Information processing method and device, computer equipment and storage medium

Abstract

The invention relates to the technical field of vehicles, and discloses an information processing method and device, computer equipment and a storage medium, and the method comprises the steps: determining a second pedestrian crossing object set from a first pedestrian crossing object set, each first pedestrian crossing object being detected by carrying out target detection on an environment image collected by a camera of a vehicle, the target second pedestrian crossing object is matched with at least one first pedestrian crossing object except the target second pedestrian crossing object, and the target second pedestrian crossing object is any second pedestrian crossing object; determining a main direction corresponding to the second pedestrian crossing object set according to the main direction of the bounding box of each second pedestrian crossing object, and determining a third pedestrian crossing object set from the second pedestrian crossing object set according to the main direction corresponding to the second pedestrian crossing object set; and generating a bounding box of the target pedestrian crossing object according to the bounding box of the third pedestrian crossing object set.

Description

Technical Field

[0001] The present application relates to the field of vehicle technology, and in particular to information processing methods, devices, computer equipment, and storage media. Background Art

[0002] Tasks like vehicle path planning and high-precision map creation all require information about crosswalks, a key road feature. The crosswalk object's bounding box, which indicates the crosswalk's outline, is the most crucial piece of information.

[0003] In related technologies, how to obtain a highly accurate bounding box of a crosswalk object used in a task becomes a technical problem that needs to be solved. Summary of the Invention

[0004] In view of this, embodiments of the present application provide an information processing method, apparatus, computer device, and storage medium.

[0005] In a first aspect, an embodiment of the present application provides an information processing method, the method comprising:

[0006] Determining a second crosswalk object set from the first crosswalk object set, wherein each first crosswalk object is detected by performing object detection on an environment image captured by a camera of the vehicle, the target second crosswalk object matches at least one first crosswalk object other than the target second crosswalk object, and the target second crosswalk object is any one of the second crosswalk objects;

[0007] determining a main direction corresponding to the second set of crosswalk objects based on a main direction of a bounding box of each second crosswalk object, and determining a third set of crosswalk objects from the second set of crosswalk objects based on the main direction corresponding to the second set of crosswalk objects;

[0008] A bounding box of a target crosswalk object is generated based on the bounding boxes of the third crosswalk object set.

[0009] In one possible implementation, determining the second crosswalk object set from the first crosswalk object set includes:

[0010] When a bounding box of the target first crosswalk object and a bounding box of at least one other first crosswalk object except the target first crosswalk object enclose the same area, determining the target first crosswalk object as the second crosswalk object, wherein the target first crosswalk object is any one of the first crosswalk objects;

[0011] The target first crosswalk object is determined as the second crosswalk object when there is a bounding box of another first crosswalk object whose distance from at least one of the following bounding boxes is less than a distance threshold: the bounding box of the target first crosswalk object, and the bounding box obtained by extending the bounding box of the target first crosswalk object in the main direction of the bounding box of the target first crosswalk object.

[0012] In one possible implementation, generating a bounding box of a target crosswalk object based on the bounding boxes of the third set of crosswalk objects includes:

[0013] Generate an initial bounding box according to the bounding box of the third crosswalk object set;

[0014] Based on the initial bounding box, at least one solving operation is performed to determine the bounding box of the target crosswalk object, wherein the bounding box targeted by the first solving operation is the initial bounding box, and the solving operation includes: when an error between the bounding box targeted by the solving operation and the bounding box of the third set of crosswalk objects satisfies an error threshold condition, determining the bounding box targeted by the solving operation as the bounding box of the target crosswalk object; when the error does not satisfy the error threshold condition, solving the bounding box targeted by the next solving operation based on the constraint condition and the bounding box targeted by the solving operation.

[0015] In a possible implementation, the constraints include: the height of the solved bounding box is the height of the initial bounding box, and the center point of the solved bounding box is in the main direction of the initial bounding box.

