Cargo loading overrun retest method based on laser radar

Abstract

The invention discloses a cargo loading overrun remeasurement method based on a laser radar, and the method comprises the following steps: 1, carrying out the at least three fourths of a to-be-measured object through a handheld three-dimensional laser scanner, and obtaining a three-dimensional point cloud containing the to-be-measured object, moving track data of the handheld three-dimensional laser scanner are automatically recorded; step 2, converting moving track data of the handheld three-dimensional laser scanner into point cloud, projecting the point cloud to an XOY plane, solving a directed surrounding rectangle of the track point cloud, and expanding the directed surrounding rectangle in equal proportion to ensure that the cut point cloud contains all point cloud data of the object to be measured; through a full-automatic identification and calculation method, a worker does not need to climb to the top of the cargo for manual measurement, the labor intensity is greatly reduced, the working safety is improved, the laser radar and the advanced point cloud processing technology are utilized, rapid and accurate detection of cargo loading overrun is achieved, and the detection efficiency and accuracy are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of a method for re-measuring over-limit cargo loading based on a laser radar, and more specifically to a method for re-measuring over-limit cargo loading based on a laser radar. Background Art

[0002] In recent years, the speed of freight trains has been continuously improved, and railway limits have become one of the basic guarantees for safe railway operation. Improper cargo loading position and improper design of locomotive and vehicle dimensions can cause unsafe driving accidents, especially for large and irregular cargo loading or improper reinforcement, which may cause vehicle derailment and affect railway driving safety, directly endangering the safety of railway freight transportation. Therefore, it is necessary to accurately measure the cargo and determine the over-limit status to meet the needs of special cargo transportation business.

[0003] Handheld laser radar can quickly and accurately collect point cloud data of the object being measured, and has the advantages of high-resolution collection, high precision, and strong anti-interference ability.

[0004] Traditional detection methods have the disadvantages of high cost, poor flexibility and insufficient precision. The application of handheld 3D laser scanners in cargo measurement has not been fully utilized, especially in complex environments and irregular cargo shapes. Therefore, a new technical solution is needed to solve this problem. Summary of the invention

[0005] The purpose of the present invention is to provide a method for re-measurement of over-limit cargo loading based on laser radar, which solves the problems of high cost, poor flexibility and insufficient accuracy of traditional detection methods, and the application of handheld three-dimensional laser scanners in cargo measurement has not been fully utilized, especially in complex environments and irregular cargo shapes.

[0006] To achieve the above object, the present invention provides the following technical solution: a method for re-testing over-limit cargo loading based on laser radar, the method for re-testing over-limit cargo loading based on laser radar comprises the following steps:

[0007] Step 1. The handheld 3D laser scanner moves around at least three quarters of the object to be measured, obtains a 3D point cloud containing the object to be measured, and automatically records the movement trajectory data of the handheld 3D laser scanner;

[0008] Step 2. Convert the moving trajectory data of the handheld 3D laser scanner into a point cloud and project it onto the XOY plane. Calculate the directed enclosing rectangle of the trajectory point cloud and expand it proportionally to ensure that the cropped point cloud contains all the point cloud data of the object to be measured. Set the maximum and minimum Z values ​​according to the empirical values ​​and construct a polyhedron with the directed enclosing rectangle.

[0009] Step 3. Crop out all point clouds contained in the polyhedron, and exclude point clouds above the object to be measured, such as the station ceiling and the ground below the object to be measured;

[0010] Step 4. Perform voxel filtering on the cropped point cloud to reduce the number of point clouds and improve computing efficiency;

[0011] Step 5. Perform plane fitting on the point cloud, filter out planes with point cloud numbers greater than a given threshold, exclude interference from planes with too few points, and then filter out planes parallel to the ground based on the normal angles of the remaining planes, exclude planes perpendicular to the ground such as the side of the vehicle, and leave two plane point clouds that meet the requirements: the platform plane point cloud and the flatbed car plane point cloud, and their plane parameters;

[0012] Step 6. Perform density projection on the point cloud in step 4 in the Z-axis direction, filter out the point cloud with a density value greater than a given threshold, cluster the point cloud, and only retain the point cloud with the largest number of points;

[0013] Step 7. Calculate the convex hull point cloud and establish a closed polygon based on the two plane point clouds obtained in step 5;

