Target object detection method, system, device and storage medium

Abstract

A method, system, device and storage medium for detecting a target object. A point cloud cluster corresponding to the target object is obtained by clustering a three-dimensional point cloud detected by a detection device mounted on a movable platform. During the clustering process, the height of the cluster center of the point cloud cluster needs to meet a preset height condition. Further, a target detection model is determined based on the distance of the target object relative to the movable platform and the correspondence between the distance and the detection model. The point cloud cluster corresponding to the target object is detected by the target detection model so that the target detection model determines the object type of the target object. That is, different detection models are used to detect target objects at different distances from the movable platform, thereby improving the detection accuracy of the target object.

Description

Technical Field

[0001] The embodiments of the present application relate to the field of mobile platforms, and in particular, to a method, system, device and storage medium for detecting a target object. Background Art

[0002] In an autonomous driving system or an assisted driving system, vehicles on the road need to be detected in order to avoid them.

[0003] In the prior art, an automatic driving system or an assisted driving system is usually provided with a photographing device, and surrounding vehicles are detected through two-dimensional images captured by the photographing device. However, the accuracy of vehicle detection is insufficient when only the two-dimensional images are used to detect surrounding vehicles. Summary of the invention

[0004] The embodiments of the present application provide a method, system, device and storage medium for detecting a target object to improve the detection accuracy of the target object.

[0005] A first aspect of an embodiment of the present application is to provide a method for detecting a target object, which is applied to a movable platform, wherein the movable platform is provided with a detection device, and the detection device is used to detect the surrounding environment of the movable platform to obtain a three-dimensional point cloud. The method includes:

[0006] Acquire the three-dimensional point cloud;

[0007] Clustering the three-dimensional point cloud to obtain a point cloud cluster corresponding to the first target object, wherein the height of the cluster center of the clustered point cloud cluster meets a preset height condition;

[0008] Determining a target detection model according to a distance of the first target object relative to the movable platform and a corresponding relationship between the distance and the detection model;

[0009] The point cloud cluster corresponding to the first target object is detected by the target detection model to determine the object type of the first target object.

[0010] A second aspect of an embodiment of the present application is to provide a target object detection system, including: a detection device, a memory, and a processor;

[0011] The detection device is used to detect the surrounding environment of the movable platform to obtain a three-dimensional point cloud;

[0012] The memory is used to store program code;

[0013] The processor calls the program code, and when the program code is executed, is used to perform the following operations:

[0014] Acquire the three-dimensional point cloud;

[0015] Clustering the three-dimensional point cloud to obtain a point cloud cluster corresponding to the first target object, wherein the height of the cluster center of the clustered point cloud cluster meets a preset height condition;

[0016] Determining a target detection model according to a distance of the first target object relative to the movable platform and a corresponding relationship between the distance and the detection model;

[0017] The point cloud cluster corresponding to the first target object is detected by the target detection model to determine the object type of the first target object.

[0018] A third aspect of the embodiments of the present application is to provide a movable platform, including:

[0019] body;

[0020] A power system, installed on the fuselage, for providing movement power;

[0021] And a target object detection system as described in the second aspect.

[0022] A fourth aspect of the embodiments of the present application is to provide a computer-readable storage medium on which a computer program is stored, and the computer program is executed by a processor to implement the method described in the first aspect.

[0023] The target object detection method, system, device and storage medium provided in the present embodiment obtain a point cloud cluster corresponding to the target object by clustering the three-dimensional point cloud obtained by the detection device carried on the movable platform. During the clustering process, the height of the cluster center of the point cloud cluster needs to meet a preset height condition. Furthermore, according to the distance of the target object relative to the movable platform and the correspondence between the distance and the detection model, the target detection model is determined, and the point cloud cluster corresponding to the target object is detected by the target detection model, so that the target detection model can determine the object type of the target object. That is to say, different detection models are used to detect target objects at different distances relative to the movable platform, thereby improving the detection accuracy of the target object. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following is a brief introduction to the drawings required for use in the description of the embodiments. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0025] Figure 1 A schematic diagram of an application scenario provided in an embodiment of the present application;

[0026] Figure 2A flow chart of a method for detecting a target object provided in an embodiment of the present application;

[0027] Figure 3 A schematic diagram of another application scenario provided by an embodiment of the present application;

[0028] Figure 4 A schematic diagram of another application scenario provided by an embodiment of the present application;

[0029] Figure 5 A schematic diagram of a detection model provided in an embodiment of the present application;

[0030] Figure 6 A flowchart of a method for detecting a target object provided by another embodiment of the present application;

[0031] Figure 7 A schematic diagram of projecting a three-dimensional point cloud onto a two-dimensional image provided in an embodiment of the present application;

[0032] Figure 8 A schematic diagram of a two-dimensional feature point provided in an embodiment of the present application;

[0033] Fig. 9 A flowchart of a method for detecting a target object provided by another embodiment of the present application;

[0034] Fig.10 A schematic diagram of a three-dimensional point cloud provided in an embodiment of the present application;

[0035] Fig.11 A schematic diagram of another three-dimensional point cloud provided in an embodiment of the present application;

[0036] Fig.12 A schematic diagram of another three-dimensional point cloud provided in an embodiment of the present application;

[0037] Fig.13 A structural diagram of a target object detection system provided in an embodiment of the present application.

[0038] Reference numerals:

[0039] 11: Vehicle; 12: Server; 13: Vehicle;

[0040] 14: vehicle; 15: 3D point cloud; 30: ground point cloud;

[0041] 31: point cloud cluster; 32: point cloud cluster;

[0042] 41: first target object; 42: first target object; 80: first image;

[0043] 81: projection area; 82: two-dimensional feature points; 1001: right area;

[0044] 1002: upper left corner image; 1003: lower left corner image; 100: white arc;

[0045] 101: first target object; 102: first target object;

[0046] 103: first target object; 104: three-dimensional point cloud;

[0047] 105: circle; 106: identification box; 130: target object detection system;

[0048] 131: detection device; 132: memory; 133: processor. DETAILED DESCRIPTION

[0049] The following will be combined with the drawings in the embodiments of the present application to clearly describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0050] It should be noted that when a component is said to be "fixed to" another component, it can be directly on the other component or there can be a central component. When a component is said to be "connected to" another component, it can be directly connected to the other component or there can be a central component at the same time.

[0051] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art to which this application belongs. The terms used herein in the specification of this application are only for the purpose of describing specific embodiments and are not intended to limit this application. The term "and / or" used herein includes any and all combinations of one or more of the related listed items.

[0052] In conjunction with the accompanying drawings, some embodiments of the present application are described in detail below. In the absence of conflict, the following embodiments and features in the embodiments can be combined with each other.

[0053] The embodiment of the present application provides a method for detecting a target object. The method is applied to a movable platform, and the movable platform is provided with a detection device, and the detection device is used to detect the surrounding environment of the movable platform to obtain a three-dimensional point cloud. In this embodiment, the movable platform can be a drone, a mobile robot or a vehicle.

[0054] In the embodiment of the present application, the movable platform is a vehicle as an example. The vehicle may be an unmanned vehicle or a vehicle equipped with an Advanced Driver Assistance Systems (ADAS) system. Figure 1 As shown, the vehicle 11 is a carrier equipped with a detection device, which can be a binocular stereo camera, a time of flight (TOF) camera and / or a laser radar. When the vehicle 11 is driving, the detection device detects the surrounding environment of the vehicle 11 in real time to obtain a three-dimensional point cloud. The surrounding environment of the vehicle 11 includes objects around the vehicle 11. Among them, the objects around the vehicle 11 include the ground, pedestrians, vehicles, etc. around the vehicle 11.

[0055] Take laser radar as an example. When a laser beam emitted by the laser radar hits the surface of an object, the surface of the object will reflect the laser beam. The laser radar can determine the position, distance and other information of the object relative to the laser radar based on the laser reflected from the surface of the object. If the laser beam emitted by the laser radar is scanned along a certain trajectory, such as a 360-degree rotation scan, a large number of laser points will be obtained, thus forming the laser point cloud data of the object, that is, a three-dimensional point cloud.

[0056] In addition, this embodiment does not limit the execution subject of the target object detection method. The target object detection method can be executed by an on-board device in the vehicle, or by other devices with data processing functions other than the on-board device, for example, Figure 1 The server 12 shown, the vehicle 11 and the server 12 can perform wireless communication or wired communication, and the vehicle 11 can send the three-dimensional point cloud detected by the detection device to the server 12, and the server 12 executes the detection method of the target object. The following introduces the target object detection method provided in the embodiment of the present application by taking the vehicle-mounted device as an example. Among them, the vehicle-mounted device can be a device with data processing function integrated in the vehicle center console, or it can also be a tablet computer, mobile phone, laptop computer, etc. placed in the vehicle.

[0057] Figure 2 Flow chart of the target object detection method provided in the embodiment of the present application. Figure 2 As shown, the method in this embodiment may include:

[0058] S201: Acquire the three-dimensional point cloud.

[0059] like Figure 1As shown, during the driving process of the vehicle 11, the detection device carried by the vehicle 11 detects the surrounding environment of the vehicle 11 in real time to obtain a three-dimensional point cloud, and the detection device can be connected to the vehicle-mounted device on the vehicle 11 in communication, so that the vehicle-mounted device on the vehicle 11 can obtain the three-dimensional point cloud detected by the detection device in real time. For example, the three-dimensional point cloud of the ground around the vehicle 11, the three-dimensional point cloud of pedestrians, and the three-dimensional point cloud of other vehicles such as the vehicle 13 and the vehicle 14.