[0016] In one possible implementation, the error threshold condition includes: a total angle error of the bounding box for which the solution operation is performed is less than an angle error threshold, and a total distance error of the bounding box for which the solution operation is performed is less than a distance error threshold; and the method further includes:

[0017] Determining a total angular error of the bounding box for the solve operation based on a difference between a principal direction angle of the bounding box of each third crosswalk object and a principal direction angle of the bounding box for the solve operation;

[0018] determining a first distance between a first bounding edge of the edges of the bounding box of the third set of crosswalk objects and a first bounding edge of the bounding box for which the solve operation is performed, and a second distance between a second bounding edge of the edges of the bounding box of the third set of crosswalk objects and a second bounding edge of the bounding box for which the solve operation is performed;

[0019] The sum of the first distance and the second distance is determined as a total distance error for the bounding box for which the solve operation is performed.

[0020] In one possible implementation, generating an initial bounding box according to the bounding box of the third crosswalk object set includes:

[0021] Determine the position mean of the center points of the bounding boxes of the third set of crosswalk objects as the position of the center point of the initial bounding box;

[0022] Determine the mean of the main directions of the bounding boxes of the third set of crosswalk objects as the main direction of the initial bounding box;

[0023] Determine the mean height of the bounding boxes of the third set of crosswalk objects as the height of the initial bounding box;

[0024] For each vertex in the vertex set of the third crosswalk object set, project the vertex onto the target horizontal axis to obtain a projection point of the vertex, wherein the center point of the initial bounding box is on the target horizontal axis and the target horizontal axis is in the main direction of the initial bounding box;

[0025] The difference between the maximum and minimum values ​​of the coordinates of all the projection points on the target abscissa axis is determined as the width of the initial bounding box.

[0026] In a second aspect, an embodiment of the present application provides an information processing device, the information processing device comprising:

[0027] a first determining unit configured to determine a second crosswalk object set from the first crosswalk object set, wherein each first crosswalk object is detected by performing object detection on an environment image captured by a camera of the vehicle, the target second crosswalk object matches at least one first crosswalk object other than the target second crosswalk object, and the target second crosswalk object is any one of the second crosswalk objects;

[0028] a second determining unit, configured to determine a main direction corresponding to the second crosswalk object set based on the main direction of the bounding box of each second crosswalk object, and to determine a third crosswalk object set from the second crosswalk object set based on the main direction corresponding to the second crosswalk object set;

[0029] The generating unit is configured to generate a bounding box of a target crosswalk object according to the bounding boxes of the third crosswalk object set.

[0030] In one possible implementation, the first determination unit is further configured to determine the target first crosswalk object as the second crosswalk object when the bounding box of the target first crosswalk object and the bounding box of at least one other first crosswalk object other than the target first crosswalk object surround the same area, wherein the target first crosswalk object is any one of the first crosswalk objects; and to determine the target first crosswalk object as the second crosswalk object when there is a bounding box of other first crosswalk objects whose distance from at least one of the following bounding boxes is less than a distance threshold: the bounding box of the target first crosswalk object, and the bounding box obtained by extending the bounding box of the target first crosswalk object in the main direction of the bounding box of the target first crosswalk object.

[0031] In one possible implementation, the generation unit is further configured to generate an initial bounding box based on the bounding box of the third crosswalk object set; perform at least one solving operation based on the initial bounding box to determine the bounding box of the target crosswalk object, wherein the bounding box targeted by the first solving operation is the initial bounding box, and the solving operation includes: when the error between the bounding box targeted by the solving operation and the bounding box of the third crosswalk object set meets an error threshold condition, determining the bounding box targeted by the solving operation as the bounding box of the target crosswalk object; when the error does not meet the error threshold condition, solving the bounding box targeted by the next solving operation based on the constraint condition and the bounding box targeted by the solving operation.

[0032] In a possible implementation, the constraints include: the height of the solved bounding box is the height of the initial bounding box, and the center point of the solved bounding box is in the main direction of the initial bounding box.