[0014] Step 8. Obtain the edge point cloud of the flatbed plane in step 7;

[0015] Step 9. Perform straight line fitting on the edge point cloud in step 8;

[0016] Step 10. Calculate the projection plane according to the normal vector of the flatbed plane and the direction vector of the flatbed edge in step 9 and a point on the edge;

[0017] Step 11. Calculate the intersection of the centerline of the flatbed truck in step 9 and the projection plane in step 10;

[0018] Step 12. Substitute the x and y coordinate values ​​of the intersection point in step 11 into the plane equation of the flatbed truck, calculate the z value, and create a new point. Use the x, y values ​​of the intersection point and the calculated z value as the coordinate values ​​of the new point and use it as the midpoint of the edge of the flatbed truck.

[0019] Step 13. Use the flatbed truck plane closed polygon established in step 6 to clip the point cloud in step 4 in the Z-axis direction to obtain the cargo and flatbed truck plane point clouds;

[0020] Step 14. Project the point cloud in step 13 onto the projection plane in step 10;

[0021] Step 15. Read the bounding data in text format and convert it into a point cloud;

[0022] Step 16. Translate the limit point cloud so that the midpoint of its bottom edge coincides with the origin, and rotate the limit point cloud so that it is located in the XOZ plane, and its normal vector is parallel to the Y axis;

[0023] Step 17. Calculate the rotation matrix based on its normal vector and the projection plane normal vector, and rotate the bounding point cloud to make it parallel to the projection plane;

[0024] Step 18. Calculate the rotation matrix based on the bounding bottom edge vector and the flatbed vehicle edge vector, and rotate the bounding point cloud so that its bottom edge is parallel to the flatbed vehicle edge;

[0025] Step 19. Translate the limit point cloud so that the midpoint of its bottom edge coincides with the midpoint of the edge of the flatbed truck;

[0026] Step 20. Read the flatbed truck model, obtain the flatbed truck height, and translate the limit downward along the Z axis by the flatbed truck height;

[0027] Step 21. At this time, the projection of the flatbed truck and the goods to be measured are in the same plane as the limit, and their relative positions with the limit are consistent with the actual situation;

[0028] Step 22. Calculate the closed polygon of the bounding point cloud;

[0029] Step 23. Use convex polygons to clip their corresponding point clouds, and you can get the points inside and outside the polygons respectively;

[0030] Step 24. Cluster the external points and find the bounding rectangle of each cluster;

[0031] Step 25. Perform the same processing as steps 22, 23 and 24 on the outer point and the first-level overrun limit;

[0032] Step 26. If there are still outliers after processing in step 25, use the secondary overlimit limit to perform overlimit detection.

[0033] Step 27. Combining the above results to obtain the over-limit detection result;

[0034] Step 28. Visualize the bounding polygon and the projected point cloud, generate a picture, and display the level of over-limit. If there is an over-limit, frame the over-limit part in the picture according to the rectangle in step 24.

[0035] As a preferred implementation of the present invention, the efficiency of using the directed enclosing rectangle of the trajectory point cloud to construct a polyhedron to crop the point cloud in step 3 is much higher than directly using the trajectory point cloud for cropping.

[0036] As a preferred embodiment of the present invention, in step 7, the point cloud in step 6 is clipped along the Z axis according to the closed polygon. If the polygon clips the point cloud in step 6, it means that the plane corresponding to the polygon is the flatbed truck plane.

[0037] As a preferred embodiment of the present invention, the straight line fitting in step 9 selects out the two long sides and short sides of the flatbed plane rectangle, calculates the midlines of the two long sides on the flatbed plane based on the two long sides, and uses the short sides as the edge lines of the flatbed.

[0038] As a preferred embodiment of the present invention, the closed polygon in step 22 is usually a concave polygon, which is converted into a minimum number of convex polygons according to the height, and the projected point cloud is clipped according to the height of the top and bottom edges of the convex polygon.