[0060] S202: Clustering the three-dimensional point cloud to obtain a point cloud cluster corresponding to the first target object, wherein the height of the cluster center of the clustered point cloud cluster meets a preset height condition.

[0061] like Figure 3 As shown, the three-dimensional point cloud 15 is a three-dimensional point cloud detected by the detection equipment carried by the vehicle 11. The three-dimensional point cloud 15 includes multiple three-dimensional points, that is, the three-dimensional point cloud is a collection composed of many three-dimensional points. In addition, three-dimensional points can also be called point cloud points. Since the point cloud points in the three-dimensional point cloud detected and acquired by the detection equipment at each sampling moment carry position information, the position information can specifically be the three-dimensional coordinates of the point cloud point in the three-dimensional coordinate system. This embodiment does not limit the three-dimensional coordinate system. For example, the three-dimensional coordinate system can specifically be a vehicle body coordinate system, an earth coordinate system, or a world coordinate system. Therefore, according to the position information of each point cloud point, the height of each point cloud point relative to the ground can be determined.

[0062] In the process of clustering the three-dimensional point cloud 15, the k-means clustering algorithm can be used to weight the point cloud points in the three-dimensional point cloud 15 whose height from the ground is close to the preset height, so that the height value of the cluster center is close to the preset height, which is recorded as Where H represents the height of the vehicle. Usually, the height of a car is about 1.6 meters, and the height of a large vehicle such as a bus is about 3 meters. Here, the vehicle height H can be set to 1.1 meters. Alternatively, H can have two values, one is H1 = 0.8 meters, and the other is H2 = 1.5 meters. By clustering H1 and H2 respectively, the height value of the cluster center is close to The height values ​​of the clusters and cluster centers are close Clustering. Here, taking the value of H as 1.1 meters as an example, assuming that P1 and P2 are any two 3D points in the 3D point cloud 15, respectively, P1 and P2 respectively correspond to a 3D coordinate, where the coordinate of P1 on the z-axis, i.e., the height direction, can be recorded as P1(z), and the coordinate of P2 on the z-axis, i.e., the height direction, can be recorded as P2(z). If the function value Loss calculated by the following formula (1) is less than or equal to a certain threshold, it is determined that P1 and P2 can be aggregated into one cluster.

[0063]

[0064] Wherein, k can be a constant. It is understandable that when clustering the 3D point cloud 15, the aggregation process between different 3D points in the 3D point cloud 15 can be similar to the aggregation process described in the above formula (1), which will not be described in detail here.

[0065] like Figure 3 As shown in FIG. 1 , after clustering the three-dimensional point cloud 15, a point cloud cluster 31 and a point cloud cluster 32 are obtained, wherein the heights of the cluster centers of the point cloud cluster 31 and the point cloud cluster 32 are both close to the preset height. Further, according to the point cloud cluster 31, the following can be obtained: Figure 4 The first target object 41 shown in FIG. 4 can be obtained according to the point cloud cluster 32 as follows: Figure 4 A first target object 42 is shown.

[0066] It is understandable that the first target objects are merely schematically illustrated here, and the number of the first target objects is not limited.

[0067] S203: Determine a target detection model according to the distance of the first target object relative to the movable platform and the corresponding relationship between the distance and the detection model.

[0068] like Figure 3 The point cloud cluster 31 and the point cloud cluster 32 shown include multiple point cloud points respectively. Since the point cloud points in the three-dimensional point cloud detected and acquired by the detection device at each sampling moment carry position information, the distance between the point cloud point and the detection device can be calculated based on the position information of each point cloud point. Further, the distance between the point cloud cluster and the vehicle body equipped with the detection device can be calculated based on the distance between the multiple point cloud points in the point cloud cluster and the detection device, and then the distance between the first target object corresponding to the point cloud cluster and the vehicle body can be obtained, for example, the distance between the first target object 41 and the vehicle 11 and the distance between the first target object 42 and the vehicle 11.

[0069] like Figure 4 As shown, the distance of the first target object 41 relative to the vehicle 11 is less than the distance of the first target object 42 relative to the vehicle 11. For example, the distance of the first target object 41 relative to the vehicle 11 is recorded as L1, and the distance of the first target object 42 relative to the vehicle 11 is recorded as L2. In this embodiment, the vehicle-mounted device can determine the target detection model corresponding to L1 based on the distance L1 of the first target object 41 relative to the vehicle 11 and the correspondence between the distance and the detection model. According to the distance L2 of the first target object 42 relative to the vehicle 11 and the correspondence between the distance and the detection model, the target detection model corresponding to L2 is determined.

[0070] In an optional implementation, verification models corresponding to different distances may be pre-trained.

[0071] For example Figure 5 As shown, specifically, according to the distance between the sample object and the movable platform for detecting the sample object, such as the collection vehicle, the sample object can be divided into sample objects within the range of 0-90 meters, sample objects within the range of 75 meters to 165 meters, and sample objects within the range of 125 meters to 200 meters relative to the collection vehicle. The collection vehicle can be the vehicle 11 as described above, or it can be a vehicle other than the vehicle 11. Specifically, the detection model obtained by training the sample objects within the range of 0-90 meters relative to the collection vehicle is detection model 1, the detection model obtained by training the sample objects within the range of 75 meters to 165 meters relative to the collection vehicle is detection model 2, and the detection model obtained by training the sample objects within the range of 125 meters to 200 meters relative to the collection vehicle is detection model 3, thereby obtaining the correspondence between the distance and the detection model.

[0072] In another optional implementation, the detection model can adjust the adaptability point according to the actual distance obtained. For example, a parameter that can be set in the inspection model and can be adjusted according to the distance. In specific implementation, the distance of the first target object is obtained, and then the parameters in the inspection model are set according to the distance to obtain the target inspection model.

[0073] S204: Detect a point cloud cluster corresponding to the first target object using the target detection model to determine an object type of the first target object.

[0074] For example, if the on-board device determines that the distance L1 of the first target object 41 relative to the vehicle 11 is within the range of 0-90 meters, the detection model 1 is used to detect the point cloud cluster corresponding to the first target object 41 to determine the object type of the first target object 41. If the distance L2 of the first target object 42 relative to the vehicle 11 is within the range of 75 meters to 165 meters, the detection model 2 is used to detect the point cloud cluster corresponding to the first target object 42 to determine the object type of the first target object 42.

[0075] It is worth noting that the point cloud distribution characteristics of vehicles in different distance ranges are different. For example, the point cloud corresponding to long-range targets is sparsely distributed, while the point cloud corresponding to short-range targets is densely distributed. The point cloud corresponding to short-range vehicles often presents the point cloud of the side of the vehicle, while the point cloud corresponding to medium-range vehicles presents more point clouds of the rear of the vehicle. Therefore, training multiple detection models with different distances can more accurately identify targets.

[0076] In addition, the object types mentioned above may include road markings, vehicles, pedestrians, road signs, etc. Furthermore, the specific type of vehicle may be identified based on the features of the point cloud cluster, for example, engineering vehicles, cars, buses, etc. may be identified.

[0077] It can be understood that the first target object in this embodiment is only used to distinguish it from the second target object in subsequent embodiments. Both the first target object and the second target object may refer to target objects detectable by the detection device.

[0078] This embodiment clusters the three-dimensional point cloud obtained by the detection equipment carried on the movable platform to obtain a point cloud cluster corresponding to the target object. During the clustering process, the height of the cluster center of the point cloud cluster needs to meet a preset height condition. Furthermore, according to the distance of the target object relative to the movable platform and the correspondence between the distance and the detection model, the target detection model is determined, and the point cloud cluster corresponding to the target object is detected by the target detection model, so that the target detection model can determine the object type of the target object. In other words, different detection models are used to detect target objects at different distances relative to the movable platform, thereby improving the detection accuracy of the target object.

[0079] Based on the above embodiment, before clustering the three-dimensional point cloud to obtain a point cloud cluster corresponding to the first target object, the method further includes: removing specific point clouds in the three-dimensional point cloud, where the specific point clouds include ground point clouds.

[0080] like Figure 3 As shown, the three-dimensional point cloud 15 detected by the detection device includes not only the point cloud corresponding to the target object, but may also include specific point clouds, such as ground point clouds 30. Therefore, before clustering the three-dimensional point cloud 15, the ground point cloud 30 in the three-dimensional point cloud 15 can be identified by a plane fitting method, and the ground point cloud 30 in the three-dimensional point cloud 15 can be removed. Further, the three-dimensional point cloud after the ground point cloud 30 is removed can be clustered.

[0081] This embodiment removes specific point clouds from the three-dimensional point cloud obtained by the detection equipment carried on the movable platform, and clusters the three-dimensional point cloud after removing the specific point clouds to obtain a point cloud cluster corresponding to the target object. This can avoid the influence of the specific point clouds on the detection of the target object, thereby further improving the detection accuracy of the target object.

[0082] An embodiment of the present application provides a method for detecting a target object. Figure 6 This is a flow chart of a method for detecting a target object provided by another embodiment of the present application. Figure 6As shown, on the basis of the above embodiment, before detecting the point cloud cluster corresponding to the first target object through the target detection model and determining the object type of the first target object, the method also includes: determining the movement direction of the first target object; adjusting the movement direction of the first target object to a preset direction.