[0033] In one possible implementation, the error threshold condition includes: a total angle error of the bounding box for which the solution operation is performed is less than an angle error threshold, and a total distance error of the bounding box for which the solution operation is performed is less than a distance error threshold; and the information processing apparatus further includes:

[0034] An error determination unit is configured to determine a total angular error of the bounding box targeted by the solution operation based on a difference between a main direction angle of the bounding box of each third crosswalk object and a main direction angle of the bounding box targeted by the solution operation; determine a first distance between a first boundary edge among the edges of the bounding box of the third crosswalk object set and a first boundary edge of the bounding box targeted by the solution operation, and a second distance between a second boundary edge among the edges of the bounding box of the third crosswalk object set and a second boundary edge of the bounding box targeted by the solution operation; and determine the sum of the first distance and the second distance as the total distance error of the bounding box targeted by the solution operation.

[0035] In one possible implementation, the generation unit is further used to determine the position mean of the center point of the bounding box of the third crosswalk object set as the position of the center point of the initial bounding box; determine the main direction mean of the bounding box of the third crosswalk object set as the main direction of the initial bounding box; determine the height mean of the bounding box of the third crosswalk object set as the height of the initial bounding box; for each vertex in the vertex set of the third crosswalk object set, project the vertex onto the target horizontal coordinate axis to obtain the projection point of the vertex, wherein the center point of the initial bounding box is on the target horizontal coordinate axis and the target horizontal coordinate axis is in the main direction of the initial bounding box; and determine the difference between the maximum and minimum values ​​of the coordinates of all the obtained projection points on the target horizontal coordinate axis as the width of the initial bounding box.

[0036] In a third aspect, an embodiment of the present application provides a computer device comprising: a memory and a processor, the memory and the processor being communicatively connected to each other, computer instructions being stored in the memory, and the processor executing the method of the first aspect or any corresponding embodiment thereof by executing the computer instructions.

[0037] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium having computer instructions stored thereon, the computer instructions being used to enable a computer to execute the method of the above-mentioned first aspect or any corresponding embodiment thereof.

[0038] In a fifth aspect, the present invention provides a computer program product comprising computer instructions for causing a computer to execute the method of the first aspect or any corresponding embodiment thereof.

[0039] The information processing method provided in embodiments of the present application determines a second set of crosswalk objects from a first set of crosswalk objects, wherein a target second crosswalk object matches at least one first crosswalk object other than the target second crosswalk object. The principal direction corresponding to the second crosswalk object set is determined based on the principal direction of the bounding box of each second crosswalk object. Furthermore, a third set of crosswalk objects is determined from the second set of crosswalk objects based on the principal direction of the second crosswalk object set. For each third crosswalk object, the third crosswalk object is both matched with at least one other crosswalk object and determined based on the principal direction corresponding to the second crosswalk object set, which reflects the principal direction of the entire second crosswalk object set. The third crosswalk objects have a high degree of correlation. Using the bounding box of the third crosswalk object set to generate a bounding box for the target crosswalk object can generate a highly accurate bounding box for the target crosswalk object. The target crosswalk object can then be applied to a corresponding task. Consequently, a relatively accurate bounding box for the crosswalk object used in the task can be generated. On the other hand, the generated bounding box of the target crosswalk object may reflect the overall outline of the crosswalk to which the target crosswalk object belongs. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] In order to more clearly illustrate the specific implementation methods of the present application or the technical solutions in the prior art, the following is a brief introduction to the drawings required for use in the specific implementation methods or the description of the prior art. Obviously, the drawings described below are some implementation methods of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0041] Figure 1 is a schematic diagram of an example of a bounding box of a target crosswalk object obtained by the information processing method provided by this application;

[0042] Figure 2 This is a flowchart of the information processing method provided by the embodiment of the present application;