[0039] Compared with the prior art, the present invention has the following beneficial effects:

[0040] In order to solve the problem of over-limit detection of cargo loading, the present invention adopts a handheld three-dimensional laser scanner to quickly obtain cargo point cloud data, and automatically performs over-limit detection, thereby improving detection efficiency and accuracy. The flexibility and high accuracy of the handheld three-dimensional laser scanner are utilized to realize the rapid collection and processing of cargo three-dimensional point cloud data, thereby solving the problem that traditional measurement methods cannot adapt to complex environments. Through algorithms such as plane fitting and straight line fitting, the flatbed truck is intelligently identified and the projection surface is calculated. The cargo point cloud is projected onto the projection surface and compared with the limit, thereby realizing automatic registration and over-limit level judgment. Point cloud processing technologies such as voxel filtering and density projection are adopted to effectively reduce the number of point clouds, improve computing efficiency, retain key information, and improve detection accuracy. The limit polygon and the projected point cloud are visualized, a picture is generated, and the over-limit level is displayed. If there is an over-limit, the over-limit part is framed, thereby making the detection result more intuitive and easy to understand. Through the fully automated recognition and calculation method, there is no need for staff to climb onto the top of the cargo for manual measurement, thereby greatly reducing the labor intensity and improving the safety of work. Laser radar and advanced point cloud processing technology are utilized to realize the rapid and accurate detection of over-limit cargo loading, thereby improving detection efficiency and accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 This is the final result of the present invention: a super over-limit view, in which the pink part is the part that exceeds the locomotive limit, the blue part is the part that exceeds the first-level limit, and the red part is the part that exceeds the second-level limit;

[0042] Figure 2 The point cloud image processed by step 4 of the present invention is firstly clipped according to the trajectory point cloud enclosing rectangle, then the point cloud above and below the object to be measured is clipped according to the height, and finally the point cloud is processed by speed-up filtering;

[0043] Figure 3 The point cloud image processed in step 5 of the present invention, that is, the two candidate planes selected by performing RANSAC plane fitting;

[0044] Figure 4The point cloud image processed in step 9 of the present invention is the point cloud of two long sides and one short side identified by linear fitting of the edge point cloud of the flatbed truck;

[0045] Figure 5 It includes the projection plane point cloud processed by step 14 of the present invention, the cargo and flatbed truck point clouds processed by step 13, and the long and short side point clouds processed by step 9. DETAILED DESCRIPTION

[0046] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0047] See also Figure 1-5 The present invention provides a technical solution: a method for re-testing over-limit cargo loading based on laser radar, the method for re-testing over-limit cargo loading based on laser radar comprises the following steps:

[0048] Step 1. The handheld 3D laser scanner moves around at least three quarters of the object to be measured, obtains a 3D point cloud containing the object to be measured, and automatically records the movement trajectory data of the handheld 3D laser scanner;

[0049] Step 2. Convert the moving trajectory data of the handheld 3D laser scanner into a point cloud and project it onto the XOY plane. Calculate the directed enclosing rectangle of the trajectory point cloud and expand it proportionally to ensure that the cropped point cloud contains all the point cloud data of the object to be measured. Set the maximum and minimum Z values ​​according to the empirical values ​​and construct a polyhedron with the directed enclosing rectangle.

[0050] Step 3. Crop out all point clouds contained in the polyhedron, and exclude point clouds above the object to be measured, such as the station ceiling and the ground below the object to be measured;

[0051] Step 4. Perform voxel filtering on the cropped point cloud to reduce the number of point clouds and improve computing efficiency;

[0052] Step 5. Perform plane fitting on the point cloud, filter out planes with point cloud numbers greater than a given threshold, exclude interference from planes with too few points, and then filter out planes parallel to the ground based on the normal angles of the remaining planes, exclude planes perpendicular to the ground such as the side of the vehicle, and leave two plane point clouds that meet the requirements: the platform plane point cloud and the flatbed car plane point cloud, and their plane parameters;

[0053] Step 6. Perform density projection on the point cloud in step 4 in the Z-axis direction, filter out the point cloud with a density value greater than a given threshold, cluster the point cloud, and only retain the point cloud with the largest number of points;

[0054] Step 7. Calculate the convex hull point cloud and establish a closed polygon based on the two plane point clouds obtained in step 5;

[0055] Step 8. Obtain the edge point cloud of the flatbed plane in step 7;

[0056] Step 9. Perform straight line fitting on the edge point cloud in step 8;

[0057] Step 10. Calculate the projection plane according to the normal vector of the flatbed plane and the direction vector of the flatbed edge in step 9 and a point on the edge;

[0058] Step 11. Calculate the intersection of the centerline of the flatbed truck in step 9 and the projection plane in step 10;

[0059] Step 12. Substitute the x and y coordinate values ​​of the intersection point in step 11 into the plane equation of the flatbed truck, calculate the z value, and create a new point. Use the x, y values ​​of the intersection point and the calculated z value as the coordinate values ​​of the new point and use it as the midpoint of the edge of the flatbed truck.