[0083] As a possible implementation manner, determining the movement direction of the first target object includes: determining the movement direction of the first target object based on a three-dimensional point cloud corresponding to the first target object at a first moment and a three-dimensional point cloud corresponding to the first target object at a second moment.

[0084] Specifically, the first moment is the previous moment, and the second moment is the current moment. Taking the first target object 41 as an example, since the first target object 41 may be in motion, the position information of the first target object 41 may change in real time. In addition, the detection device on the vehicle 11 is detecting the surrounding environment in real time, so the on-board device can obtain and process the three-dimensional point cloud detected by the detection device in real time. The three-dimensional point cloud corresponding to the first target object 41 at the previous moment and the three-dimensional point cloud corresponding to the first target object 41 at the current moment may change, so the movement direction of the first target object 41 can be determined based on the three-dimensional point cloud corresponding to the first target object 41 at the previous moment and the three-dimensional point cloud corresponding to the first target object 41 at the current moment.

[0085] Optionally, determining the movement direction of the first target object based on the three-dimensional point cloud corresponding to the first target object at the first moment and the three-dimensional point cloud corresponding to the first target object at the second moment includes: projecting the three-dimensional point cloud corresponding to the first target object at the first moment and the three-dimensional point cloud corresponding to the first target object at the second moment into a world coordinate system, respectively; determining the movement direction of the first target object based on the three-dimensional point cloud corresponding to the first target object at the first moment and the three-dimensional point cloud corresponding to the first target object at the second moment in the world coordinate system.

[0086] For example, the three-dimensional point cloud corresponding to the first target object 41 at the previous moment and the three-dimensional point cloud corresponding to the first target object 41 at the current moment are respectively projected into the world coordinate system. Further, the relative position relationship between the three-dimensional point cloud corresponding to the first target object 41 at the previous moment and the three-dimensional point cloud corresponding to the first target object 41 at the current moment is calculated by an iterative closest point (ICP) algorithm. The relative position relationship includes a rotation relationship and a translation relationship. The movement direction of the first target object 41 can be determined according to the translation relationship. In a possible implementation method, the translation relationship is the movement direction of the first target object 41.

[0087] As another possible implementation manner, the determining the moving direction of the first target object includes the following steps:

[0088] S601: Project the three-dimensional point cloud corresponding to the first target object at the first moment into the two-dimensional image at the first moment to obtain a first projection point.

[0089] S602: Project the three-dimensional point cloud corresponding to the first target object at the second moment onto the two-dimensional image at the second moment to obtain a second projection point.

[0090] In this embodiment, the vehicle 11 may also be equipped with a shooting device, which can be used to capture images of the surrounding environment of the vehicle 11, which are specifically two-dimensional images. The period of the detection device detecting and obtaining the three-dimensional point cloud and the period of the shooting device shooting images may be the same or different. For example, at the previous moment, while the detection device detects and obtains the three-dimensional point cloud of the first target object 41, the shooting device captures a frame of two-dimensional image. At the current moment, while the detection device detects and obtains the three-dimensional point cloud of the first target object 41, the shooting device captures another frame of two-dimensional image. Here, the two-dimensional image captured by the shooting device at the previous moment can be recorded as the first image, and the two-dimensional image captured by the shooting device at the current moment can be recorded as the second image. Specifically, the three-dimensional point cloud of the first target object 41 at the previous moment can be projected onto the first image to obtain a first projection point. The three-dimensional point cloud of the first target object 41 at the current moment is projected onto the second image to obtain a second projection point. Figure 7 As shown, the left area represents the three-dimensional point cloud detected by the detection device at a certain moment, and the right area represents the projection of the three-dimensional point cloud onto the two-dimensional image to obtain the projection area of ​​the three-dimensional point cloud on the two-dimensional image, which includes the projection points.

[0091] In an optional implementation, projecting the three-dimensional point cloud into the two-dimensional image includes: projecting some or all of the point cloud points in the three-dimensional point cloud onto the two-dimensional plane along the Z axis. The Z axis may be the Z axis of the vehicle body coordinate system. Alternatively, if the coordinates of the three-dimensional point cloud have been corrected to the earth coordinate system, the Z axis may be the Z axis of the earth coordinate system.

[0092] S603. Determine three-dimensional information of the first feature point based on the first projection point and a first feature point in the two-dimensional image at the first moment, wherein the first feature point is a feature point whose positional relationship with the first projection point conforms to a preset positional relationship.

[0093] For ease of distinction, the projection point of the three-dimensional point cloud of the first target object 41 at the previous moment on the first image is recorded as the first projection point, and the feature point on the first image is recorded as the first feature point. The positional relationship between the first feature point and the first projection point conforms to the preset positional relationship.

[0094] Optionally, determining the three-dimensional information of the first feature point based on the first projection point and the first feature point in the two-dimensional image at the first moment includes: determining a weight coefficient corresponding to the first projection point based on a distance between the first projection point and the first feature point in the two-dimensional image at the first moment; and determining the three-dimensional information of the first feature point based on the weight coefficient corresponding to the first projection point and the three-dimensional information of the first projection point.

[0095] like Figure 8 As shown, 80 represents the first image captured by the shooting device at the previous moment, and 81 represents the projection area formed by projecting the three-dimensional point cloud of the first target object 41 at the previous moment onto the first image 80. A two-dimensional feature point, i.e., the first feature point, can be extracted from the projection area 81. The two-dimensional feature point is not necessarily a projection point, that is, the two-dimensional feature point does not necessarily have three-dimensional information. Here, the three-dimensional information of the two-dimensional feature point can be estimated by Gaussian distribution. Figure 8 As shown, 82 represents any two-dimensional feature point in the projection area 81. Further, a preset range around the two-dimensional feature point 82 is determined, for example, the projection points within a 10*10 pixel area. For example, A, B, C, and D are projection points within the preset range. The distance of projection point A relative to the two-dimensional feature point 82 is recorded as d1, the distance of projection point B relative to the two-dimensional feature point 82 is recorded as d2, the distance of projection point C relative to the two-dimensional feature point 82 is recorded as d3, and the distance of projection point D relative to the two-dimensional feature point 82 is recorded as d4. Among them, (μ1, v1) represents the pixel coordinates of the projection point A on the first image 80 , and (μ0, v0) represents the pixel coordinates of the two-dimensional feature point 82 on the first image 80 . (μ2, v2) represents the pixel coordinates of the projection point B on the first image 80. (μ3, v3) represents the pixel coordinates of the projection point C on the first image 80. (μ4, v4) represents the pixel coordinates of the projection point D on the first image 80. In addition, the three-dimensional information of the three-dimensional point corresponding to the projection point A is recorded as P1, the three-dimensional information of the three-dimensional point corresponding to the projection point B is recorded as P2, the three-dimensional information of the three-dimensional point corresponding to the projection point C is recorded as P3, and the three-dimensional information of the three-dimensional point corresponding to the projection point D is recorded as P4. Among them, P1, P2, P3, and P4 are vectors, respectively, including xyz three-axis coordinates.

[0096] The three-dimensional information of the two-dimensional feature point 82 is recorded as P0, and P0 can be calculated by the following formulas (2) and (3):

[0097]

[0098]

[0099] Where n represents the number of projection points within a preset range around the two-dimensional feature point 82, ω i represents the weight coefficient. Different projection points may correspond to different weight coefficients or the same weight coefficient. σ is an adjustable parameter, for example, a parameter adjusted based on experience.

[0100] It can be understood that the calculation process of the three-dimensional information of other two-dimensional feature points in the projection area 81 is similar to the calculation process of the three-dimensional information of the two-dimensional feature point 82 as described above, and will not be repeated here.

[0101] S604. Determine three-dimensional information of the second feature point based on the second projection point and the second feature point in the two-dimensional image at the second moment, wherein the second feature point is a feature point whose positional relationship with the second projection point conforms to a preset positional relationship, and the second feature point corresponds to the first feature point.

[0102] For ease of distinction, the projection point of the three-dimensional point cloud of the first target object 41 at the current moment on the second image is recorded as the second projection point, and the feature point on the second image is recorded as the second feature point. The positional relationship between the second feature point and the second projection point conforms to the preset positional relationship.

[0103] According to the first feature point on the first image 80 , a second feature point corresponding to the first feature point on the second image can be calculated using a corner tracking algorithm (Kanade-Lucas-Tomasi Tracking, KLT).

[0104] Optionally, determining the three-dimensional information of the second feature point based on the second projection point and the second feature point in the two-dimensional image at the second moment includes: determining a weight coefficient corresponding to the second projection point based on a distance between the second projection point and the second feature point in the two-dimensional image at the second moment; and determining the three-dimensional information of the second feature point based on the weight coefficient corresponding to the second projection point and the three-dimensional information of the second projection point.

[0105] Specifically, the process of calculating the three-dimensional information of the second feature point on the second image is similar to the process of calculating the three-dimensional information of the first feature point on the first image, which will not be described in detail here.

[0106] S605: Determine a moving direction of the first target object according to the three-dimensional information of the first feature point and the three-dimensional information of the second feature point.

[0107] Specifically, the three-dimensional information of the first feature point is the three-dimensional information P0 of the two-dimensional feature point 82 as described above, and the three-dimensional information of the second feature point is the three-dimensional information of the two-dimensional feature point corresponding to the two-dimensional feature point 82 in the second image, which is recorded as P′0. The movement direction of the first target object 41 can be determined based on P0 and P′0. Specifically, the position change between P0 and P′0 is the movement direction of the first target object 41.