[0043] Figure 3 is a flowchart of another information processing method provided in an embodiment of the present application;

[0044] Figure 4 is a diagram of distances between bounding edges of edges of a bounding box of a third set of crosswalk objects and bounding edges of a bounding box for which a solve operation is performed;

[0045] Figure 5 This is a schematic diagram of the effect of the center point of the bounding box targeted by the solution operation on the main direction of the initial bounding box;

[0046] Figure 6is a schematic diagram of the parameters of the bounding box for the solution operation;

[0047] Figure 7 It is a structural diagram of the computer device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0048] To make the purpose, technical solutions, and advantages of the embodiments of the present application more clear, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are 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 those skilled in the art without making creative efforts shall fall within the scope of protection of this application.

[0049] refer to Figure 1 , which shows a schematic diagram of an example of a bounding box of a target crosswalk object obtained by the information processing method provided by this application.

[0050] In this example, the first crosswalk object set includes six first crosswalk objects. Figure 1 The bounding boxes 101 of the six first crosswalk objects are shown. Among them, the areas enclosed by the bounding boxes 101 of the six first crosswalk objects all represent a part of the same crosswalk. It is often the case that the area enclosed by the crosswalk objects detected by the target detection network represents a part of the crosswalk, which is caused by the limited field of view of the camera and related obstacles such as barriers. For example, for a crosswalk involving multiple lanes in the same direction with a barrier, the area enclosed by the crosswalk objects detected by the target detection network represents a part of the crosswalk. For example, for a crosswalk crossing lanes in the forward and reverse directions, the area enclosed by the crosswalk objects detected by the target detection network represents a part of the crosswalk. In the related art, a bounding box enclosing a part of the crosswalk with a high confidence level is usually selected as the bounding box applied to the task, resulting in the crosswalk objects applied to the task being unable to accurately reflect the overall outline of the crosswalk.

[0051] Figure 1 Also shown are a road solid line object 102 , a road arrow object 103 , a road dotted line object 104 , and a median strip object 105 .

[0052] In this example, two third crosswalk objects are determined. Figure 1 Bounding boxes 106 of two third crosswalk objects are shown.

[0053] In this example, a target crosswalk object's bounding box 107 is generated based on the bounding boxes 106 of two third crosswalk objects. Target crosswalk object's bounding box 107 can reflect the overall outline of the crosswalk. By converting the coordinates of the target crosswalk object's vertices to a world coordinate system, the bounding box of the crosswalk represented by the target crosswalk object in the real world can be obtained.

[0054] refer to Figure 2 , which shows a flow chart of the information processing method provided in an embodiment of the present application.

[0055] It should be noted that, for each step, the coordinates of the point used when executing the step can be called the coordinates of the point in the target coordinate system. For each step, the point used when executing the step can be the center point of the bounding box, the vertex of the bounding box, etc.

[0056] In a possible implementation, the target coordinate system is an image coordinate system based on the environment image.

[0057] In another possible implementation, the target coordinate system is the base coordinate system. The origin of the base coordinate system (c x ′ , c y ′ ) is: the coordinates of the position mean of the center points of the bounding boxes of the first crosswalk object set, which is obtained by averaging the positions of the center points of the bounding boxes of the first crosswalk object set. The x-axis direction of the base coordinate system is: the main direction of the bounding box of the first crosswalk object whose bounding box has the largest width. After establishing the base coordinate system, the homogeneous transformation matrix T between the image coordinate system based on the environment image and the base coordinate system can be determined. Using the homogeneous transformation matrix T, the coordinates of the point in the image coordinate system based on the environment image are converted to the coordinates of the point in the base coordinate system. The homogeneous transformation matrix T can be expressed as:

[0058]

[0059] Among them, θ ′ It is the counterclockwise angle between the x-axis of the base coordinate system and the x-axis of the image coordinate system based on the environment image.

[0060] In step S201, a second crosswalk object set is determined from the first crosswalk object set.

