork volume of the cut area are obtained from the area to be tested. The GIS map of the area to be tested generates a distribution map of the fill and cut areas of the area to be tested and distinguishes the fill area and cut area with different colors in the distribution map of the fill and cut areas.

3. The AI-based earthwork measurement method for water conservancy projects according to claim 1, characterized in that, The feature region includes at least one of the following regions: vegetation cover region, building cover region, water body cover region, complex slope region; the local scanning step specifically includes obtaining local point cloud data of each feature region by setting up a corresponding scanning device in each feature region; the scanning device is a total station or a handheld mobile scanner.

4. The AI-based earthwork measurement method for water conservancy projects according to claim 1, characterized in that, The point cloud fusion step further includes: preprocessing the original point cloud data of the area to be tested and the local point cloud data of each feature region; the preprocessing includes removing invalid point clouds, abnormal point clouds and point clouds that do not belong to the area to be tested.

5. The AI-based earthwork measurement method for water conservancy projects according to claim 1, characterized in that, Also includes: Dynamic demonstration steps: The completed filling and excavation areas of the test area are acquired at regular intervals and combined with 3D images of the test area to generate a 3D animation model, which is then displayed to supervisors through a background display terminal and a mobile terminal.

6. The AI-based earthwork measurement method for water conservancy projects according to claim 1, characterized in that, The overall scanning step, which sets the scanning path based on the terrain features and location range of the area to be tested, specifically includes: acquiring multiple images of the area to be tested through a network or camera, extracting the terrain features and location range of the area to be tested from the multiple images based on image processing technology, simulating the area to be tested based on the terrain features and location range to obtain a two-dimensional planar image of the area to be tested, then designing multiple different scanning paths for the two-dimensional planar image of the area to be tested using a large AI model, simulating point cloud scanning for each scanning path, and finally selecting the optimal scanning path as the final scanning path based on the obtained point cloud data results.

7. The AI-based earthwork measurement method for water conservancy projects according to claim 1, characterized in that, Also includes: Task generation steps: Based on the coordinate range of the area to be measured, the required earthwork volume, and the number of available excavation and filling vehicles in the system, combined with big data mining and models, the number of days required for excavation and filling is determined. Based on the start date, the required number of days, and future weather forecasts, the end date and excavation and filling schedule are generated. A daily excavation and filling task schedule is then allocated based on the schedule, start date, end date, and the number of vehicles dispatched daily. This schedule is then pushed to the vehicle dispatch center for vehicle scheduling. The daily task schedule includes multiple excavation and filling tasks, each including the coordinate range and earthwork volume. When the daily task schedule is completed, the required earthwork volume and start date for the area to be measured are updated, and the process returns to the task generation steps to re-estimate the required number of days and generate a new schedule. This process continues until the required earthwork volume for the area to be measured is completed. Based on the coordinate range of the area to be measured and the required earthwork volume, combined with big data mining and modeling, the number of days required for filling and excavation is determined. Specifically, this involves: firstly, collecting data on the earthwork volume, number of days, and vehicles used for filling and excavation of similar historical water conservancy projects, and then using model training to derive a correlation model between the number of days, vehicles, and earthwork volume; and then estimating the number of days based on the correlation model and the coordinate range of the area to be measured, the required earthwork volume, and the vehicles used for filling and excavation. Task detection steps: When each cutting and filling vehicle completes its corresponding cutting and filling task, it reports the completion time of the cutting and filling task and judges whether there are any abnormalities in the work of each cutting and filling vehicle based on the completion time of the cutting and filling task of each vehicle, so as to maintain the cutting and filling vehicles with abnormalities in a timely manner and dispatch new cutting and filling vehicles to work.

8. An AI-based earthwork measurement system for water conservancy projects, characterized in that: include: The overall scanning module is used to set a scanning path according to the terrain features and location range of the area to be tested, and then scan the area to be tested by launching a drone carrying a lidar device according to the scanning path to obtain the original point cloud data of the area to be tested. The local scanning module is used to set multiple feature regions according to the terrain features of the area to be measured and to perform point cloud scanning on each feature region to obtain local point cloud data of each feature region. The point cloud fusion module is used to convert the original point cloud data of the area to be tested and the local point cloud data of each feature region into the same coordinate system. Then, based on the coordinate position of each feature region in the area to be tested, the local point cloud data of each feature region is supplemented into the original point cloud data of the area to be tested to obtain the overall point cloud data of the area to be tested. Then, a point cloud model is constructed based on the overall point cloud data of the area to be tested. The point cloud fusion module is also used to acquire 3D images of the area to be tested using UAV based on oblique photogrammetry technology, divide the 3D images of the area to be tested into grids, and combine them with a GIS map to obtain a 3D grid map of the area to be tested. The original point cloud data of the area to be tested is matched with the 3D grid map of the area to be tested to obtain the point cloud set in each grid. The point cloud data of each feature area is obtained by combining the grid corresponding to each feature area. After deleting the point cloud data of each feature area from the original point cloud data of the area to be tested, the local point cloud data of each feature area is filled into the original point cloud data of the area to be tested to obtain the point cloud data of the area to be tested. The calculation module is used to calculate the required earthwork volume for the area to be measured based on the point cloud model of the area to be measured and the cut and fill standards of the area to be measured.

Images