[0108] Optionally, before determining the movement direction of the first target object based on the three-dimensional information of the first feature point and the three-dimensional information of the second feature point, the method also includes: converting the three-dimensional information of the first feature point and the three-dimensional information of the second feature point into a world coordinate system, respectively.

[0109] For example, P0 and P′0 are respectively converted into the world coordinate system, and the position change between P0 and P′0 is calculated in the world coordinate system, and the position change is the movement direction of the first target object 41 .

[0110] It can be understood that the movement directions of other first target objects except the first target object 41 can also be determined by the several possible implementation methods as described above, which will not be described in detail here.

[0111] After determining the moving direction of the first target object, the moving direction of the first target object may be further adjusted to a preset direction. Optionally, the preset direction is the moving direction of a sample object used to train the detection model.

[0112] For example, the movement direction of the sample object used to train the detection model is toward the north, or toward the front or rear of the collection vehicle detecting the sample object. Taking the north direction as an example, in order to enable the detection model to accurately detect the first target object 41 or the first target object 42, the movement direction of the first target object 41 or the first target object 42 needs to be adjusted to the north. For example, the angle between the movement direction of the first target object 41 or the first target object 42 and the north direction is θ, then the three-dimensional point cloud corresponding to the first target object 41 or the three-dimensional point cloud corresponding to the first target object 42 is rotated according to the following formula (4) as described in the formula R z (θ) is rotated so that the movement direction of the first target object 41 or the first target object 42 is northward:

[0113]

[0114] This embodiment determines the movement direction of the target object and adjusts the movement direction of the target object to a preset direction. Since the preset direction is the movement direction of the sample object used to train the detection model, after the movement direction of the target object is adjusted to the preset direction, the detection model is used for detection, which can further improve the detection accuracy of the target object.

[0115] The embodiment of the present application provides a method for detecting a target object. Based on the above embodiment, after detecting the point cloud cluster corresponding to the first target object through the target detection model and determining the object type of the first target object, the method further includes: if the first target object is determined to be a vehicle through the target detection model, verifying the detection result of the target detection model according to preset conditions.

[0116] For example, when it is determined through the target detection model that the first target object 41 is a vehicle, the detection result is further verified through preset conditions.

[0117] Optionally, the preset condition includes at least one of the following: the size of the first target object meets a preset size; the spatial overlap between the first target object and other target objects around the first target object is less than a preset threshold.

[0118] For example, when the target detection model detects that the first target object 41 is a vehicle, further, it is detected whether the width of the first target object 41 exceeds a preset width range, and the preset width range can be the width range of a normal vehicle, for example, 2.8 meters to 3 meters. If the width of the first target object 41 exceeds the preset width range, it is determined that the detection result of the detection model for the first target object 41 is biased, that is, the first target object 41 may not be a vehicle. If the width of the first target object 41 is within the preset width range, the spatial overlap between the first target object 41 and other surrounding target objects is further detected, wherein the spatial overlap between the first target object 41 and other surrounding target objects can be specifically the spatial overlap between the identification frame used to characterize the first target object 41 and the identification frame used to characterize other surrounding target objects. If the spatial overlap is greater than a preset threshold, it is determined that the detection result of the detection model for the first target object 41 is biased, that is, the first target object 41 may not be a vehicle. If the spatial overlap is less than the preset threshold, it is determined that the detection result of the detection model for the first target object 41 is correct.

[0119] In this embodiment, after detecting the target object according to the distance of the target object relative to the movable platform through the target detection model corresponding to the distance, if it is determined that the object type of the target object is a vehicle, the detection result of the target detection model is further verified according to preset conditions. When the preset conditions are met, it is determined that the detection result of the target detection model is correct; when the preset conditions are not met, it is determined that the detection result of the target detection model has a deviation, thereby further improving the detection accuracy of the target object.

[0120] An embodiment of the present application provides a method for detecting a target object. Fig. 9 This is a flow chart of a method for detecting a target object provided by another embodiment of the present application. Fig. 9 As shown, based on the above embodiment, the distance between the first target object and the movable platform is less than or equal to the first preset distance. Fig.10 As shown, the right area 1001 is the 3D point cloud detected by the detection device, the upper left corner image 1002 represents the image of the 3D point cloud after removing the height information, and the lower left corner image 1003 represents the 2D image. The white circles in the right area 1001 represent the ground point cloud, and the white arc 100 represents the first preset distance relative to the detection device, such as 80 meters away. 101, 102, and 103 respectively represent the first target object with a distance less than or equal to 80 meters relative to the detection device. According to Fig.10 It can be seen that there are no circles of white coils beyond 80 meters, that is, no ground point cloud is detected beyond 80 meters. This embodiment provides a method for detecting a ground point cloud beyond a first preset distance and detecting a second target object beyond the first preset distance.

[0121] After detecting the point cloud cluster corresponding to the first target object by the target detection model and determining the object type of the first target object, the method further comprises the following steps:

[0122] S901: If the first target object is determined to be a vehicle through the target detection model, determine a ground point cloud outside the first preset distance according to the position of the first target object.

[0123] In this embodiment, assuming that the vehicle-mounted device uses the target detection model corresponding to the distance of the first target object 101 relative to the detection device to determine that the first target object 101 is a vehicle, uses the target detection model corresponding to the distance of the first target object 102 relative to the detection device to determine that the first target object 102 is a vehicle, and uses the target detection model corresponding to the distance of the first target object 103 relative to the detection device to determine that the first target object 103 is a vehicle, then the vehicle-mounted device can also determine the ground point cloud 80 meters away from the detection device based on the positions of the first target object 101, the first target object 102 and the first target object 103.

[0124] Optionally, determining the ground point cloud outside the first preset distance based on the position of the first target object includes: determining the slope of the ground where the first target object is located based on the position of the first target object; and determining the ground point cloud outside the first preset distance based on the slope of the ground.

[0125] Specifically, the slope of the ground where the first target objects 101, 102 and 103 are located is determined according to the positions of the first target objects 101, 102 and 103, and the ground point cloud 80 meters away from the detection device is determined according to the slope of the ground. It can be understood that this embodiment does not limit the number of first target objects.

[0126] Optionally, determining the slope of the ground where the first target object is located based on the position of the first target object includes: determining the slope of a plane formed by at least three first target objects based on the positions of at least three first target objects, the slope of the plane being the slope of the ground where the first target object is located.

[0127] For example, when the first target object 101, the first target object 102, and the first target object 103 are all vehicles, the three vehicles can determine a plane. For example, the coordinates of the first target object 101 are marked as A (x1, y1, z1), the coordinates of the first target object 102 are marked as B (x2, y2, z2), and the coordinates of the first target object 103 are marked as C (x3, y3, z3), then the vector AB = (x2-x1, y2-y1, z2-z1), and the vector AC = (x3-x1, y3-y1, z3-z1). The normal vector of the plane where AB and AC are located is AB×AC = (a, b, c), where:

[0128] a=(y2-y1)(z3-z1)-(z2-z1)(y3-y1)

[0129] b=(z2-z1)(x3-x1)-(z3-z1)(x2-x1)

[0130] c=(x2-x1)(y3-y1)-(x3-x1)(y2-y1)

[0131] Specifically, according to the normal vectors of the planes where AB and AC are located, the slope of the plane formed by the first target object 101, the first target object 102 and the first target object 103 can be determined. The slope of the plane can specifically be the slope of the ground where the first target object 101, the first target object 102 and the first target object 103 are located.

[0132] It can be understood that when the number of first target objects is greater than 3, every 3 first target objects can determine a plane, so that multiple planes can be obtained. The slopes of multiple planes can be calculated by the plane slope calculation method as described above. At this time, the ground slope can be fitted according to the slopes of the multiple planes.

[0133] It can be understood that, based on the ground slope, it can be determined whether the ground is horizontal, an viaduct or a slope. In some embodiments, the ground where the first target object is located may not be horizontal, for example, it may be an viaduct or a slope. Therefore, based on the ground slope, it can also be determined whether the first target object is on a viaduct or a slope.

[0134] After determining the ground slope of the ground where the first target object is located, the ground where the first target object is located can be extended according to the ground slope to obtain the ground point cloud 80 meters away. For example, a straight line extension is performed to a distance of 80 meters away according to the width of the road where the first target object is located. Here, the situation that the ground 80 meters away is horizontal can be considered, and the situation that there is a slope or viaduct 80 meters away can be temporarily ignored.

[0135] S902: Determine, according to the ground point cloud outside the first preset distance, an object type of a second target object outside the first preset distance.

[0136] For example, the object type of the second target object 80 meters away is determined based on the ground point cloud 80 meters away.

[0137] Optionally, determining the object type of a second target object outside the first preset distance based on the ground point cloud outside the first preset distance includes: determining a point cloud cluster corresponding to the second target object outside the first preset distance based on the ground point cloud outside the first preset distance, the bottom of the second target object and the bottom of the first target object being in the same plane; and determining the object type of the second target object by detecting the point cloud cluster corresponding to the second target object through a detection model corresponding to the distance of the second target object relative to the movable platform.

