WO2021081958A1 - Terrain detection method, movable platform, control device, system, and storage medium - Google Patents

Abstract

A terrain detection method, a movable platform, a control device, a system, and a storage medium. The method comprises: acquiring point cloud data (S101); obtaining a two-dimensional image according to the point cloud data (S102); performing pixel assignment on the two-dimensional image to obtain a completed two-dimensional image (S103); performing processing on the completed two-dimensional image to obtain a processed two-dimensional image (S104); reconstructing the point cloud data according to the processed two-dimensional image (S105); and determining terrain information according to the reconstructed point cloud data (S106).

Description

Terrain detection method, movable platform, control equipment, system and storage medium Technical field

The invention relates to the field of detection technology, in particular to a terrain detection method, a movable platform, a control device, a system and a storage medium.

Background technique

At present, when using detection equipment to detect the terrain of the ground, three-dimensional point cloud data is generally collected, and the detection equipment is, for example, a radar, a laser device, or a camera. In order to improve the accuracy of terrain detection, it is necessary to remove noise from the collected 3D point cloud data. However, most of the existing noise removal methods are based on prior information, set certain clustering rules and search methods, The original observation points in the point cloud data are clustered. The original observation points that do not meet the clustering rules and cannot form a point cluster are regarded as noise removal, and the remaining original observation points are divided according to the point cluster to which they belong. However, this method relies on the choice of clustering rules and search methods. If the appropriate clustering rules and search methods are not selected, the final clustering effect will deviate greatly from the actual situation, resulting in relatively large errors in the terrain detection results. Large, and the clustering algorithm has higher requirements on the processor, and consumes more computing resources.

Therefore, how to improve the accuracy of terrain detection becomes an urgent problem to be solved.

Summary of the invention

Based on this, this application provides a terrain detection method, a movable platform, a control device, a system, and a storage medium to improve the accuracy of terrain detection.

In the first aspect, this application provides a terrain detection method, which includes:

Obtain 3D point cloud data containing terrain information;

Obtaining a two-dimensional image corresponding to the three-dimensional point cloud data according to the three-dimensional point cloud data;

Pixel assignment is performed on pixels lacking pixel values in the two-dimensional image to obtain a two-dimensional image with pixel complementation;

Performing morphological processing on the two-dimensional image after the pixel complementation to obtain a processed two-dimensional image;

Reconstructing three-dimensional point cloud data according to the processed two-dimensional image; and

According to the reconstructed 3D point cloud data, the terrain information is determined.

In the second aspect, the present application also provides a movable platform, which includes a detection device, a memory, and a processor;

The detection device is used for terrain detection and collecting three-dimensional point cloud data containing terrain information;

The memory is used to store a computer program;

The processor is configured to execute the computer program and, when executing the computer program, implement the following steps:

Obtain 3D point cloud data containing terrain information;

Obtaining a two-dimensional image corresponding to the three-dimensional point cloud data according to the three-dimensional point cloud data;

Pixel assignment is performed on pixels lacking pixel values in the two-dimensional image to obtain a two-dimensional image with pixel complementation;

Performing morphological processing on the two-dimensional image after the pixel complementation to obtain a processed two-dimensional image;

Reconstructing three-dimensional point cloud data according to the processed two-dimensional image; and

According to the reconstructed 3D point cloud data, the terrain information is determined.

In a third aspect, the present application also provides a control device, the control device including a memory and a processor;

The memory is used to store a computer program;

The processor is configured to execute the computer program and, when executing the computer program, implement the steps of the above-mentioned terrain detection method, and send the determined terrain information to the movable platform.

In a fourth aspect, the present application also provides a control system, the control system includes an aircraft and the control device as described in the third aspect; wherein, the movable platform is used to collect three-dimensional point cloud data and include The three-dimensional point cloud data of the terrain information is sent to the control device.

In a fifth aspect, the present application also provides a computer-readable storage medium that stores a computer program, and when the computer program is executed by a processor, the processor implements the above-mentioned terrain detection method.

The terrain detection method, movable platform, control equipment, system and storage medium proposed by the present invention can improve the accuracy of terrain detection.

It should be understood that the above general description and the following detailed description are only exemplary and explanatory, and cannot limit the application.

Description of the drawings

In order to explain the technical solutions of the embodiments of the present application more clearly, the following will briefly introduce the drawings used in the description of the embodiments. Obviously, the drawings in the following description are some embodiments of the present application. Ordinary technicians can obtain other drawings based on these drawings without creative work.

Fig. 1 is a schematic block diagram of a control system provided by an embodiment of the present application;

FIG. 2 is a schematic structural diagram of an aircraft provided by an embodiment of the present application;

FIG. 3 is a schematic diagram of a radar when collecting terrain information according to an embodiment of the present application;

4 is a schematic flowchart of steps of a terrain detection method provided by an embodiment of the present application;

FIG. 5 is a schematic diagram of a Delaunay triangulation network constructed according to an embodiment of the present application;

Fig. 6 is a schematic block diagram of a movable platform provided by an embodiment of the present application;

Fig. 7 is a schematic block diagram of a control device provided by an embodiment of the present application.

Detailed ways

The technical solutions in the embodiments of the present application will be described clearly and completely in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, rather than all of them. Based on the embodiments in this application, all other embodiments obtained by those of ordinary skill in the art without creative work shall fall within the protection scope of this application.

The flowchart shown in the drawings is only an example, and does not necessarily include all contents and operations/steps, nor does it have to be executed in the described order. For example, some operations/steps can also be decomposed, combined or partially combined, so the actual execution order may be changed according to actual conditions.

Hereinafter, some embodiments of the present application will be described in detail with reference to the accompanying drawings. In the case of no conflict, the following embodiments and features in the embodiments can be combined with each other.

The embodiments of the present application provide a terrain detection method, a movable platform, a control device, a system, and a storage medium, which are used to improve the accuracy of terrain detection by the detection device mounted on the movable platform. Specifically, in the embodiment of the present invention, the three-dimensional point cloud data is converted into a two-dimensional image, the two-dimensional image is processed, the noise is removed from the two-dimensional image, and other operations are performed according to the processed two-dimensional image. Image reconstruction of three-dimensional point cloud data, to achieve accurate detection of terrain information, and improve the accuracy of terrain detection.

Among them, the control system includes a movable platform and control equipment.

Exemplarily, the movable platform includes an aircraft, a robot, or an autonomous vehicle, etc. The movable platform is equipped with a detection device, and the detection device includes a radar, a ranging sensor, etc., for ease of description, this application uses the control device as the radar for detailed introduction .

Exemplarily, the control device includes a ground control platform, mobile phone, tablet computer, notebook computer, PC computer, and the like.

Exemplarily, as shown in FIG. 1, the control system is a terrain detection system, and the terrain detection system 100 includes an aircraft 110 and a control device 120.

The aircraft 110 includes a drone, which includes a rotary-wing drone, such as a four-rotor drone, a hexarotor drone, and an eight-rotor drone. It can also be a fixed-wing drone or a rotary-wing drone. The combination of type and fixed-wing UAV is not limited here.

FIG. 2 is a schematic structural diagram of an aircraft 110 according to an embodiment of the present specification. In this embodiment, a rotary wing unmanned aerial vehicle is taken as an example for description.

The aircraft 110 may include a power system, a flight control system, and a frame. The aircraft 110 may communicate with the control device 120 wirelessly, and the control device 120 may display flight information of the aircraft. The control device 120 may communicate with the aircraft 110 in a wireless manner for remote control of the aircraft 110.

Wherein, the frame may include a fuselage 111 and a tripod 112 (also referred to as a landing gear). The fuselage 111 may include a center frame 1111 and one or more arms 1112 connected to the center frame 1111, and the one or more arms 1112 extend radially from the center frame. The tripod 112 is connected to the fuselage 111 for supporting the aircraft 110 when the aircraft 110 is landing.