[0060] Step 13. Use the flatbed truck plane closed polygon established in step 6 to clip the point cloud in step 4 in the Z-axis direction to obtain the cargo and flatbed truck plane point clouds;

[0061] Step 14. Project the point cloud in step 13 onto the projection plane in step 10;

[0062] Step 15. Read the bounding data in text format and convert it into a point cloud;

[0063] Step 16. Translate the limit point cloud so that the midpoint of its bottom edge coincides with the origin, and rotate the limit point cloud so that it is located in the XOZ plane, and its normal vector is parallel to the Y axis;

[0064] Step 17. Calculate the rotation matrix based on its normal vector and the projection plane normal vector, and rotate the bounding point cloud to make it parallel to the projection plane;

[0065] Step 18. Calculate the rotation matrix based on the bounding bottom edge vector and the flatbed vehicle edge vector, and rotate the bounding point cloud so that its bottom edge is parallel to the flatbed vehicle edge;

[0066] Step 19. Translate the limit point cloud so that the midpoint of its bottom edge coincides with the midpoint of the edge of the flatbed truck;

[0067] Step 20. Read the flatbed truck model, obtain the flatbed truck height, and translate the limit downward along the Z axis by the flatbed truck height;

[0068] Step 21. At this time, the projection of the flatbed truck and the goods to be measured are in the same plane as the limit, and their relative positions with the limit are consistent with the actual situation;

[0069] Step 22. Calculate the closed polygon of the bounding point cloud;

[0070] Step 23. Use convex polygons to clip their corresponding point clouds, and you can get the points inside and outside the polygons respectively;

[0071] Step 24. Cluster the external points and find the bounding rectangle of each cluster;

[0072] Step 25. Perform the same processing as steps 22, 23 and 24 on the outer point and the first-level overrun limit;

[0073] Step 26. If there are still outliers after processing in step 25, use the secondary overlimit limit to perform overlimit detection.

[0074] Step 27. Combining the above results to obtain the over-limit detection result;

[0075] Step 28. Visualize the bounding polygon and the projected point cloud, generate a picture, and display the level of over-limit. If there is an over-limit, frame the over-limit part in the picture according to the rectangle in step 24.

[0076] Further improved, the efficiency of cropping the point cloud by constructing a polyhedron using the directed enclosing rectangle of the trajectory point cloud in step 3 is much higher than directly using the trajectory point cloud for cropping. By constructing a polyhedron, the point cloud data containing the object to be measured can be more effectively screened out, while excluding irrelevant point clouds, such as the station ceiling and ground, thereby improving the overall processing speed.

[0077] As a further improvement, in step 7, the point cloud in step 6 is cropped along the Z axis according to the closed polygon. If the polygon crops the point cloud in step 6, it means that the plane corresponding to the polygon is the flatbed truck plane. By cropping the point cloud along the Z axis and verifying the cropping result, the flatbed truck plane can be effectively identified, providing an accurate basis for subsequent over-limit detection.

[0078] Further improved, the straight line fitting in step 9 selects out the two long sides and short sides of the flatbed truck plane rectangle, calculates the center line of the two long sides on the flatbed truck plane based on the two long sides, and the short side serves as the edge line of the flatbed truck. By identifying the long and short sides of the flatbed truck through straight line fitting and calculating the center line, the edge position of the flatbed truck can be determined more accurately, providing more accurate data support for subsequent over-limit detection.

[0079] As a further improvement, the closed polygon in step 22 is usually a concave polygon, and the concave polygon is converted into a minimum number of convex polygons according to the height, and the projected point cloud is cropped according to the height of the top and bottom edges of the convex polygon. By converting the concave polygon into a convex polygon, the limit point cloud can be processed more conveniently, and cropped according to the height of the top and bottom edges of the convex polygon, thereby ensuring the accurate alignment of the projected point cloud with the limit point cloud and improving the accuracy of over-limit detection.