[0138] For example, based on the ground point cloud 80 meters away, determine the point cloud cluster corresponding to the second target object 80 meters away, such as Fig.11 As shown, since the second target object 80 meters away will be blocked by nearby objects, the number of the 3D point cloud 104 in the distance is small, that is, the 3D point cloud 104 in the distance may be only a part of the 3D point cloud of the upper part of the second target object. At this time, it is necessary to fill the remaining part of the 3D point cloud of the second target object according to the ground point cloud 80 meters away. For example, it is necessary to fill the 3D point cloud of the lower half of the second target object so that the bottom of the second target object is in the same plane as the bottoms of the first target object 101, the first target object 102 and the first target object 103. Among them, the partial 3D point cloud of the upper part of the second target object and the filled 3D point cloud of the lower half can constitute the point cloud cluster corresponding to the second target object.

[0139] Further, according to the distance of the second target object relative to the detection device, the point cloud cluster corresponding to the second target object is detected using a detection model corresponding to the distance, that is, the second target object is detected by the detection model as a pedestrian, a vehicle or other object. In addition, the number of second target objects is not limited here, and may be one or more. Since the distance of the second target object relative to the detection device is greater than the first preset distance, the second target object can be detected using a detection model corresponding to a second preset distance greater than the first preset distance.

[0140] Optionally, determining a point cloud cluster corresponding to a second target object outside the first preset distance based on the ground point cloud outside the first preset distance includes: clustering the three-dimensional point cloud outside the first preset distance after removing the ground point cloud to obtain a partial point cloud corresponding to the second target object; determining a point cloud cluster corresponding to the second target object based on the partial point cloud corresponding to the second target object and the ground point cloud outside the first preset distance.

[0141] For example, a 3D point cloud detected by a detection device at a distance of 80 meters from the detection device is obtained. Since the 3D point cloud at a distance of 80 meters may include ground point clouds, it is necessary to remove the ground point clouds from the 3D point clouds at a distance of 80 meters, and cluster the 3D point clouds after removing the ground point clouds to obtain a partial point cloud corresponding to the second target object, for example Fig.11 A three-dimensional point cloud 104 is shown.

[0142] Alternatively, after determining the ground slope of the ground where the first target object is located, the ground where the first target object is located is extended according to the ground slope to obtain a ground point cloud 80 meters away. When detecting the second target object 80 meters away, it is necessary to remove the extended ground point cloud 80 meters away, and cluster the three-dimensional point cloud after removing the ground point cloud from the three-dimensional point cloud 80 meters away, so as to obtain a partial point cloud corresponding to the second target object. Further, based on the partial point cloud corresponding to the second target object and the ground point cloud 80 meters away, the point cloud cluster corresponding to the second target object is determined. Specifically, the lower half of the second target object is padded so that the bottom of the second target object is in the same plane as the bottoms of the first target object 101, the first target object 102, and the first target object 103.

[0143] Specifically, the clustering process is similar to the clustering process described above, and will not be repeated here. The difference is that the vehicle height H used in the clustering process here is larger than the vehicle height H used in the clustering process described above. For example, the vehicle height H used in the clustering process here can be 1.6 meters or 2.5 meters. Optionally, the method also includes: if the second target object is a vehicle, and the width of the second target object is less than or equal to the first width, then the three-dimensional point cloud with a height greater than or equal to the first height in the point cloud cluster corresponding to the second target object is removed to obtain the remaining three-dimensional point cloud corresponding to the second target object; if the second target object is a vehicle, the width of the second target object is greater than the first width, and the width of the second target object is less than or equal to the second width, then the three-dimensional point cloud with a height greater than or equal to the second height in the point cloud cluster corresponding to the second target object is removed to obtain the remaining three-dimensional point cloud corresponding to the second target object; according to the remaining three-dimensional point cloud corresponding to the second target object, an identification box for characterizing the vehicle is generated, and the identification box is used for the movable platform to make navigation decisions; wherein, the second width is greater than the first width, and the second height is greater than the first height.

[0144] It is understandable that, since there may be tiny objects such as road signs or tree branches above the second target object, and since tiny objects such as road signs or tree branches may be very close to the second target object, when the point cloud cluster corresponding to the second target object is obtained by clustering, the point cloud cluster corresponding to the second target object may include the three-dimensional point cloud of tiny objects such as road signs or tree branches. Therefore, when the vehicle-mounted device uses the detection model corresponding to the distance of the second target object relative to the detection device to determine that the second target object is a vehicle, it is necessary to further process the point cloud cluster corresponding to the second target object.

[0145] Specifically, based on the width of the second target object, determine whether the second target object is a small car or a large car. For example, if the width of the second target object is less than or equal to the first width, then determine that the second target object is a small car. If the width of the second target object is greater than the first width, and the width of the second target object is less than or equal to the second width, then determine that the second target object is a large car. Specifically, the second width is greater than the first width. Further, if the second target object is a small car, remove the three-dimensional point cloud with a height greater than or equal to the first height, such as a height of more than 1.8 meters, in the point cloud cluster corresponding to the second target object, to obtain the remaining three-dimensional point cloud corresponding to the second target object. If the second target object is a large car, remove the three-dimensional point cloud with a height greater than or equal to the second height, such as a height of more than 3.2 meters, in the point cloud cluster corresponding to the second target object, to obtain the remaining three-dimensional point cloud corresponding to the second target object. Fig.12 As shown, the 3D point cloud in the circle 105 is the 3D point cloud corresponding to the tree branch. Further, based on the remaining 3D point clouds corresponding to the second target object, an identification frame for representing the vehicle is generated. Fig.11 The three-dimensional point cloud 104 is based on the removal of Fig.12 The remaining three-dimensional point cloud corresponding to the second target object is obtained by using the three-dimensional point cloud corresponding to the branch in the circle 105 shown in the figure. Further, the lower half of the second target object is filled according to the ground point cloud 80 meters away. For example, the three-dimensional point cloud of the lower half of the second target object needs to be filled so that the bottom of the second target object is in the same plane as the bottoms of the first target object 101, the first target object 102 and the first target object 103, and the following is obtained: Fig.12 The second target object shown is the identification frame 106 representing the vehicle. Further, a vehicle equipped with a detection device, such as the vehicle 11, can make navigation decisions based on the identification frame 106, for example, plan a route based on the identification frame 106, plan the driving route of the vehicle 11 in advance, or control the vehicle 11 to switch to other lanes in advance, or control the speed of the vehicle 11 in advance, etc.

[0146] This embodiment determines the distant ground point cloud based on the position of the first target object nearby, and detects the distant second target object based on the distant ground point cloud, so that the movable platform equipped with the detection device makes navigation decisions based on the distant second target object, thereby improving the safety of the movable platform. In addition, by detecting whether the second target object is a large vehicle or a small vehicle, and according to the height corresponding to the large vehicle or the height corresponding to the small vehicle, the three-dimensional point cloud corresponding to the second target object that may be a small object such as a road sign or a tree branch is removed, thereby improving the detection accuracy of the second target object. In addition, based on the positions of at least three first target objects, the slope of the plane formed by the at least three first target objects is determined, and the slope of the ground where the first target objects are located is determined based on the slope of the plane. Based on the slope of the ground where the first target objects are located, it can also be determined whether the ground is horizontal, an elevated bridge or a slope, etc., thereby improving the detection accuracy of ground recognition. When removing the ground point cloud, not only the point cloud of the horizontal ground can be removed, but also the point cloud of the road surface such as elevated bridges or slopes can be removed, thereby reducing the influence of the point cloud of the horizontal ground and the point cloud of the road surface such as elevated bridges or slopes on the detection of the first target object or the second target object, thereby further improving the detection accuracy of the first target object or the second target object.

[0147] An embodiment of the present application provides a system for detecting a target object. Fig.13 The structure diagram of the target object detection system provided in the embodiment of the present application is as follows: Fig.13 As shown, the target object detection system 130 includes: a detection device 131, a memory 132 and a processor 133. Among them, the detection device 131 is used to detect the surrounding environment of the movable platform to obtain a three-dimensional point cloud. The processor 133 can specifically be a component in the vehicle-mounted device in the above embodiment, or other components, devices or components with data processing functions carried in the vehicle. Specifically, the memory 132 is used to store program code; the processor 133 calls the program code, and when the program code is executed, it is used to perform the following operations: obtain the three-dimensional point cloud; cluster the three-dimensional point cloud to obtain a point cloud cluster corresponding to the first target object, wherein the height of the cluster center of the clustered point cloud cluster meets the preset height condition; determine the target detection model according to the distance of the first target object relative to the movable platform, and the correspondence between the distance and the detection model; detect the point cloud cluster corresponding to the first target object through the target detection model, and determine the object type of the first target object.

[0148] Optionally, the processor 133 detects the point cloud cluster corresponding to the first target object through the target detection model, and before determining the object type of the first target object, is also used to: determine the movement direction of the first target object; and adjust the movement direction of the first target object to a preset direction.

[0149] Optionally, the preset direction is a movement direction of a sample object used to train the detection model.

[0150] Optionally, when the processor 133 determines the movement direction of the first target object, it is specifically used to: determine the movement direction of the first target object according to the three-dimensional point cloud corresponding to the first target object at the first moment and the three-dimensional point cloud corresponding to the first target object at the second moment.