The power system may include one or more electronic governors (referred to as ESCs for short), one or more propellers 113, and one or more motors 114 corresponding to the one or more propellers 113, where the motors 114 are connected to the electronic governors. Between the speed controller and the propeller 113, the motor 114 and the propeller 113 are arranged on the arm 1112 of the aircraft 110; the electronic governor is used to receive the driving signal generated by the flight control system, and provide a driving current to the motor according to the driving signal to control the motor 114 speed. The motor 114 is used to drive the propeller 113 to rotate, so as to provide power for the flight of the aircraft 110, and the power enables the aircraft 110 to realize movement of one or more degrees of freedom. In some embodiments, the aircraft 110 may rotate about one or more rotation axes. For example, the aforementioned rotation axis may include a roll axis, a yaw axis, and a pitch axis. It should be understood that the motor 114 may be a DC motor or an AC motor. In addition, the motor 114 may be a brushless motor or a brushed motor.

The flight control system may include a flight controller and a sensing system. The sensing system is used to measure the attitude information of the unmanned aerial vehicle, that is, the position information and state information of the aerial vehicle 110 in space, such as three-dimensional position, three-dimensional angle, three-dimensional velocity, three-dimensional acceleration, and three-dimensional angular velocity. The sensing system may include, for example, at least one of sensors such as a gyroscope, an ultrasonic sensor, an electronic compass, an inertial measurement unit (IMU), a vision sensor, a global navigation satellite system, and a barometer. For example, the global navigation satellite system may be the Global Positioning System (GPS). The flight controller is used to control the flight of the aircraft 110, for example, it can control the flight of the aircraft 110 according to the attitude information measured by the sensing system. It should be understood that the flight controller may control the aircraft 110 according to pre-programmed program instructions, or may control the aircraft 110 by responding to one or more control instructions from the control device 120.

As shown in FIG. 1, the tripod 112 of the aircraft 110 is equipped with a radar 115, and the radar 115 is used to realize the function of surveying terrain information. Among them, the aircraft 110 may include two or more tripods 112, and the radar 115 is mounted on one of the tripods 112.

The radar mainly includes an RF front-end module and a signal processing module. The RF front-end module includes a transmitting antenna and a receiving antenna. The signal processing module is responsible for generating modulated signals and processing and analyzing the collected intermediate frequency signals.

Specifically, the RF front-end module receives the modulated signal to generate a high-frequency signal whose frequency changes linearly with the modulated signal, and radiates downward through the transmitting antenna. The electromagnetic wave encounters the ground, targets or obstacles and is reflected back, and then is received by the receiving antenna and transmitted. The signal and the intermediate frequency are mixed to obtain an intermediate frequency signal, and the speed information and distance information can be obtained according to the frequency of the intermediate frequency signal.

Radar encounters a target object by radiating electromagnetic waves in space, and the scattered echo from the target object is received by the radar to detect the target object. When the radar is flying with the movable platform, it continuously collects the coordinates of the observation point by radiating electromagnetic waves, and finally collects three-dimensional point cloud data including terrain information. The movable platform sends the collected three-dimensional point cloud data including terrain information to the control device, and the control device processes the three-dimensional point cloud data to determine the terrain information, and the control device can also send the determined terrain information to the movable platform .

As shown in Figure 3, it is a schematic diagram of radar collecting terrain information. The data collected by the radar includes the depth of field detection distance, the horizontal detection distance, and the height position information, that is, the elevation value. It should be noted that the three-dimensional point clouds used for the two-dimensional image construction are all in the geodetic coordinate system.

It should be understood that the aforementioned naming of the various components of the aircraft is only for identification purposes, and should not be understood as a limitation to the embodiments of this specification.

Please refer to FIG. 4, which is a schematic flowchart of steps of a terrain detection method according to an embodiment of the present application. This method can be applied to control equipment to improve the accuracy of terrain detection.

The terrain detection method will be introduced in detail below in conjunction with the control system in FIG. 1. It should be understood that the control system in FIG. 1 does not constitute a limitation on the application scenario of the terrain detection method.

As shown in Fig. 4, the terrain detection method includes step S101 to step S106.

S101. Acquire three-dimensional point cloud data containing terrain information.

Wherein, the three-dimensional point cloud data includes a plurality of observation points, each of the observation points includes first position information, second position information, and height position information, wherein the first position information and the second position information are different. In the specific implementation process, the first position information, the second position information, and the height position information are perpendicular to each other. In this embodiment, the first position information may be a depth of field detection distance, the second position information may be a horizontal detection distance, and the height position information may be an elevation value.

During the flight of the aircraft, the radar mounted on the aircraft scans the terrain information of the flight area of the aircraft to obtain three-dimensional point cloud data containing terrain information, and send the obtained three-dimensional point cloud data containing terrain information to controlling device.

S102. Obtain a two-dimensional image corresponding to the three-dimensional point cloud data according to the three-dimensional point cloud data.

According to an embodiment of the present invention, the three-dimensional point cloud data is mapped to the corresponding two-dimensional image according to the three-dimensional point cloud data.

Specifically, the control device projects the three-dimensional point cloud data into a two-dimensional image according to the acquired three-dimensional point cloud data to obtain a two-dimensional image corresponding to the three-dimensional point cloud data.

In some embodiments, the method of obtaining a two-dimensional image corresponding to the three-dimensional point cloud data according to the three-dimensional point cloud data is specifically: determining a two-dimensional matrix according to the three-dimensional point cloud data, and comparing all data according to the two-dimensional matrix. The three-dimensional point cloud data is projected to obtain a two-dimensional image.

Among them, the corresponding two-dimensional matrix is determined according to the three-dimensional point cloud data. Each observation point in the three-dimensional point cloud data corresponds to a matrix unit in the two-dimensional matrix, where the two-dimensional matrix is a two-dimensional pixel matrix that constitutes a two-dimensional image. Each matrix unit in the matrix is a pixel in the two-dimensional image. According to the two-dimensional matrix, all observation points in the three-dimensional point cloud data are projected one by one to obtain a two-dimensional image corresponding to the three-dimensional point cloud data.

In the specific implementation process, one observation point can correspond to one matrix unit, or multiple observation points can correspond to one matrix unit.

Determine the two-dimensional matrix according to the three-dimensional point cloud data, and project the three-dimensional point cloud data according to the correspondence between the matrix units in the two-dimensional matrix and the observation points in the three-dimensional point cloud data to improve the two-dimensional image obtained by the projection of the three-dimensional point cloud data Accuracy.

In some embodiments, the manner of determining the two-dimensional matrix according to the three-dimensional point cloud data is specifically: determining the first target location information and the second location information of the observation points in the three-dimensional point cloud data. The second target location information; the distance resolution is acquired, and a two-dimensional matrix is determined according to the first location information, the second location information, and the distance resolution.

Among them, the range resolution refers to the resolution used by the radar mounted on the aircraft to scan terrain information.

Exemplarily, determining the first target location information and the second target location information according to the first location information and the second location information of the observation points in the three-dimensional point cloud data may be calculated by calculating the average of the first location information of the multiple observation points. The value determines the first target position information, and the average value of the second position information of the multiple observation points is calculated to determine the second target position information.

In the specific implementation process, obtain the first position information of all observation points in the 3D point cloud data, that is, the depth of field detection distance, calculate the sum of the depth detection distances of all observation points, and then divide by the sum of the depth detection distances of all observation points According to the number of observation points, the average value of the depth-of-field detection distance is obtained, and the average value of the depth-of-field detection distance is used as the first target position information. Obtain the second position information of all observation points in the 3D point cloud data, that is, the horizontal detection distance, calculate the sum of the horizontal detection distances of all observation points, and then divide the sum of the horizontal detection distances of all observation points by the number of observation points , The average value of the horizontal detection distance is obtained, and the average value of the horizontal detection distance is used as the second target position information.