[0080] Working principle: The present invention uses a handheld 3D laser scanner to rotate at least three-quarters around the object to be measured, obtains 3D point cloud data containing the object to be measured, and automatically records the movement trajectory data of the handheld 3D laser scanner; converts the movement trajectory data into a point cloud and projects it to the XOY plane, calculates the directed enclosing rectangle of the trajectory point cloud, and expands it proportionally, constructs a polyhedron to crop out all the point cloud data containing the object to be measured, and excludes point clouds above the object to be measured (such as the station ceiling) and point clouds below the object to be measured (such as the ground); performs voxel filtering on the cropped point cloud to reduce the number of point clouds and improve computing efficiency; performs plane fitting on the point cloud, screens out planes parallel to the ground (platform plane and flatbed car plane), and excludes planes with too few points or perpendicular to the ground; performs density projection on the point cloud in the Z-axis direction, screens out point clouds with density values ​​greater than a given threshold, clusters the point cloud, retains only the point cloud with the largest number of points, and calculates the convex hull Point cloud, establish a closed polygon, identify the edge point cloud of the flatbed plane, and perform straight line fitting to calculate the center line and edge line of the flatbed; calculate the projection plane according to the normal vector and edge direction vector of the flatbed plane, project the cargo and flatbed plane point cloud onto the projection plane, read the limit data in text format, convert it into a point cloud, and perform translation and rotation operations to make it parallel to the projection plane and the edge of the flatbed, and the midpoint of the bottom edge coincides; calculate the closed polygon of the limit point cloud, use a convex polygon to clip the projected point cloud, and obtain the points inside and outside the polygon respectively, cluster the external points, find the circumscribed rectangle of each class, and compare it with the first-level over-limit limit. If there are still external points, use the second-level over-limit limit for over-limit detection; combine the above results to obtain the over-limit detection result, visualize the limit polygon and the projected point cloud, generate a picture, and display the over-limit level. If there is an over-limit, frame the over-limit part in the picture according to the circumscribed rectangle.

[0081] The above shows and describes the basic principles and main features of the present invention and the advantages of the present invention. It is obvious to those skilled in the art that the present invention is not limited to the details of the above exemplary embodiments, and the present invention can be implemented in other specific forms without departing from the spirit or basic features of the present invention. Therefore, no matter from which point of view, the embodiments should be regarded as exemplary and non-restrictive. The scope of the present invention is defined by the attached claims rather than the above description, and it is intended that all changes falling within the meaning and scope of the equivalent elements of the claims are included in the present invention. Any figure mark in the claims should not be regarded as limiting the claims involved.

[0082] Finally, a few points should be explained: First, in the description of this application, it should be noted that, unless otherwise specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, and can be mechanical or electrical connections, or internal connectivity between two components, or direct connections. "Up", "down", "left", "right", etc. are only used to indicate relative position relationships. When the absolute position of the object being described changes, the relative position relationship may change.