[0151] Optionally, when the processor 133 determines the movement direction of the first target object based on the three-dimensional point cloud corresponding to the first target object at the first moment and the three-dimensional point cloud corresponding to the first target object at the second moment, it is specifically used to: project the three-dimensional point cloud corresponding to the first target object at the first moment and the three-dimensional point cloud corresponding to the first target object at the second moment into the world coordinate system, respectively; determine the movement direction of the first target object based on the three-dimensional point cloud corresponding to the first target object at the first moment and the three-dimensional point cloud corresponding to the first target object at the second moment in the world coordinate system.

[0152] Optionally, when the processor 133 determines the movement direction of the first target object, it is specifically used to: project the three-dimensional point cloud corresponding to the first target object at the first moment into the two-dimensional image at the first moment to obtain a first projection point; project the three-dimensional point cloud corresponding to the first target object at the second moment into the two-dimensional image at the second moment to obtain a second projection point; determine the three-dimensional information of the first feature point based on the first projection point and the first feature point in the two-dimensional image at the first moment, wherein the first feature point is a feature point whose positional relationship with the first projection point conforms to a preset positional relationship; determine the three-dimensional information of the second feature point based on the second projection point and the second feature point in the two-dimensional image at the second moment, wherein the second feature point is a feature point whose positional relationship with the second projection point conforms to a preset positional relationship, and the second feature point corresponds to the first feature point; determine the movement direction of the first target object based on the three-dimensional information of the first feature point and the three-dimensional information of the second feature point.

[0153] Optionally, when the processor 133 determines the three-dimensional information of the first feature point based on the first projection point and the first feature point in the two-dimensional image at the first moment, it is specifically used to: determine the weight coefficient corresponding to the first projection point based on the distance between the first projection point and the first feature point in the two-dimensional image at the first moment; determine the three-dimensional information of the first feature point based on the weight coefficient corresponding to the first projection point and the three-dimensional information of the first projection point.

[0154] Optionally, when the processor 133 determines the three-dimensional information of the second feature point based on the second projection point and the second feature point in the two-dimensional image at the second moment, it is specifically used to: determine the weight coefficient corresponding to the second projection point based on the distance between the second projection point and the second feature point in the two-dimensional image at the second moment; determine the three-dimensional information of the second feature point based on the weight coefficient corresponding to the second projection point and the three-dimensional information of the second projection point.

[0155] Optionally, before determining the movement direction of the first target object based on the three-dimensional information of the first feature point and the three-dimensional information of the second feature point, the processor 133 is also used to: convert the three-dimensional information of the first feature point and the three-dimensional information of the second feature point into a world coordinate system, respectively.

[0156] Optionally, the processor 133 detects the point cloud cluster corresponding to the first target object through the target detection model, and after determining the object type of the first target object, is also used to: if it is determined through the target detection model that the first target object is a vehicle, verify the detection result of the target detection model according to preset conditions.

[0157] Optionally, the preset condition includes at least one of the following: the size of the first target object meets a preset size; the overlap between the first target object and other target objects around the first target object is less than a preset threshold.

[0158] Optionally, before clustering the three-dimensional point cloud to obtain a point cloud cluster corresponding to the first target object, the processor 133 is further configured to: remove specific point clouds from the three-dimensional point cloud, where the specific point clouds include ground point clouds.

[0159] Optionally, the distance of the first target object relative to the movable platform is less than or equal to a first preset distance; after the processor 133 detects the point cloud cluster corresponding to the first target object through the target detection model and determines the object type of the first target object, it is also used to: if the first target object is determined to be a vehicle through the target detection model, determine the ground point cloud outside the first preset distance according to the position of the first target object; determine the object type of the second target object outside the first preset distance according to the ground point cloud outside the first preset distance.

[0160] Optionally, when the processor 133 determines the ground point cloud outside the first preset distance based on the position of the first target object, it is specifically used to: determine the slope of the ground where the first target object is located based on the position of the first target object; and determine the ground point cloud outside the first preset distance based on the slope of the ground.

[0161] Optionally, when the processor 133 determines the slope of the ground where the first target object is located based on the position of the first target object, it is specifically used to: determine the slope of a plane formed by at least three first target objects based on the positions of at least three first target objects, the slope of the plane being the slope of the ground where the first target object is located.

[0162] Optionally, when the processor 133 determines the object type of the second target object outside the first preset distance based on the ground point cloud outside the first preset distance, it is specifically used to: determine a point cloud cluster corresponding to the second target object outside the first preset distance based on the ground point cloud outside the first preset distance, and the bottom of the second target object and the bottom of the first target object are in the same plane; detect the point cloud cluster corresponding to the second target object through the detection model corresponding to the distance of the second target object relative to the movable platform, and determine the object type of the second target object.

[0163] Optionally, when the processor 133 determines the point cloud cluster corresponding to the second target object outside the first preset distance based on the ground point cloud outside the first preset distance, it is specifically used to: cluster the three-dimensional point cloud outside the first preset distance after removing the ground point cloud to obtain a partial point cloud corresponding to the second target object; determine the point cloud cluster corresponding to the second target object based on the partial point cloud corresponding to the second target object and the ground point cloud outside the first preset distance.

[0164] Optionally, the processor 133 is also used to: if the second target object is a vehicle and the width of the second target object is less than or equal to the first width, then remove the three-dimensional point cloud whose height is greater than or equal to the first height in the point cloud cluster corresponding to the second target object, and obtain the remaining three-dimensional point cloud corresponding to the second target object; if the second target object is a vehicle, the width of the second target object is greater than the first width, and the width of the second target object is less than or equal to the second width, then remove the three-dimensional point cloud whose height is greater than or equal to the second height in the point cloud cluster corresponding to the second target object, and obtain the remaining three-dimensional point cloud corresponding to the second target object; based on the remaining three-dimensional point cloud corresponding to the second target object, generate an identification box for characterizing the vehicle, and the identification box is used for the movable platform to make navigation decisions; wherein, the second width is greater than the first width, and the second height is greater than the first height.

[0165] The specific principles and implementation methods of the target object detection system provided in the embodiments of the present application are similar to those in the above embodiments and will not be repeated here.

[0166] An embodiment of the present application provides a movable platform. The movable platform includes: a fuselage, a power system, and a target object detection system as described in the above embodiment. The power system is installed on the fuselage to provide mobile power. The target object detection system can implement the target object detection method as described above. The specific principles and implementation methods of the target object detection method are similar to those in the above embodiment and are not repeated here. This embodiment does not limit the specific form of the movable platform. For example, the movable platform can be a drone, a movable robot or a vehicle.

[0167] In addition, this embodiment further provides a computer-readable storage medium on which a computer program is stored. The computer program is executed by a processor to implement the target object detection method described in the above embodiment.

[0168] In the several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0169] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0170] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of hardware plus software functional units.

[0171] The above-mentioned integrated unit implemented in the form of a software functional unit can be stored in a computer-readable storage medium. The above-mentioned software functional unit is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) or a processor to perform some steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk and other media that can store program code.

[0172] Those skilled in the art can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional modules is used as an example for illustration. In actual applications, the above-mentioned functions can be assigned to different functional modules as needed, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. The specific working process of the device described above can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.

[0173] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit it. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for detecting a target object, characterized in that: Applied to a movable platform, the movable platform is provided with a detection device, the detection device is used to detect the surrounding environment of the movable platform to obtain a three-dimensional point cloud, the method comprises: Acquire the three-dimensional point cloud; Clustering the three-dimensional point cloud to obtain a point cloud cluster corresponding to the first target object, wherein the height of the cluster center of the clustered point cloud cluster meets a preset height condition; Determining a target detection model according to a distance of the first target object relative to the movable platform and a corresponding relationship between the distance and the detection model; The point cloud cluster corresponding to the first target object is detected by the target detection model to determine the object type of the first target object.

2. The method according to claim 1, characterized in that Before detecting the point cloud cluster corresponding to the first target object by using the target detection model to determine the object type of the first target object, the method further includes: determining a moving direction of the first target object; The moving direction of the first target object is adjusted to a preset direction.

3. The method according to claim 2, characterized in that The preset direction is the movement direction of the sample object used to train the detection model.

4. The method according to claim 3, characterized in that: The determining the movement direction of the first target object includes: A moving direction of the first target object is determined according to a three-dimensional point cloud corresponding to the first target object at a first moment and a three-dimensional point cloud corresponding to the first target object at a second moment.

5. The method according to claim 4, characterized in that The determining the movement direction of the first target object according to the three-dimensional point cloud corresponding to the first target object at the first moment and the three-dimensional point cloud corresponding to the first target object at the second moment includes: Projecting the three-dimensional point cloud corresponding to the first target object at the first moment and the three-dimensional point cloud corresponding to the first target object at the second moment into the world coordinate system respectively; The moving direction of the first target object is determined according to the three-dimensional point cloud corresponding to the first target object at the first moment and the three-dimensional point cloud corresponding to the first target object at the second moment in the world coordinate system.

6. The method according to claim 3, characterized in that: The determining the movement direction of the first target object includes: Projecting the three-dimensional point cloud corresponding to the first target object at the first moment into the two-dimensional image at the first moment to obtain a first projection point; Projecting the three-dimensional point cloud corresponding to the first target object at the second moment into the two-dimensional image at the second moment to obtain a second projection point; Determine three-dimensional information of the first feature point according to the first projection point and a first feature point in the two-dimensional image at the first moment, wherein the first feature point is a feature point whose positional relationship with the first projection point conforms to a preset positional relationship; Determine three-dimensional information of the second feature point according to the second projection point and a second feature point in the two-dimensional image at the second moment, wherein the second feature point is a feature point whose positional relationship with the second projection point conforms to a preset positional relationship, and the second feature point corresponds to the first feature point; The moving direction of the first target object is determined according to the three-dimensional information of the first feature point and the three-dimensional information of the second feature point.