Exemplarily, the method of determining the first target location information and the second target location information according to the first location information and the second location information of the observation point in the three-dimensional point cloud data may be: the observation point from the three-dimensional point cloud data The largest first location information and the largest second location information are determined in the first location information and the second location information as the first target location information and the second target location information, respectively.

In the specific implementation process, the first position information of all observation points in the three-dimensional point cloud data, that is, the depth-of-field detection distance, is acquired, and the largest depth-of-field detection distance is determined from the depth-of-field detection distances of multiple observation points as the first target position information. Obtain the second position information of all observation points in the three-dimensional point cloud data, that is, the horizontal detection distance, and determine the largest horizontal detection distance from the horizontal detection distances of the multiple observation points as the first target position information.

After obtaining the first target location information and the second target location information, the length and width of the two-dimensional matrix are determined according to the distance resolution.

Exemplarily, the length of the two-dimensional matrix may be twice the first target location information divided by the distance resolution, and the width of the two-dimensional matrix may be twice the second target location information divided by the distance resolution. It is understandable that the length and width of the two-dimensional matrix can be interchanged.

In some embodiments, the method of projecting the three-dimensional point cloud data according to the two-dimensional matrix to obtain a two-dimensional image is specifically: determining that the observation point of the three-dimensional point cloud data corresponds in the two-dimensional matrix The matrix index of the observation point; assign the height position information of the observation point to the matrix element corresponding to the matrix index of the observation point; use the matrix index of the two-dimensional matrix as a pixel and correspond to the matrix element of the two-dimensional matrix The height position information is used as the pixel value of the pixel to obtain a two-dimensional image.

Among them, determine the matrix index of each observation point in the three-dimensional point cloud data in the two-dimensional matrix, use the matrix index corresponding to the observation point as the pixel point, and use the height position information of the observation point, that is, the elevation value of the observation point as the pixel The pixel value of the point to obtain a two-dimensional image. Determine the matrix index corresponding to each observation point in the 3D point cloud data, use the matrix index to project the 3D point cloud data, improve the accuracy of the 3D point cloud data during projection, and improve the accuracy of the obtained 2D image .

Exemplarily, the step of determining the matrix index corresponding to the observation point of the three-dimensional point cloud data in the two-dimensional matrix is specifically as follows:

Calculate the sum of the first position information of the observation point and the largest first position information, and divide the sum of the first position information and the largest first position information of the observation point by the quotient of the distance resolution as the first matrix index An index value;

Calculate the sum of the second position information of the observation point and the largest second position information, and divide the sum of the second position information and the largest second position information of the observation point by the quotient of the distance resolution as the first matrix index Two index value.

Wherein, the first index value and the second index value are used to determine the matrix unit corresponding to the observation point in a two-dimensional matrix. In a specific implementation process, the first index value may be that the observation point is in the The coordinates in the length direction in the two-dimensional matrix, and the second index value may be the coordinates in the width direction of the observation point in the two-dimensional matrix.

It should be noted that when the length of the two-dimensional matrix can be twice the first target position information divided by the distance resolution, and the width of the two-dimensional matrix can be twice the second target position information divided by the distance resolution, The first index value may be the coordinates of the observation point in the length direction of the two-dimensional matrix, and the second index value may be the coordinates of the observation point in the width direction of the two-dimensional matrix.

When the width of the two-dimensional matrix is twice the first target position information divided by the distance resolution, and the length of the two-dimensional matrix is twice the second target position information divided by the distance resolution, the first index value may be The coordinates of the observation point in the width direction of the two-dimensional matrix, and the second index value may be the coordinates of the observation point in the length direction of the two-dimensional matrix.

Exemplarily, after the matrix index corresponding to the observation point is determined, the height position information of the observation point is assigned to the matrix element corresponding to the matrix index of the observation point, thereby obtaining the pixel value of the pixel point corresponding to the observation point .

For example, the coordinates of the observation point are (x _i , y _i ), where x _i is the first position information of the observation point, which is the depth of field detection distance, and x _i is the second position information of the observation point, which is the horizontal detection distance .

Matrix index corresponding to the point of observation (I _i, J _i), where, I _i is the first index value matrix index, J _i for the second index value of the index matrix. Then there is

Among them, r is the range resolution, L _x is the maximum first position information, that is, the maximum depth of field detection distance, and _Ly is the maximum second position information, that is, the maximum horizontal detection distance.

Matrix index (I _i, J _i) corresponding to the matrix _{elements: Img (I i, J i} ) = z i, where, z _i is the height of the position information, i.e. elevation values.

Exemplarily, the step of assigning the height position information of the observation point to the matrix element corresponding to the matrix index of the observation point is specifically as follows:

If there are multiple observation points corresponding to the same matrix index, determine the maximum height corresponding to the multiple observation points; assign the maximum height to the matrix element corresponding to the same matrix index.

Among them, the matrix index corresponding to the observation point is judged separately, and if multiple observation points correspond to the same matrix index, the largest value in the height position information of the several observation points is assigned to the matrix element corresponding to the matrix index.

When multiple observation points are cited in a matrix index, in order to prevent the pixel value from being assigned multiple times and cause the loss of the most reliable height position information of the observation point, therefore, the maximum height position information of the multiple observation points is used as the matrix index Corresponding matrix elements to improve the integrity of the information retention of the three-dimensional point cloud data in the two-dimensional image obtained by the projection.

S103: Perform pixel assignment on pixels lacking pixel values in the two-dimensional image to obtain a two-dimensional image with pixel complementation.

When the radar scans terrain information, because the radar scan has a scanning blind area, and when the radar uses different range resolutions to scan the same scanning area multiple times, it will appear that the three-dimensional point cloud data is projected into a two-dimensional image. Some of the pixels lack pixel values. Therefore, pixel values need to be assigned to pixels lacking pixel values to complement the missing parts in the two-dimensional image to avoid errors in the subsequent morphological processing of the two-dimensional image.

In some embodiments, the pixel assignment of the pixels lacking pixel values in the two-dimensional image specifically includes: determining the pixels lacking pixel values in the two-dimensional image; and determining the pixels lacking pixel values according to an image interpolation algorithm. The pixel points are subjected to interpolation processing to complement the pixel value of the pixel point lacking pixel value.

Among them, the image interpolation algorithm includes one of the nearest neighbor interpolation method, linear interpolation method and bilinear interpolation method.

Exemplarily, the steps of assigning values to pixels of pixels lacking pixel values in a two-dimensional image are specifically as follows:

Construct a Delaunay triangulation network according to the three-dimensional point cloud data; determine the height position information corresponding to the pixels lacking pixel values in the two-dimensional image according to the Delaunay triangulation network; and assign the determined height position information to The pixel points lacking pixel values in the two-dimensional image are used to complement the pixel values of the pixel points lacking pixel values.

Among them, the Delaunay Triangle includes multiple Delaunay Triangles. According to the Delaunay Triangulation, the height position information corresponding to the pixels lacking pixel values in the two-dimensional image is determined, and the determined height position information is assigned to the pixels lacking pixel values in the two-dimensional image.

Exemplarily, constructing a Delaunay triangulation based on three-dimensional point cloud data is specifically: constructing a plurality of triangles with the observation points in the three-dimensional point cloud data as vertices, and forming the Delaunay triangulation for the plurality of triangles; Wherein, there are no other observation points in the circumcircle of any of the triangles in the Delaunay triangulation network.

Among them, Figure 5 is a Delaunay triangulation constructed based on the obtained three-dimensional point cloud data. Take the observation points in the three-dimensional point cloud data as the vertices to construct multiple triangles, where the constructed triangles do not intersect each other and there are no other observation points in the circumcircle of any triangle. The constructed multiple triangles constitute Delaunay Triangulation.

In some embodiments, the step of the Delaunay triangulation determining the height position information corresponding to the pixel points lacking pixel values in the two-dimensional image is specifically:

Determine a target triangle in the Delaunay triangulation, where the target triangle is a triangle including the pixel points lacking pixel values; according to the target triangle, determine the corresponding pixel points in the two-dimensional image that lack pixel values Height position information.