[0061] It should be noted that the first crosswalk object set consists of each first crosswalk object.

[0062] For each first crosswalk object, the first crosswalk object is detected by performing object detection on an environment image captured by a camera of the vehicle.

[0063] The environment image is an image of the environment in which the vehicle is located. The environment image indicates the environment in which the vehicle is located.

[0064] Before step S201 , for each first crosswalk object in the first crosswalk object set, the first crosswalk object is detected by performing target detection on the environment image using a target detection network, such as a target detection network of the YOLO series.

[0065] Before step S201 , for each first crosswalk object in the first crosswalk object set, the object detection network outputs a detection result for the first crosswalk object.

[0066] For each first crosswalk object in the first crosswalk object set, the detection result for the first crosswalk object includes: a bounding box of the first crosswalk object.

[0067] It should be noted that the bounding box of the crosswalk object is a rectangular box. The area enclosed by the bounding box of the crosswalk object is in the environment image. The outline of the bounding box of the crosswalk object can be the outline of the area enclosed by the bounding box of the crosswalk object. The area enclosed by the bounding box of the crosswalk object is the area of ​​the bounding box of the crosswalk object.

[0068] In the embodiment of the present application, the width of the bounding box of the crosswalk object may be: the length of the longest side of the bounding box of the crosswalk object.

[0069] In the embodiment of the present application, the height of the bounding box of the crosswalk object may be: the length of the shortest side of the bounding box of the crosswalk object.

[0070] In the embodiment of the present application, the main direction of the bounding box of the crosswalk object may be: the direction in which the longest side of the bounding box of the crosswalk object is located.

[0071] That is, the longest side of the bounding box of the crosswalk object is in the main direction of the bounding box of the crosswalk object.

[0072] In the embodiment of the present application, the target second crosswalk object matches at least one first crosswalk object other than the target second crosswalk object.

[0073] The target second crosswalk object is any second crosswalk object.

[0074] For example, the first crosswalk object set consists of the first crosswalk object 1, the first crosswalk object 2, the first crosswalk object 3, the first crosswalk object 4, the first crosswalk object 5, and the first crosswalk object 6. If the first crosswalk object 1 matches at least one of the first crosswalk object 2, the first crosswalk object 3, the first crosswalk object 4, the first crosswalk object 5, and the first crosswalk object 6, the first crosswalk object 1 is used as the second crosswalk object.

[0075] In one possible implementation, in step S201, when the bounding box of the target first crosswalk object and the bounding box of at least one other first crosswalk object other than the target first crosswalk object enclose the same area, the target first crosswalk object is determined to be the second crosswalk object, wherein the target first crosswalk object is any one of the first crosswalk objects.

[0076] The target first crosswalk object is any first crosswalk object.

[0077] In another possible implementation, in step S201 , when there are other first crosswalk objects whose distance from the bounding box of the target first crosswalk object is less than a distance threshold, the target first crosswalk object is determined as the second crosswalk object.

[0078] In the embodiment of the present application, the distance between the bounding box of the crosswalk object A and the bounding box of the crosswalk object B may be: the minimum value of the distances between the vertices of the bounding box of the crosswalk object A and the vertices of the crosswalk object B.

[0079] Among them, the pedestrian crossing object A and the pedestrian crossing object B are any two pedestrian crossing objects.

[0080] In step S202, the main direction corresponding to the second crosswalk object set is determined according to the main direction of the bounding box of each second crosswalk object, and the third crosswalk object set is determined from the second crosswalk object set according to the main direction corresponding to the second crosswalk object set.

[0081] The second crosswalk object set is composed of each second crosswalk object.

[0082] It should be noted that the main direction corresponding to the second crosswalk object set can be understood as: the overall main direction of the second crosswalk object set.