[0083] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments or replace some of the technical features therein by equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for retesting over-limit cargo loading based on laser radar, characterized in that: The method for retesting over-limit loading of cargo based on laser radar includes the following steps: Step 1. The handheld 3D laser scanner moves around at least three quarters of the object to be measured, obtains a 3D point cloud containing the object to be measured, and automatically records the movement trajectory data of the handheld 3D laser scanner; Step 2. Convert the moving trajectory data of the handheld 3D optical scanner into a point cloud and project it onto the XOY plane. Calculate the directed enclosing rectangle of the trajectory point cloud and expand it proportionally to ensure that the cropped point cloud contains all the point cloud data of the object to be measured. Set the maximum and minimum Z values ​​according to the empirical values ​​and construct a polyhedron with the directed enclosing rectangle. Step 3. Crop out all point clouds contained in the polyhedron, and exclude point clouds above the object to be measured, such as the station ceiling and the ground below the object to be measured; Step 4. Perform voxel filtering on the cropped point cloud to reduce the number of point clouds and improve computing efficiency; Step 5. Perform plane fitting on the point cloud, filter out planes with point cloud numbers greater than a given threshold, exclude interference from planes with too few points, and then filter out planes parallel to the ground based on the normal angles of the remaining planes, exclude planes perpendicular to the ground such as the side of the vehicle, and leave two plane point clouds that meet the requirements: the platform plane point cloud and the flatbed car plane point cloud, and their plane parameters; Step 6. Perform density projection on the point cloud in step 4 in the Z-axis direction, filter out the point cloud with a density value greater than a given threshold, cluster the point cloud, and only retain the point cloud with the largest number of points; Step 7. Calculate the convex hull point cloud and establish a closed polygon based on the two plane point clouds obtained in step 5; Step 8. Obtain the edge point cloud of the flatbed plane in step 7; Step 9. Perform straight line fitting on the edge point cloud in step 8; Step 10. Calculate the projection plane according to the normal vector of the flatbed plane and the direction vector of the flatbed edge in step 9 and a point on the edge; Step 11. Calculate the intersection of the centerline of the flatbed truck in step 9 and the projection plane in step 10; Step 12. Substitute the x and y coordinate values ​​of the intersection point in step 11 into the plane equation of the flatbed truck, calculate the z value, and create a new point. Use the x, y values ​​of the intersection point and the calculated z value as the coordinate values ​​of the new point and use it as the midpoint of the edge of the flatbed truck. Step 13. Use the flatbed truck plane closed polygon established in step 6 to clip the point cloud in step 4 in the Z-axis direction to obtain the cargo and flatbed truck plane point clouds; Step 14. Project the point cloud in step 13 onto the projection plane in step 10; Step 15. Read the bounding data in text format and convert it into a point cloud; Step 16. Translate the limit point cloud so that the midpoint of its bottom edge coincides with the origin, and rotate the limit point cloud so that it is located in the XOZ plane, and its normal vector is parallel to the Y axis; Step 17. Calculate the rotation matrix based on its normal vector and the projection plane normal vector, and rotate the bounding point cloud to make it parallel to the projection plane; Step 18. Calculate the rotation matrix based on the bounding bottom edge vector and the flatbed vehicle edge vector, and rotate the bounding point cloud so that its bottom edge is parallel to the flatbed vehicle edge; Step 19. Translate the limit point cloud so that the midpoint of its bottom edge coincides with the midpoint of the edge of the flatbed truck; Step 20. Read the flatbed truck model, obtain the flatbed truck height, and translate the limit downward along the Z axis by the flatbed truck height; Step 21. At this time, the projection of the flatbed truck and the goods to be measured are in the same plane as the limit, and their relative positions with the limit are consistent with the actual situation; Step 22. Calculate the closed polygon of the bounding point cloud; Step 23. Use convex polygons to clip their corresponding point clouds, and you can get the points inside and outside the polygons respectively; Step 24. Cluster the external points and find the bounding rectangle of each cluster; Step 25. Perform the same processing as steps 22, 23 and 24 on the outer point and the first-level overrun limit; Step 26. If there are still outliers after processing in step 25, use the secondary overrun limit to perform overrun detection; Step 27. Combining the above results to obtain the over-limit detection result; Step 28. Visualize the bounding polygon and the projected point cloud, generate a picture, and display the level of over-limit. If there is an over-limit, frame the over-limit part in the picture according to the rectangle in step 24.

2. The method for retesting over-limit cargo loading based on laser radar according to claim 1 is characterized in that: The efficiency of using the directed enclosing rectangle of the trajectory point cloud to construct a polyhedron to crop the point cloud in step 3 is much higher than directly using the trajectory point cloud for cropping.

3. The method for retesting over-limit cargo loading based on laser radar according to claim 1 is characterized in that: In the step 7, the point cloud in step 6 is clipped along the Z axis according to the closed polygon. If the polygon clips out the point cloud in step 6, it means that the plane corresponding to the polygon is the flatbed car plane.

4. The method for retesting over-limit cargo loading based on laser radar according to claim 1 is characterized in that: The straight line fitting in step 9 selects two long sides and short sides of the flatbed plane rectangle, calculates the midlines of the two long sides on the flatbed plane based on the two long sides, and uses the short sides as the edge lines of the flatbed.

5. The method for retesting over-limit cargo loading based on laser radar according to claim 1 is characterized in that: The closed polygon in step 22 is usually a concave polygon. The concave polygon is converted into a minimum number of convex polygons according to the height, and the projected point cloud is clipped according to the height of the top and bottom edges of the convex polygon.

Images