The target triangle is a triangle including pixels lacking pixel values in the triangle plane. After the target triangle is determined, the height position information corresponding to the pixels lacking pixel values in the two-dimensional image can be determined according to the target triangle.

Exemplarily, the step of determining, according to the target triangle, the height position information corresponding to the pixel point lacking pixel value in the two-dimensional image is specifically:

Acquire first position information, second position information, and height position information of the three vertices of the target triangle; calculate the plane of the target triangle according to the first position information, second position information, and height position information of the three vertices Equation; determine the first position information and second position information corresponding to the pixel point lacking pixel value in the three-dimensional point cloud data; according to the plane equation and the first position corresponding to the pixel point lacking pixel value The information and the second position information calculate the height position information of the pixel point lacking pixel value.

Wherein, the calculation process of determining the first position information and the second position information corresponding to the pixel point lacking pixel value in the three-dimensional point cloud data may be the calculation of the matrix index corresponding to the observation point in the three-dimensional point cloud data The inverse operation of the calculation process will not be described in detail here.

After determining the target triangle, obtain the first position information, second position information, and height position information of the three vertices of the target triangle, and calculate the target triangle's position information according to the first position information, second position information, and height position information of the three vertices. Plane equation in ternary space. Substituting the first position information and the second position information of the pixels lacking pixel values into the ternary space plane equation, and solving the height position information of the pixels lacking pixel values. When the terrain changes drastically, reduce the factor The smooth distortion caused by pixel completion improves the accuracy of the calculated height position information of the pixels lacking pixel values.

Exemplarily, in order to quickly determine the height position information corresponding to the pixels lacking pixel values in the two-dimensional image, the step of determining the height position information corresponding to the pixels lacking pixel values in the two-dimensional image according to the target triangle is specifically for:

Calculate the distance between the pixel point lacking pixel value and the three vertices of the target triangle; determine the vertex closest to the pixel point lacking pixel value as the target vertex according to the distance, and determine the height position of the target vertex The information serves as height position information corresponding to the pixel point lacking pixel value.

Among them, after determining the target triangle, calculate the distance between the pixel point lacking pixel value and the three vertices of the target triangle, and use the vertex corresponding to the calculated minimum distance as the closest target to the pixel point lacking pixel value Vertex, the height position information of the target vertex is taken as the height position information of the pixel point lacking pixel value, and the calculation amount of the height position information is reduced when the terrain changes relatively smoothly.

When the number of target vertices corresponding to the calculated minimum distance is two or more, the average value of the height position information of the target vertices is calculated, and the calculated average value is used as the height position information of pixels lacking pixel values.

S104: Perform morphological processing on the two-dimensional image after the pixel complementation to obtain a processed two-dimensional image.

Among them, the morphological processing includes: closing operation, opening operation, closing operation first and then opening operation, and opening operation first and then closing operation. Morphological processing is performed on the two-dimensional image after pixel complementation, and the noise in the two-dimensional image is eliminated.

Exemplarily, the closing operation operation includes: performing an expansion operation operation first, and then performing an erosion operation operation. After the closing operation, the noise in the two-dimensional image is filtered out, that is, the noise in the three-dimensional point cloud data is filtered out.

In the specific implementation process, the calculation formula of the closed operation is as follows:

Among them, the kernel is the structural element (that is, the convolution kernel),

Is the expansion operator, and Θ is the corrosion operator. I _i is the first index value _{of the matrix index, and J i} is the second index value of the matrix index.

Exemplarily, the opening operation operation includes: first performing a corrosion operation operation, and then performing an expansion operation operation. After the open operation, the local lowest point is kept while removing the noise.

In some embodiments, the step of performing morphological processing on the two-dimensional image after the pixel complementation to obtain a processed two-dimensional image is specifically:

A plurality of convolution kernels of different sizes are selected to perform morphological processing on the pixels of the two-dimensional image after the pixel complementation, and different weights are assigned to the results of the calculation processing to obtain a two-dimensional image with noise removed.

Among them, due to morphological processing, that is, the effect of morphological filtering largely depends on the size of the selected convolution kernel, too small convolution kernel will retain too many obstacles and targets, and too large convolution kernel will cause The terrain is smooth and distorted. In order to eliminate the smooth distortion of terrain caused by morphological processing and solve the problem of retaining too many obstacles on the ground, a series of convolution kernels of different sizes can be used to perform morphological operations on the pixels in the two-dimensional image, and assign different results to the calculation results. To obtain the filtered pixel value of each pixel to obtain a two-dimensional image with noise removed.

For example, the convolution kernel used is {kernal ₁ ,kernal ₂ ,kernal ₃ ,...,kernal _n }, and different weights are assigned to the corresponding morphological operation results {w ₁ ,w ₂ ,w ₃ ,...,w _n }

The resulting filtered pixel value is:

Among them, kernal _i is the i-th convolution kernel, i=1, 2,...,n. Exemplarily, when performing morphological processing on a two-dimensional image after pixel complementation, for each pixel in the two-dimensional image after pixel complementation, the closing operation and/or the opening operation are Choose different convolution kernels.

Exemplarily, when performing morphological processing on a pixel-completion two-dimensional image, for each pixel in the pixel-completion two-dimensional image, the closing operation and/or the opening operation are Different convolution kernels are selected, and different weights are set for the operation processing results of different convolution kernels.

Wherein, the terrain information includes one or more of ground height, ground flatness, and ground slope.

In some embodiments, in order to quickly determine the terrain information, the step of determining the terrain information according to the processed two-dimensional image is specifically as follows:

S105: Reconstruct 3D point cloud data according to the processed 2D image.

S106: Determine terrain information according to the reconstructed three-dimensional point cloud data.

Among them, the 3D point cloud data is reconstructed according to the processed 2D image, and then the reconstructed 3D point cloud data is fitted to obtain a fitting plane. The ground height, ground slope, and ground flatness can be extracted by fitting the plane. information.

Exemplarily, the step of reconstructing three-dimensional point cloud data from a processed two-dimensional image is specifically: acquiring a matrix index corresponding to a pixel in the processed two-dimensional image; The pixel point undergoes coordinate conversion to obtain the first position information and the second position information of the reconstructed point corresponding to the pixel point; the pixel value of the pixel point is used as the height position information of the reconstructed point to complete a three-dimensional point cloud Reconstruction of data.

Among them, the matrix index and pixel value corresponding to the pixel in the two-dimensional image are obtained, and the coordinate conversion of the pixel is performed according to the matrix index to obtain the first position information and the second position information of the reconstructed point corresponding to the pixel, and then the pixel The value is used as the height position information of the reconstructed point to complete the reconstruction of the three-dimensional point cloud data.

In a specific implementation process, performing coordinate conversion on the pixel point according to the matrix index of the pixel point to obtain the first position information and the second position information of the reconstructed point corresponding to the pixel point includes: calculating the The product of the first index value of the pixel point and the distance resolution, and the product of the first index value and the distance resolution minus the difference of the maximum depth of field distance is taken as the first reconstruction point corresponding to the pixel point Position information; calculate the product of the second index value of the pixel and the distance resolution, and take the product of the second index value and the distance resolution minus the difference of the second position information as the difference with the pixel The second location information of the corresponding reconstruction point.

For example, the coordinates of the reconstructed point are (x _i , y _i ), where x _i is the first position information of the observation point, which is the depth of field detection distance, and x _i is the second position information of the observation point, which is the horizontal detection distance .

Reconstruction dot matrix corresponding to index (I _i, J _i), where, I _i is the first index value matrix index, J _i for the second index value of the index matrix. Then there are:

x _i ＝I _i *rL _x

y _i ＝J _i *rL _y

Among them, r is the range resolution, L _x is the maximum first position information, that is, the maximum depth of field detection distance, and _Ly is the maximum second position information, that is, the maximum horizontal detection distance.