[0083] In one possible implementation of determining the main direction corresponding to the second crosswalk object set, in step S202, a main direction angle of each second crosswalk object is determined, and a mean of the main direction angles of the second crosswalk object set is determined. The mean is calculated as: the sum of the main direction angles of the second crosswalk object set divided by the number of second crosswalk objects. The main direction having a main direction angle equal to the mean is determined as the main direction corresponding to the second crosswalk object set.

[0084] It should be noted that, in the embodiment of the present application, the main direction angle of the crosswalk object may refer to: the angle between the main direction of the crosswalk object and a corresponding direction of a coordinate axis of the target coordinate system, such as the horizontal axis.

[0085] In other words, the main direction angle of the crosswalk object may refer to the angle between the side of the crosswalk object in the main direction and a coordinate axis of the target coordinate system, such as the horizontal axis.

[0086] In another possible implementation of determining the principal direction corresponding to the second crosswalk object set, in step S202, a principal direction unit vector is generated for each second crosswalk object. The principal direction unit vector of the second crosswalk object is a unit vector having a direction corresponding to the principal direction of the second crosswalk object. The mean of the principal direction unit vectors of the second crosswalk object set is calculated to obtain a mean vector, and the direction of the mean vector is determined as the principal direction corresponding to the second crosswalk object set.

[0087] In step S202, a third crosswalk object set is determined from the second crosswalk object set according to the main direction corresponding to the second crosswalk object set.

[0088] In one possible implementation, for each second crosswalk object, the angle between the main direction of the bounding box of the second crosswalk object and the main direction corresponding to the second crosswalk object set is determined; when the angle is less than an angle threshold, such as 30°, the second crosswalk object is determined to be a third crosswalk object.

[0089] In step S203, a bounding box of the target crosswalk object is generated based on the bounding boxes of the third crosswalk object set.

[0090] In a possible implementation of generating the bounding box of the target crosswalk object, in step S203 , a minimum bounding rectangle surrounding each third crosswalk object is generated, and the minimum bounding rectangle surrounding each third crosswalk object is determined as the bounding box of the target crosswalk object.

[0091] In another possible implementation of generating the bounding box of the target crosswalk object, in step S203, the position mean of the center points of the bounding boxes of the third crosswalk object set is determined as the position of the center point of the bounding box of the target crosswalk object. This position mean is obtained by averaging the positions of the center points of the bounding boxes of the third crosswalk object set. The main direction of the bounding box of the target crosswalk object is determined as the mean of the main directions of the bounding box of the third crosswalk object set. The height of the bounding box of the target crosswalk object is determined as the mean of the heights of the bounding boxes of the third crosswalk object set. The width of the bounding box of the target crosswalk object is determined as the difference between the maximum and minimum values ​​of the horizontal coordinates of the vertex set of the third crosswalk object set in the target coordinate system.

[0092] The vertex set of the third crosswalk object set includes: a vertex set of each third crosswalk object, and the vertex set of the third crosswalk object consists of four vertices of a bounding box of the third crosswalk object.

[0093] The difference between the maximum value and the minimum value is: the maximum value minus the minimum value.

[0094] To determine the mean principal direction of the bounding box of the third-party crosswalk object set, the principal direction angle of each third-party crosswalk object can be determined, and the mean principal direction angle of the third-party crosswalk object set can be determined. The mean is calculated as: the sum of the principal direction angles of the third-party crosswalk object set divided by the number of third-party crosswalk objects. The principal direction having the principal direction angle equal to the mean is determined as the mean principal direction of the bounding box of the third-party crosswalk object set.

[0095] refer to Figure 3 , which shows a flow chart of another information processing method provided in an embodiment of the present application.

[0096] In step S301, a second crosswalk object set is determined from the first crosswalk object set.

[0097] Step S301 includes steps S3011 - S3012 .

[0098] In step S3011, when the bounding box of the target first crosswalk object and the bounding box of at least one other first crosswalk object except the target first crosswalk object enclose the same area, the target first crosswalk object is determined as a second crosswalk object.