7. The method according to claim 6, characterized in that The determining, according to the first projection point and the first feature point in the two-dimensional image at the first moment, the three-dimensional information of the first feature point includes: Determine a weight coefficient corresponding to the first projection point according to a distance between the first projection point and a first feature point in the two-dimensional image at the first moment; The three-dimensional information of the first feature point is determined according to the weight coefficient corresponding to the first projection point and the three-dimensional information of the first projection point.

8. The method according to claim 6, characterized in that The determining the three-dimensional information of the second feature point according to the second projection point and the second feature point in the two-dimensional image at the second moment includes: Determine a weight coefficient corresponding to the second projection point according to a distance between the second projection point and a second feature point in the two-dimensional image at the second moment; The three-dimensional information of the second feature point is determined according to the weight coefficient corresponding to the second projection point and the three-dimensional information of the second projection point.

9. The method according to any one of claims 6 to 8, characterized in that: Before determining the movement direction of the first target object according to the three-dimensional information of the first feature point and the three-dimensional information of the second feature point, the method further includes: The three-dimensional information of the first feature point and the three-dimensional information of the second feature point are respectively converted into a world coordinate system.

10. The method according to any one of claims 1 to 8, characterized in that: After detecting the point cloud cluster corresponding to the first target object by using the target detection model and determining the object type of the first target object, the method further includes: If it is determined through the target detection model that the first target object is a vehicle, the detection result of the target detection model is verified according to preset conditions.

11. The method according to claim 9, characterized in that After detecting the point cloud cluster corresponding to the first target object by using the target detection model and determining the object type of the first target object, the method further includes: If it is determined through the target detection model that the first target object is a vehicle, the detection result of the target detection model is verified according to preset conditions.

12. The method according to claim 10, characterized in that The preset conditions include at least one of the following: The size of the first target object meets a preset size; A spatial overlap between the first target object and other target objects around the first target object is less than a preset threshold.

13. The method according to claim 11, characterized in that The preset conditions include at least one of the following: The size of the first target object meets a preset size; A spatial overlap between the first target object and other target objects around the first target object is less than a preset threshold.

14. The method according to any one of claims 1-8 or 11-13, characterized in that: Before clustering the three-dimensional point cloud to obtain a point cloud cluster corresponding to the first target object, the method further includes: Specific point clouds in the three-dimensional point cloud are removed, where the specific point clouds include ground point clouds.

15. The method according to claim 9, characterized in that Before clustering the three-dimensional point cloud to obtain a point cloud cluster corresponding to the first target object, the method further includes: Specific point clouds in the three-dimensional point cloud are removed, where the specific point clouds include ground point clouds.

16. The method according to claim 10, characterized in that Before clustering the three-dimensional point cloud to obtain a point cloud cluster corresponding to the first target object, the method further includes: Specific point clouds in the three-dimensional point cloud are removed, where the specific point clouds include ground point clouds.

17. The method according to any one of claims 1 to 8, characterized in that: The distance between the first target object and the movable platform is less than or equal to a first preset distance; After detecting the point cloud cluster corresponding to the first target object by using the target detection model and determining the object type of the first target object, the method further includes: If the first target object is determined to be a vehicle through the target detection model, determining a ground point cloud outside the first preset distance according to the position of the first target object; An object type of a second target object outside the first preset distance is determined according to the ground point cloud outside the first preset distance.

18. The method according to claim 9, characterized in that The distance between the first target object and the movable platform is less than or equal to a first preset distance; After detecting the point cloud cluster corresponding to the first target object by using the target detection model and determining the object type of the first target object, the method further includes: If the first target object is determined to be a vehicle through the target detection model, determining a ground point cloud outside the first preset distance according to the position of the first target object; An object type of a second target object outside the first preset distance is determined according to the ground point cloud outside the first preset distance.

19. The method according to claim 17, characterized in that The step of determining the ground point cloud outside the first preset distance according to the position of the first target object includes: Determining the slope of the ground where the first target object is located according to the position of the first target object; A ground point cloud outside the first preset distance is determined according to the slope of the ground.

20. The method according to claim 18, characterized in that The step of determining the ground point cloud outside the first preset distance according to the position of the first target object includes: Determining the slope of the ground where the first target object is located according to the position of the first target object; A ground point cloud outside the first preset distance is determined according to the slope of the ground.

21. The method according to claim 19, characterized in that The step of determining the slope of the ground where the first target object is located according to the position of the first target object includes: According to the positions of at least three of the first target objects, the slope of a plane formed by the at least three first target objects is determined, and the slope of the plane is the slope of the ground where the first target objects are located.

22. The method according to claim 20, characterized in that The step of determining the slope of the ground where the first target object is located according to the position of the first target object includes: According to the positions of at least three of the first target objects, the slope of a plane formed by the at least three first target objects is determined, and the slope of the plane is the slope of the ground where the first target objects are located.

23. The method according to claim 17, characterized in that The determining, based on the ground point cloud outside the first preset distance, the object type of the second target object outside the first preset distance includes: Determining, according to the ground point cloud outside the first preset distance, a point cloud cluster corresponding to a second target object outside the first preset distance, wherein the bottom of the second target object is in the same plane as the bottom of the first target object; The point cloud cluster corresponding to the second target object is detected by using a detection model corresponding to the distance of the second target object relative to the movable platform to determine the object type of the second target object.

24. The method according to any one of claims 18 to 22, characterized in that: The determining, based on the ground point cloud outside the first preset distance, the object type of the second target object outside the first preset distance includes: Determining, according to the ground point cloud outside the first preset distance, a point cloud cluster corresponding to a second target object outside the first preset distance, wherein the bottom of the second target object is in the same plane as the bottom of the first target object; The point cloud cluster corresponding to the second target object is detected by using a detection model corresponding to the distance of the second target object relative to the movable platform to determine the object type of the second target object.

25. The method according to claim 23, characterized in that The step of determining, based on the ground point cloud outside the first preset distance, a point cloud cluster corresponding to a second target object outside the first preset distance includes: Clustering the three-dimensional point cloud beyond the first preset distance after removing the ground point cloud to obtain a partial point cloud corresponding to the second target object; A point cloud cluster corresponding to the second target object is determined according to the partial point cloud corresponding to the second target object and the ground point cloud outside the first preset distance.

26. The method according to claim 24, characterized in that The step of determining, based on the ground point cloud outside the first preset distance, a point cloud cluster corresponding to a second target object outside the first preset distance includes: Clustering the three-dimensional point cloud beyond the first preset distance after removing the ground point cloud to obtain a partial point cloud corresponding to the second target object; A point cloud cluster corresponding to the second target object is determined according to the partial point cloud corresponding to the second target object and the ground point cloud outside the first preset distance.

27. The method according to claim 25 or 26, characterized in that The method further comprises: If the second target object is a vehicle, and the width of the second target object is less than or equal to the first width, removing the three-dimensional point cloud whose height is greater than or equal to the first height in the point cloud cluster corresponding to the second target object, to obtain the remaining three-dimensional point cloud corresponding to the second target object; If the second target object is a vehicle, the width of the second target object is greater than the first width, and the width of the second target object is less than or equal to the second width, then removing the three-dimensional point cloud with a height greater than or equal to the second height in the point cloud cluster corresponding to the second target object to obtain the remaining three-dimensional point cloud corresponding to the second target object; generating an identification frame for characterizing the vehicle according to the remaining three-dimensional point cloud corresponding to the second target object, wherein the identification frame is used for the movable platform to make navigation decisions; The second width is greater than the first width, and the second height is greater than the first height.

28. A target object detection system, characterized in that: include: Probe devices, memory, and processors; The detection device is used to detect the surrounding environment of the movable platform to obtain a three-dimensional point cloud; The memory is used to store program code; The processor calls the program code, and when the program code is executed, is used to perform the following operations: Acquire the three-dimensional point cloud; Clustering the three-dimensional point cloud to obtain a point cloud cluster corresponding to the first target object, wherein the height of the cluster center of the clustered point cloud cluster meets a preset height condition; Determining a target detection model according to a distance of the first target object relative to the movable platform and a corresponding relationship between the distance and the detection model; The point cloud cluster corresponding to the first target object is detected by the target detection model to determine the object type of the first target object.

29. The system according to claim 28, characterized in that The processor detects the point cloud cluster corresponding to the first target object by using the target detection model, and before determining the object type of the first target object, is further configured to: determining a moving direction of the first target object; The movement direction of the first target object is adjusted to a preset direction.

30. The system according to claim 29, characterized in that The preset direction is the movement direction of the sample object used to train the detection model.

31. The system according to claim 30, characterized in that When the processor determines the moving direction of the first target object, it is specifically used to: A moving direction of the first target object is determined according to a three-dimensional point cloud corresponding to the first target object at a first moment and a three-dimensional point cloud corresponding to the first target object at a second moment.

32. The system according to claim 31, characterized in that The processor determines the movement direction of the first target object according to the three-dimensional point cloud corresponding to the first target object at the first moment and the three-dimensional point cloud corresponding to the first target object at the second moment, specifically for: Projecting the three-dimensional point cloud corresponding to the first target object at the first moment and the three-dimensional point cloud corresponding to the first target object at the second moment into the world coordinate system respectively; The moving direction of the first target object is determined according to the three-dimensional point cloud corresponding to the first target object at the first moment and the three-dimensional point cloud corresponding to the first target object at the second moment in the world coordinate system.