In some embodiments, the step of determining terrain information according to the reconstructed three-dimensional point cloud data is specifically: fitting the reconstructed three-dimensional point cloud data to obtain a fitting plane, and determining the terrain information according to the fitting plane.

Among them, the 3D point cloud data is reconstructed according to the processed 2D image, and then the reconstructed 3D point cloud data is fitted to obtain a fitting plane. The ground height, ground slope, and ground flatness can be extracted by fitting the plane. information.

Exemplarily, the average value is calculated according to the height position information of the multiple reconstruction points in the fitting plane, and the ground flatness of the scanning area is determined according to the average value.

Exemplarily, the slope of the fitting plane is determined according to the height position information of multiple reconstruction points.

The foregoing embodiment obtains three-dimensional point cloud data containing terrain information; obtains a two-dimensional image corresponding to the three-dimensional point cloud data according to the three-dimensional point cloud data; performs processing on the pixels in the two-dimensional image that lack pixel values Pixel assignment is used to obtain a two-dimensional image after pixel complementation; morphological processing is performed on the two-dimensional image after pixel complementation to obtain a processed two-dimensional image; terrain information is determined according to the processed two-dimensional image. Based on digital image morphology processing, the three-dimensional point cloud data is projected into the two-dimensional image, and the two-dimensional image is morphologically filtered. The ground can be separated from the ground target while removing the noise, so as to realize the accurate estimation of the ground.

Please refer to FIG. 6, which is a schematic block diagram of a movable platform provided by an embodiment of the present application. The mobile platform 11 includes a processor 111, a memory 112, and a detection device 113. The processor 111, the memory 112, and the detection device 113 are connected by a bus, such as an I2C (Inter-integrated Circuit) bus or the detection device 113 and the processing device 113. The device 111 is connected via the CAN bus.

Among them, the movable platform includes aircraft, robots or autonomous unmanned vehicles.

Specifically, the processor 111 may be a micro-controller unit (MCU), a central processing unit (CPU), a digital signal processor (Digital Signal Processor, DSP), or the like.

Specifically, the memory 112 may be a Flash chip, a read-only memory (ROM, Read-Only Memory) disk, an optical disk, a U disk, or a mobile hard disk.

Specifically, the detection device 113 is used for terrain detection and collecting three-dimensional point cloud data containing terrain information.

Wherein, the processor is used to run a computer program stored in a memory, and implement the following steps when executing the computer program:

Obtain 3D point cloud data containing terrain information;

Obtaining a two-dimensional image corresponding to the three-dimensional point cloud data according to the three-dimensional point cloud data;

Pixel assignment is performed on pixels lacking pixel values in the two-dimensional image to obtain a two-dimensional image with pixel complementation;

Performing morphological processing on the two-dimensional image after the pixel complementation to obtain a processed two-dimensional image;

Reconstructing three-dimensional point cloud data according to the processed two-dimensional image; and

According to the reconstructed 3D point cloud data, the terrain information is determined.

In some embodiments, the processor implementing the step of obtaining a two-dimensional image corresponding to the three-dimensional point cloud data according to the three-dimensional point cloud data includes:

A two-dimensional matrix is determined according to the three-dimensional point cloud data, and the three-dimensional point cloud data is projected according to the two-dimensional matrix to obtain a two-dimensional image.

In some embodiments, the three-dimensional point cloud data includes a plurality of observation points, and each of the observation points includes first position information, second position information, and height position information, wherein the first position information and the second position information The location information is different.

In some embodiments, the processor implementing the step of determining a two-dimensional matrix based on the three-dimensional point cloud data includes:

Determining the first target location information and the second target location information according to the first location information and the second location information of the observation point in the three-dimensional point cloud data;

Obtain the distance resolution, and determine a two-dimensional matrix according to the first position information, the second position information, and the distance resolution.

In some embodiments, the processor implementing the step of determining the first target location information and the second target location information based on the first location information and the second location information of the observation points in the three-dimensional point cloud data includes:

The maximum first position information and the maximum second position information are determined from the first position information and the second position information of the observation point in the three-dimensional point cloud data, as the first target position information and the second target position information, respectively.

In some embodiments, the processor implementing the step of projecting the three-dimensional point cloud data according to the two-dimensional matrix to obtain a two-dimensional image includes:

Determine the matrix index corresponding to the observation point of the three-dimensional point cloud data in the two-dimensional matrix;

Assigning the height position information of the observation point to the matrix element corresponding to the matrix index of the observation point;

Taking the matrix index of the two-dimensional matrix as a pixel and taking the height position information corresponding to the matrix element of the two-dimensional matrix as the pixel value of the pixel, a two-dimensional image is obtained.

In some embodiments, the processor implementing the step of determining the matrix index corresponding to the observation point of the three-dimensional point cloud data in the two-dimensional matrix includes:

Calculate the sum of the first position information of the observation point and the largest first position information, and divide the sum of the first position information and the largest first position information of the observation point by the quotient of the distance resolution as the first matrix index An index value;

Calculate the sum of the second position information of the observation point and the largest second position information, and divide the sum of the second position information and the largest second position information of the observation point by the quotient of the distance resolution as the first matrix index Two index value.

In some embodiments, the processor implementing the step of assigning the height position information of the observation point to the matrix element corresponding to the matrix index of the observation point includes:

If there are multiple observation points corresponding to the same matrix index, determine the maximum height corresponding to the multiple observation points;

Assign the maximum value of the height to the matrix element corresponding to the same matrix index.

In some embodiments, the processor implementing the step of pixel assignment to pixels lacking pixel values in the two-dimensional image includes:

Determine the pixel points lacking pixel values in the two-dimensional image;

Perform interpolation processing on the pixel point lacking pixel value according to an image interpolation algorithm to complement the pixel value of the pixel point lacking pixel value.

In some embodiments, the image interpolation algorithm includes one of: nearest neighbor interpolation, linear interpolation, and bilinear interpolation.

In some embodiments, the processor implementing the step of pixel assignment to pixels lacking pixel values in the two-dimensional image includes:

Constructing a Delaunay triangulation network according to the three-dimensional point cloud data;

Determining, according to the Delaunay Triangulation, the height and position information corresponding to the pixels lacking pixel values in the two-dimensional image; and

The determined height position information is assigned to the pixel point lacking pixel value in the two-dimensional image to complete the pixel value of the pixel point lacking pixel value.

In some embodiments, the processor implementing the step of constructing Delaunay's triangle based on the three-dimensional point cloud data includes:

Constructing a plurality of triangles with the observation points in the three-dimensional point cloud data as vertices, and forming the Delaunay triangulation network;

Wherein, there are no other observation points in the circumcircle of any of the triangles in the Delaunay triangulation network.

In some embodiments, the processor implementing the step of determining, according to the Delaunay triangulation, the step of determining height position information corresponding to pixels lacking pixel values in the two-dimensional image includes:

Determining a target triangle in the Delaunay Triangulation, where the target triangle is a triangle including the pixel points lacking pixel values;

The height position information corresponding to the pixel point lacking pixel value in the two-dimensional image is determined according to the target triangle.

In some embodiments, the processor implementing the step of determining height position information corresponding to pixels lacking pixel values in the two-dimensional image according to the target triangle includes:

Acquiring first position information, second position information, and height position information of the three vertices of the target triangle;

Calculating the plane equation of the target triangle according to the first position information, the second position information, and the height position information of the three vertices;

Determining first position information and second position information corresponding to the pixel point lacking pixel value under the three-dimensional point cloud data;

The height position information of the pixel point lacking pixel value is calculated according to the plane equation and the first position information and second position information corresponding to the pixel point lacking pixel value.

In some embodiments, the processor implementing the step of determining height position information corresponding to pixels lacking pixel values in the two-dimensional image according to the target triangle includes:

Calculating the distance between the pixel point lacking pixel value and the three vertices of the target triangle;

The vertex closest to the pixel point lacking pixel value is determined as the target vertex according to the distance, and the height position information of the target vertex is used as the height position information corresponding to the pixel point lacking pixel value.

In some embodiments, the morphological processing includes one of a closing operation, an opening operation, a closing operation first and then an opening operation, and an opening operation before the closing operation.

In some embodiments, the closing operation includes: performing an expansion operation first, and then performing an erosion operation; or, the opening operation includes: performing an erosion operation first, and then an expansion operation.

In some embodiments, the step of performing morphological processing on the two-dimensional image after pixel complementation by the processor to obtain a processed two-dimensional image includes:

A plurality of convolution kernels of different sizes are selected to perform morphological processing on the pixels of the two-dimensional image after the pixel complementation, and different weights are assigned to the results of the calculation processing to obtain a two-dimensional image with noise removed.

In some embodiments, for each pixel in the two-dimensional image after the pixel complementation, a different convolution kernel is selected for the closing operation and/or the opening operation.

In some embodiments, for each pixel in the pixel-completion two-dimensional image, the closing operation and/or the opening operation both select different convolution kernels, and for different convolution kernels Different weights are set for the operation processing results of.

In some embodiments, the processor implementing the step of reconstructing three-dimensional point cloud data from the processed two-dimensional image includes:

Acquiring a matrix index corresponding to a pixel in the processed two-dimensional image;

Performing coordinate conversion on the pixel point according to the matrix index of the pixel point to obtain the first position information and the second position information of the reconstructed point corresponding to the pixel point;

The pixel value of the pixel point is used as the height position information of the reconstructed point to complete the reconstruction of the three-dimensional point cloud data.

In some embodiments, the processor implements the coordinate conversion of the pixel point according to the matrix index of the pixel point to obtain the first position information and the second position of the reconstructed point corresponding to the pixel point The information steps include:

Calculate the product of the first index value of the pixel point and the distance resolution, and use the product of the first index value and the distance resolution minus the difference of the maximum depth of field distance as the reconstruction point corresponding to the pixel point ’S first location information;

Calculate the product of the second index value of the pixel point and the distance resolution, and use the product of the second index value and the distance resolution minus the difference of the second position information as the reconstruction corresponding to the pixel point The second location information of the point.

In some embodiments, the processor implementing the step of determining terrain information according to the reconstructed three-dimensional point cloud data includes:

Fitting the reconstructed three-dimensional point cloud data to obtain a fitting plane, and determining terrain information according to the fitting plane.

In some embodiments, the terrain information includes one or more of ground height, ground flatness, and ground slope.

Please refer to FIG. 7, which is a schematic block diagram of a control device provided by an embodiment of the present application. The control device 12 includes a processor 121 and a memory 122, and the processor 121 and the memory 122 are connected by a bus, such as an I2C (Inter-integrated Circuit) bus.

Specifically, the processor 121 may be a micro-controller unit (MCU), a central processing unit (CPU), a digital signal processor (Digital Signal Processor, DSP), or the like.

Specifically, the memory 122 may be a Flash chip, a read-only memory (ROM, Read-Only Memory) disk, an optical disk, a U disk, or a mobile hard disk, etc. The memory 122 is used to store computer programs.

Wherein, the processor is used to run a computer program stored in a memory, and implement the following steps when executing the computer program:

Obtain 3D point cloud data containing terrain information;

Obtaining a two-dimensional image corresponding to the three-dimensional point cloud data according to the three-dimensional point cloud data;

Pixel assignment is performed on pixels lacking pixel values in the two-dimensional image to obtain a two-dimensional image with pixel complementation;

Performing morphological processing on the two-dimensional image after the pixel complementation to obtain a processed two-dimensional image;

Reconstructing three-dimensional point cloud data according to the processed two-dimensional image; and

According to the reconstructed 3D point cloud data, the terrain information is determined, and the determined terrain information is sent to the movable platform.

The embodiment of the present application also provides a control system, which may be, for example, the flight control system shown in FIG. 1. The control system includes a movable platform and a control device, and the control device is communicatively connected with the movable platform;

The movable platform is used to collect three-dimensional point cloud data and send the three-dimensional point cloud data to the control device.

The embodiments of the present application also provide a computer-readable storage medium, the computer-readable storage medium stores a computer program, the computer program includes program instructions, and the processor executes the program instructions to implement the foregoing implementation The steps of the terrain detection method provided in the example.

The computer-readable storage medium may be the internal storage unit of the removable platform and the control device described in any of the foregoing embodiments, for example, the hard disk or memory of the control device. The computer-readable storage medium may also be an external storage device of the control device, such as a plug-in hard disk equipped on the control device, a smart memory card (Smart Media Card, SMC), and a Secure Digital (SD) ) Card, Flash Card, etc.

The above are only specific implementations of this application, but the protection scope of this application is not limited to this. Anyone familiar with the technical field can easily think of various equivalents within the technical scope disclosed in this application. Modifications or replacements, these modifications or replacements shall be covered within the scope of protection of this application. Therefore, the protection scope of this application should be subject to the protection scope of the claims.

Claims (51)

1. A terrain detection method, characterized in that it comprises:

   Obtain 3D point cloud data containing terrain information;

   Obtaining a two-dimensional image corresponding to the three-dimensional point cloud data according to the three-dimensional point cloud data;

   Pixel assignment is performed on pixels lacking pixel values in the two-dimensional image to obtain a two-dimensional image with pixel complementation;

   Performing morphological processing on the two-dimensional image after the pixel complementation to obtain a processed two-dimensional image;

   Reconstructing three-dimensional point cloud data according to the processed two-dimensional image; and

   According to the reconstructed 3D point cloud data, the terrain information is determined.
2. The terrain detection method according to claim 1, wherein the obtaining a two-dimensional image corresponding to the three-dimensional point cloud data according to the three-dimensional point cloud data comprises:

   A two-dimensional matrix is determined according to the three-dimensional point cloud data, and the three-dimensional point cloud data is projected according to the two-dimensional matrix to obtain a two-dimensional image.
3. The terrain detection method according to claim 2, wherein the three-dimensional point cloud data includes a plurality of observation points, and each of the observation points includes first position information, second position information, and height position information, wherein, The first location information is different from the second location information.
4. The terrain detection method according to claim 3, wherein the determining a two-dimensional matrix according to the three-dimensional point cloud data comprises:

   Determining the first target location information and the second target location information according to the first location information and the second location information of the observation point in the three-dimensional point cloud data;

   Obtain the distance resolution, and determine a two-dimensional matrix according to the first position information, the second position information, and the distance resolution.
5. The terrain detection method according to claim 4, wherein the first target location information and the second target location information are determined according to the first location information and the second location information of the observation points in the three-dimensional point cloud data, include:

   The maximum first position information and the maximum second position information are determined from the first position information and the second position information of the observation point in the three-dimensional point cloud data, as the first target position information and the second target position information, respectively.
6. The terrain detection method according to claim 3, wherein the projecting the three-dimensional point cloud data according to the two-dimensional matrix to obtain a two-dimensional image comprises:

   Determine the matrix index corresponding to the observation point of the three-dimensional point cloud data in the two-dimensional matrix;

   Assigning the height position information of the observation point to the matrix element corresponding to the matrix index of the observation point;

   Taking the matrix index of the two-dimensional matrix as a pixel and taking the height position information corresponding to the matrix element of the two-dimensional matrix as the pixel value of the pixel, a two-dimensional image is obtained.
7. The terrain detection method according to claim 6, wherein the determining the matrix index corresponding to the observation point of the three-dimensional point cloud data in the two-dimensional matrix comprises:

   Calculate the sum of the first position information of the observation point and the largest first position information, and divide the sum of the first position information and the largest first position information of the observation point by the quotient of the distance resolution as the first matrix index An index value;