33. The system according to claim 30, characterized in that When the processor determines the moving direction of the first target object, it is specifically used to: Projecting the three-dimensional point cloud corresponding to the first target object at the first moment into the two-dimensional image at the first moment to obtain a first projection point; Projecting the three-dimensional point cloud corresponding to the first target object at the second moment into the two-dimensional image at the second moment to obtain a second projection point; Determine three-dimensional information of the first feature point according to the first projection point and a first feature point in the two-dimensional image at the first moment, wherein the first feature point is a feature point whose positional relationship with the first projection point conforms to a preset positional relationship; Determine three-dimensional information of the second feature point according to the second projection point and a second feature point in the two-dimensional image at the second moment, wherein the second feature point is a feature point whose positional relationship with the second projection point conforms to a preset positional relationship, and the second feature point corresponds to the first feature point; The moving direction of the first target object is determined according to the three-dimensional information of the first feature point and the three-dimensional information of the second feature point.

34. The system according to claim 33, characterized in that When the processor determines the three-dimensional information of the first feature point according to the first projection point and the first feature point in the two-dimensional image at the first moment, it is specifically used to: Determine a weight coefficient corresponding to the first projection point according to a distance between the first projection point and a first feature point in the two-dimensional image at the first moment; The three-dimensional information of the first feature point is determined according to the weight coefficient corresponding to the first projection point and the three-dimensional information of the first projection point.

35. The system according to claim 33, characterized in that When the processor determines the three-dimensional information of the second feature point according to the second projection point and the second feature point in the two-dimensional image at the second moment, it is specifically used to: Determine a weight coefficient corresponding to the second projection point according to a distance between the second projection point and a second feature point in the two-dimensional image at the second moment; The three-dimensional information of the second feature point is determined according to the weight coefficient corresponding to the second projection point and the three-dimensional information of the second projection point.

36. The system according to any one of claims 33 to 35, characterized in that: Before the processor determines the moving direction of the first target object according to the three-dimensional information of the first feature point and the three-dimensional information of the second feature point, the processor is further configured to: The three-dimensional information of the first feature point and the three-dimensional information of the second feature point are respectively converted into a world coordinate system.

37. The system according to any one of claims 28 to 35, characterized in that: The processor detects the point cloud cluster corresponding to the first target object through the target detection model, and after determining the object type of the first target object, is further configured to: If it is determined through the target detection model that the first target object is a vehicle, the detection result of the target detection model is verified according to preset conditions.

38. The system according to claim 36, characterized in that The processor detects the point cloud cluster corresponding to the first target object through the target detection model, and after determining the object type of the first target object, is further configured to: If it is determined through the target detection model that the first target object is a vehicle, the detection result of the target detection model is verified according to preset conditions.

39. The system according to claim 37, characterized in that The preset conditions include at least one of the following: The size of the first target object meets a preset size; The degree of overlap between the first target object and other target objects around the first target object is less than a preset threshold.

40. The system of claim 38, wherein: The preset conditions include at least one of the following: The size of the first target object meets a preset size; The degree of overlap between the first target object and other target objects around the first target object is less than a preset threshold.

41. The system according to any one of claims 28-35 or 38-40, characterized in that: Before clustering the three-dimensional point cloud to obtain a point cloud cluster corresponding to the first target object, the processor is further configured to: Specific point clouds in the three-dimensional point cloud are removed, where the specific point clouds include ground point clouds.

42. The system of claim 36, wherein: Before clustering the three-dimensional point cloud to obtain a point cloud cluster corresponding to the first target object, the processor is further configured to: Specific point clouds in the three-dimensional point cloud are removed, where the specific point clouds include ground point clouds.

43. The system according to claim 37, characterized in that Before clustering the three-dimensional point cloud to obtain a point cloud cluster corresponding to the first target object, the processor is further configured to: Specific point clouds in the three-dimensional point cloud are removed, where the specific point clouds include ground point clouds.

44. The system according to any one of claims 28 to 35, characterized in that: The distance between the first target object and the movable platform is less than or equal to a first preset distance; The processor detects the point cloud cluster corresponding to the first target object through the target detection model, and after determining the object type of the first target object, is further configured to: If the first target object is determined to be a vehicle through the target detection model, determining a ground point cloud outside the first preset distance according to the position of the first target object; An object type of a second target object outside the first preset distance is determined according to the ground point cloud outside the first preset distance.

45. The system of claim 36, wherein: The distance between the first target object and the movable platform is less than or equal to a first preset distance; The processor detects the point cloud cluster corresponding to the first target object through the target detection model, and after determining the object type of the first target object, is further configured to: If the first target object is determined to be a vehicle through the target detection model, determining a ground point cloud outside the first preset distance according to the position of the first target object; An object type of a second target object outside the first preset distance is determined according to the ground point cloud outside the first preset distance.

46. ​​The system according to claim 44, characterized in that When the processor determines the ground point cloud outside the first preset distance according to the position of the first target object, it is specifically used to: Determining the slope of the ground where the first target object is located according to the position of the first target object; A ground point cloud outside the first preset distance is determined according to the slope of the ground.

47. The system according to claim 45, characterized in that When the processor determines the ground point cloud outside the first preset distance according to the position of the first target object, it is specifically used to: Determining the slope of the ground where the first target object is located according to the position of the first target object; A ground point cloud outside the first preset distance is determined according to the slope of the ground.

48. The system according to claim 46, characterized in that When the processor determines the slope of the ground where the first target object is located according to the position of the first target object, it is specifically used to: According to the positions of at least three of the first target objects, the slope of a plane formed by the at least three first target objects is determined, and the slope of the plane is the slope of the ground where the first target objects are located.

49. The system according to claim 47, characterized in that When the processor determines the slope of the ground where the first target object is located according to the position of the first target object, it is specifically used to: According to the positions of at least three of the first target objects, the slope of a plane formed by the at least three first target objects is determined, and the slope of the plane is the slope of the ground where the first target objects are located.

50. The system of claim 44, wherein: When the processor determines the object type of the second target object outside the first preset distance according to the ground point cloud outside the first preset distance, it is specifically used to: Determining, according to the ground point cloud outside the first preset distance, a point cloud cluster corresponding to a second target object outside the first preset distance, wherein the bottom of the second target object is in the same plane as the bottom of the first target object; The point cloud cluster corresponding to the second target object is detected by using a detection model corresponding to the distance of the second target object relative to the movable platform to determine the object type of the second target object.

51. The system according to any one of claims 45 to 49, characterized in that: When the processor determines the object type of the second target object outside the first preset distance according to the ground point cloud outside the first preset distance, it is specifically used to: Determining, according to the ground point cloud outside the first preset distance, a point cloud cluster corresponding to a second target object outside the first preset distance, wherein the bottom of the second target object is in the same plane as the bottom of the first target object; The point cloud cluster corresponding to the second target object is detected by using a detection model corresponding to the distance of the second target object relative to the movable platform to determine the object type of the second target object.

52. The system according to claim 50, characterized in that When the processor determines, based on the ground point cloud outside the first preset distance, a point cloud cluster corresponding to a second target object outside the first preset distance, the processor is specifically configured to: Clustering the three-dimensional point cloud outside the first preset distance after removing the ground point cloud to obtain a partial point cloud corresponding to the second target object; A point cloud cluster corresponding to the second target object is determined according to the partial point cloud corresponding to the second target object and the ground point cloud outside the first preset distance.

53. The system according to claim 51, characterized in that When the processor determines, based on the ground point cloud outside the first preset distance, a point cloud cluster corresponding to a second target object outside the first preset distance, the processor is specifically configured to: Clustering the three-dimensional point cloud outside the first preset distance after removing the ground point cloud to obtain a partial point cloud corresponding to the second target object; A point cloud cluster corresponding to the second target object is determined according to the partial point cloud corresponding to the second target object and the ground point cloud outside the first preset distance.

54. The system according to claim 52 or 53, characterized in that The processor is further configured to: If the second target object is a vehicle, and the width of the second target object is less than or equal to the first width, removing the three-dimensional point cloud whose height is greater than or equal to the first height in the point cloud cluster corresponding to the second target object, to obtain the remaining three-dimensional point cloud corresponding to the second target object; If the second target object is a vehicle, the width of the second target object is greater than the first width, and the width of the second target object is less than or equal to the second width, then removing the three-dimensional point cloud with a height greater than or equal to the second height in the point cloud cluster corresponding to the second target object to obtain the remaining three-dimensional point cloud corresponding to the second target object; Based on the remaining three-dimensional point cloud corresponding to the second target object, an identification box for characterizing the vehicle is generated, and the identification box is used for the movable platform to make navigation decisions; wherein the second width is greater than the first width, and the second height is greater than the first height.

55. A movable platform, characterized in that: include: body; A power system, installed on the fuselage, for providing movement power; And a target object detection system as described in any one of claims 28-54.

56. The movable platform according to claim 55, characterized in that: The movable platform includes: a drone, a movable robot or a vehicle.

57. A computer-readable storage medium, characterized in that A computer program is stored thereon, and the computer program is executed by a processor to implement the method according to any one of claims 1 to 27.

Images