   Calculate the sum of the second position information of the observation point and the largest second position information, and divide the sum of the second position information and the largest second position information of the observation point by the quotient of the distance resolution as the first matrix index Two index value.
8. The terrain detection method according to claim 6, wherein the assigning the height position information of the observation point to the matrix element corresponding to the matrix index of the observation point comprises:

   If there are multiple observation points corresponding to the same matrix index, determine the maximum height corresponding to the multiple observation points;

   Assign the maximum value of the height to the matrix element corresponding to the same matrix index.
9. The terrain detection method according to any one of claims 1 to 8, wherein the pixel assignment of pixels in the two-dimensional image that lack pixel values includes:

   Determine the pixel points lacking pixel values in the two-dimensional image;

   Perform interpolation processing on the pixel point lacking pixel value according to an image interpolation algorithm to complement the pixel value of the pixel point lacking pixel value.
10. The terrain detection method according to claim 9, wherein the image interpolation algorithm comprises one of: nearest neighbor interpolation method, linear interpolation method, and bilinear interpolation method.
11. The terrain detection method according to any one of claims 1 to 8, wherein the pixel assignment of pixels in the two-dimensional image that lack pixel values includes:

    Constructing a Delaunay triangulation network according to the three-dimensional point cloud data;

    Determining, according to the Delaunay Triangulation, the height and position information corresponding to the pixels lacking pixel values in the two-dimensional image; and

    The determined height position information is assigned to the pixel point lacking pixel value in the two-dimensional image to complete the pixel value of the pixel point lacking pixel value.
12. The terrain detection method according to claim 11, wherein said constructing a Delaunay triangulation network according to the three-dimensional point cloud data comprises:

    Constructing a plurality of triangles with the observation points in the three-dimensional point cloud data as vertices, and forming the Delaunay triangulation network;

    Wherein, there are no other observation points in the circumcircle of any of the triangles in the Delaunay triangulation network.
13. The terrain detection method according to claim 11, wherein the determining height position information corresponding to pixels lacking pixel values in the two-dimensional image according to the Delaunay triangulation network comprises:

    Determining a target triangle in the Delaunay Triangulation, where the target triangle is a triangle including the pixel points lacking pixel values;

    The height position information corresponding to the pixel point lacking pixel value in the two-dimensional image is determined according to the target triangle.
14. The terrain detection method according to claim 13, wherein the determining, according to the target triangle, the height and position information corresponding to the pixels lacking pixel values in the two-dimensional image comprises:

    Acquiring first position information, second position information, and height position information of the three vertices of the target triangle;

    Calculating the plane equation of the target triangle according to the first position information, the second position information, and the height position information of the three vertices;

    Determining first position information and second position information corresponding to the pixel point lacking pixel value under the three-dimensional point cloud data;

    The height position information of the pixel point lacking pixel value is calculated according to the plane equation and the first position information and second position information corresponding to the pixel point lacking pixel value.
15. The terrain detection method according to claim 13, wherein the determining, according to the target triangle, the height and position information corresponding to the pixels lacking pixel values in the two-dimensional image comprises:

    Calculating the distance between the pixel point lacking pixel value and the three vertices of the target triangle;

    The vertex closest to the pixel point lacking pixel value is determined as the target vertex according to the distance, and the height position information of the target vertex is used as the height position information corresponding to the pixel point lacking pixel value.
16. The terrain detection method according to any one of claims 1 to 8, wherein the morphological processing includes: closing operation, opening operation, closing operation before opening operation, and opening operation first The operation then performs one of the closed operations.
17. The terrain detection method according to claim 16, wherein the closing operation includes: performing an expansion operation first, and then performing an erosion operation; or, the opening operation includes: performing an erosion operation first, and then Perform expansion operations.
18. The terrain detection method according to claim 16, wherein the performing morphological processing on the two-dimensional image after the pixel complementation to obtain the processed two-dimensional image comprises:

    A plurality of convolution kernels of different sizes are selected to perform morphological processing on the pixels of the two-dimensional image after the pixel complementation, and different weights are assigned to the results of the calculation processing to obtain a two-dimensional image with noise removed.
19. The terrain detection method according to claim 18, characterized in that, for each pixel in the two-dimensional image after the pixel complementation, a different volume is selected for the closing operation and/or the opening operation. Product core.
20. The terrain detection method according to claim 18, characterized in that, for each pixel in the two-dimensional image after the pixel complementation, a different volume is selected for the closing operation and/or the opening operation. Convolution kernels, and different weights are set for the operation processing results of different convolution kernels.
21. The terrain detection method according to claim 6, wherein the reconstruction of 3D point cloud data according to the processed 2D image comprises:

    Acquiring a matrix index corresponding to a pixel in the processed two-dimensional image;

    Performing coordinate conversion on the pixel point according to the matrix index of the pixel point to obtain the first position information and the second position information of the reconstructed point corresponding to the pixel point;

    The pixel value of the pixel point is used as the height position information of the reconstructed point to complete the reconstruction of the three-dimensional point cloud data.
22. 22. The terrain detection method according to claim 21, wherein the coordinate conversion is performed on the pixel point according to the matrix index of the pixel point to obtain the first position information of the reconstructed point corresponding to the pixel point And second location information, including:

    Calculate the product of the first index value of the pixel point and the distance resolution, and use the product of the first index value and the distance resolution minus the difference of the maximum depth of field distance as the reconstruction point corresponding to the pixel point ’S first location information;

    Calculate the product of the second index value of the pixel point and the distance resolution, and use the product of the second index value and the distance resolution minus the difference of the second position information as the reconstruction corresponding to the pixel point The second location information of the point.
23. The terrain detection method according to claim 6, wherein the determining the terrain information according to the reconstructed three-dimensional point cloud data comprises:

    Fitting the reconstructed three-dimensional point cloud data to obtain a fitting plane, and determining terrain information according to the fitting plane.
24. The terrain detection method according to claim 23, wherein the terrain information includes one or more of ground height, ground flatness, and ground slope.
25. A movable platform, characterized in that, the movable platform includes a detection device, a memory, and a processor;

    The detection device is used for terrain detection and collecting three-dimensional point cloud data containing terrain information;

    The memory is used to store a computer program;

    The processor is configured to execute the computer program and, when executing the computer program, implement the following steps:

    Obtain 3D point cloud data containing terrain information;

    Obtaining a two-dimensional image corresponding to the three-dimensional point cloud data according to the three-dimensional point cloud data;

    Pixel assignment is performed on pixels lacking pixel values in the two-dimensional image to obtain a two-dimensional image with pixel complementation;

    Performing morphological processing on the two-dimensional image after the pixel complementation to obtain a processed two-dimensional image;

    Reconstructing three-dimensional point cloud data according to the processed two-dimensional image; and

    According to the reconstructed 3D point cloud data, the terrain information is determined.
26. The mobile platform according to claim 25, wherein the step of the processor implementing the step of obtaining a two-dimensional image corresponding to the three-dimensional point cloud data according to the three-dimensional point cloud data comprises:

    A two-dimensional matrix is determined according to the three-dimensional point cloud data, and the three-dimensional point cloud data is projected according to the two-dimensional matrix to obtain a two-dimensional image.
27. The mobile platform of claim 26, wherein the three-dimensional point cloud data includes a plurality of observation points, and each of the observation points includes first position information, second position information, and height position information, wherein, The first location information is different from the second location information.
28. The mobile platform according to claim 27, wherein the processor implementing the step of determining a two-dimensional matrix according to the three-dimensional point cloud data comprises:

    Determining the first target location information and the second target location information according to the first location information and the second location information of the observation point in the three-dimensional point cloud data;

    Obtain the distance resolution, and determine a two-dimensional matrix according to the first position information, the second position information, and the distance resolution.
29. The mobile platform of claim 28, wherein the processor implements the determination of the first target location information and the second location information according to the first location information and the second location information of the observation points in the three-dimensional point cloud data. 2. The steps of target location information include:

    The maximum first position information and the maximum second position information are determined from the first position information and the second position information of the observation point in the three-dimensional point cloud data, as the first target position information and the second target position information, respectively.
30. The movable platform according to claim 27, wherein the processor implements the step of projecting the three-dimensional point cloud data according to the two-dimensional matrix to obtain a two-dimensional image, comprising:

    Determine the matrix index corresponding to the observation point of the three-dimensional point cloud data in the two-dimensional matrix;

    Assigning the height position information of the observation point to the matrix element corresponding to the matrix index of the observation point;

    Taking the matrix index of the two-dimensional matrix as a pixel and taking the height position information corresponding to the matrix element of the two-dimensional matrix as the pixel value of the pixel, a two-dimensional image is obtained.
31. The movable platform according to claim 30, wherein the processor implements the step of determining the matrix index corresponding to the observation point of the three-dimensional point cloud data in the two-dimensional matrix, comprising:

    Calculate the sum of the first position information of the observation point and the largest first position information, and divide the sum of the first position information and the largest first position information of the observation point by the quotient of the distance resolution as the first matrix index An index value;

    Calculate the sum of the second position information of the observation point and the largest second position information, and divide the sum of the second position information and the largest second position information of the observation point by the quotient of the distance resolution as the first matrix index Two index value.
32. The movable platform according to claim 30, wherein the processor implements the step of assigning the height position information of the observation point to the matrix element corresponding to the matrix index of the observation point, comprising:

    If there are multiple observation points corresponding to the same matrix index, determine the maximum height corresponding to the multiple observation points;

    Assign the maximum value of the height to the matrix element corresponding to the same matrix index.
33. The movable platform according to any one of claims 25 to 32, wherein the processor implements the step of performing pixel assignment on pixels lacking pixel values in the two-dimensional image, comprising:

    Determine the pixel points lacking pixel values in the two-dimensional image;

    Perform interpolation processing on the pixel point lacking pixel value according to an image interpolation algorithm to complement the pixel value of the pixel point lacking pixel value.
34. The movable platform according to claim 33, wherein the image interpolation algorithm includes one of: nearest neighbor interpolation method, linear interpolation method, and bilinear interpolation method.
35. The movable platform according to any one of claims 25 to 32, wherein the processor implements the step of performing pixel assignment on pixels lacking pixel values in the two-dimensional image, comprising:

    Constructing a Delaunay triangulation network according to the three-dimensional point cloud data;

    Determining, according to the Delaunay Triangulation, the height and position information corresponding to the pixels lacking pixel values in the two-dimensional image; and

    The determined height position information is assigned to the pixel point lacking pixel value in the two-dimensional image to complete the pixel value of the pixel point lacking pixel value.
36. The mobile platform according to claim 35, wherein the processor implementing the step of constructing Delaunay triangle according to the three-dimensional point cloud data comprises:

    Constructing a plurality of triangles with the observation points in the three-dimensional point cloud data as vertices, and forming the Delaunay triangulation network;

    Wherein, there are no other observation points in the circumcircle of any of the triangles in the Delaunay triangulation network.
37. The movable platform according to claim 35, wherein the processor implements the step of determining, according to the Delaunay triangulation, the height position information corresponding to the pixels lacking pixel values in the two-dimensional image ,include:

    Determining a target triangle in the Delaunay Triangulation, where the target triangle is a triangle including the pixel points lacking pixel values;

    The height position information corresponding to the pixel point lacking pixel value in the two-dimensional image is determined according to the target triangle.
38. The movable platform according to claim 37, wherein the step of the processor implementing the step of determining, according to the target triangle, height position information corresponding to pixels lacking pixel values in the two-dimensional image comprises:

    Acquiring first position information, second position information, and height position information of the three vertices of the target triangle;

    Calculating the plane equation of the target triangle according to the first position information, the second position information, and the height position information of the three vertices;

    Determining first position information and second position information corresponding to the pixel point lacking pixel value under the three-dimensional point cloud data;

    The height position information of the pixel point lacking pixel value is calculated according to the plane equation and the first position information and second position information corresponding to the pixel point lacking pixel value.
39. The movable platform according to claim 37, wherein the step of the processor implementing the step of determining, according to the target triangle, height position information corresponding to pixels lacking pixel values in the two-dimensional image comprises:

    Calculating the distance between the pixel point lacking pixel value and the three vertices of the target triangle;

    The vertex closest to the pixel point lacking pixel value is determined as the target vertex according to the distance, and the height position information of the target vertex is used as the height position information corresponding to the pixel point lacking pixel value.
40. The mobile platform according to any one of claims 25 to 32, wherein the morphological processing includes: closing operation, opening operation, closing operation first, then opening operation, and opening operation first The operation then performs one of the closed operations.
41. The mobile platform according to claim 40, wherein the closing operation includes: performing an expansion operation first, and then performing an erosion operation; or, the opening operation includes: performing an erosion operation first, and then Perform expansion operations.
42. The movable platform according to claim 40, wherein the processor implements the step of performing morphological processing on the two-dimensional image after the pixel complementation to obtain the processed two-dimensional image, comprising:

    A plurality of convolution kernels of different sizes are selected to perform morphological processing on the pixels of the two-dimensional image after the pixel complementation, and different weights are assigned to the results of the calculation processing to obtain a two-dimensional image with noise removed.
43. The movable platform according to claim 42, wherein for each pixel in the two-dimensional image after the pixel complementation, the closing operation and/or the opening operation both select different volumes. Product core.
44. The movable platform according to claim 42, wherein for each pixel in the two-dimensional image after the pixel complementation, the closing operation and/or the opening operation both select different volumes. Convolution kernels, and different weights are set for the operation processing results of different convolution kernels.
45. The mobile platform according to claim 30, wherein the processor implementing the step of reconstructing three-dimensional point cloud data from the processed two-dimensional image comprises:

    Acquiring a matrix index corresponding to a pixel in the processed two-dimensional image;

    Performing coordinate conversion on the pixel point according to the matrix index of the pixel point to obtain the first position information and the second position information of the reconstructed point corresponding to the pixel point;

    The pixel value of the pixel point is used as the height position information of the reconstructed point to complete the reconstruction of the three-dimensional point cloud data.
46. The movable platform according to claim 45, wherein the processor implements the coordinate conversion of the pixel point according to the matrix index of the pixel point to obtain a reconstruction point corresponding to the pixel point The steps of the first location information and the second location information include:

    Calculate the product of the first index value of the pixel point and the distance resolution, and use the product of the first index value and the distance resolution minus the difference of the maximum depth of field distance as the reconstruction point corresponding to the pixel point ’S first location information;

    Calculate the product of the second index value of the pixel point and the distance resolution, and use the product of the second index value and the distance resolution minus the difference of the second position information as the reconstruction corresponding to the pixel point The second location information of the point.
47. The movable platform according to claim 30, wherein the step of determining the terrain information according to the reconstructed three-dimensional point cloud data by the processor comprises:

    Fitting the reconstructed three-dimensional point cloud data to obtain a fitting plane, and determining terrain information according to the fitting plane.
48. The movable platform according to claim 47, wherein the terrain information includes one or more of ground height, ground flatness, and ground slope.
49. A control device, characterized in that the control device includes a memory and a processor;

    The memory is used to store a computer program;

    The processor is configured to execute the computer program and, when executing the computer program, implement the steps of the terrain detection method according to any one of claims 1-24, and send the determined terrain information to the movable platform.
50. A control system, characterized in that, the control system comprises a movable platform and the control equipment according to claim 49; wherein, the movable platform is used to collect three-dimensional point cloud data and to combine the terrain information The three-dimensional point cloud data is sent to the control device.
51. A computer-readable storage medium, characterized in that, the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the processor realizes as described in any one of claims 1 to 24. The terrain detection method described.

Images