[0099] The target first crosswalk object is any first crosswalk object.

[0100] It should be noted that the first crosswalk objects other than the target first crosswalk object are referred to as other first crosswalk objects.

[0101] In step S3012, the target first crosswalk object is determined as the second crosswalk object when there is a bounding box of another first crosswalk object whose distance from at least one of the following bounding boxes is less than a distance threshold: the bounding box of the target first crosswalk object, and the bounding box obtained by extending the bounding box of the target first crosswalk object in the main direction of the bounding box of the target first crosswalk object.

[0102] A bounding box obtained by extending the bounding box of the target first crosswalk object in the main direction of the bounding box of the target first crosswalk object may be referred to as an extended bounding box of the target first crosswalk object.

[0103] The length of a bounding box obtained by extending the bounding box of the target first crosswalk object in a principal direction of the bounding box of the target first crosswalk object is greater than the length of the bounding box of the target first crosswalk object. The difference between the length of the bounding box obtained by extending the bounding box of the target first crosswalk object in the principal direction of the bounding box of the target first crosswalk object and the length of the bounding box of the target first crosswalk object is: the predetermined extension length.

[0104] In step S3012, if there are other first crosswalk objects whose distances to the bounding box of the target first crosswalk object are less than the distance threshold and / or there are other first crosswalk objects whose distances to the extended bounding box of the target first crosswalk object are less than the distance threshold, the target first crosswalk object is determined as the second crosswalk object.

[0105] Through steps S3011 - S3012 , the matching relationship between the first crosswalk objects can be determined more comprehensively, thereby improving the accuracy of the determined second crosswalk objects.

[0106] In step S302, the main direction corresponding to the second crosswalk object set is determined based on the main direction of the bounding box of each second crosswalk object, and the third crosswalk object set is determined from the second crosswalk object set based on the main direction corresponding to the second crosswalk object set.

[0107] The process of step S302 refers to step S202.

[0108] In step S303, an initial bounding box is generated based on the bounding boxes of the third set of crosswalk objects; and at least one solving operation is performed based on the initial bounding box to determine the bounding box of the target crosswalk object.

[0109] The bounding box targeted by the first solving operation is the initial bounding box.

[0110] The i-th solution operation includes: when the error between the bounding box targeted by the i-th solution operation and the bounding box of the third crosswalk object set meets the error threshold condition, determining the bounding box targeted by the i-th solution operation as the bounding box corresponding to the target crosswalk object; when the error between the bounding box targeted by the i-th solution operation and the bounding box of the third crosswalk object set meets the error threshold condition, solving the bounding box targeted by the next solution operation, i.e., the (i+1)-th solution operation, based on the constraint conditions and the bounding box targeted by the i-th solution operation.

[0111] The i-th solution operation is any solution operation except the last solution operation.

[0112] The bounding box targeted by the (i+1)th solving operation is represented by the parameters of the bounding box targeted by the (i+1)th solving operation. Solving the bounding box targeted by the (i+1)th solving operation may refer to solving the parameters of the bounding box targeted by the (i+1)th solving operation.

[0113] The parameters of the bounding box targeted by the (i+1)th solution operation include: the coordinates of the position of the center point of the bounding box targeted by the (i+1)th solution operation, the height of the bounding box targeted by the (i+1)th solution operation, the width of the bounding box targeted by the (i+1)th solution operation, and the main direction of the bounding box targeted by the (i+1)th solution operation.

[0114] In one possible implementation, the error between the bounding box targeted by the solution operation and the bounding box of the third crosswalk object set is: the total angular error of the bounding box targeted by the solution operation, and the error threshold condition is: the total angular error of the bounding box targeted by the solution operation is less than the angular error threshold.

[0115] In another possible implementation, the error between the bounding box targeted by the solution operation and the bounding box of the third crosswalk object set is: the total distance error of the bounding box targeted by the solution operation, and the error threshold condition is: the total distance error of the bounding box targeted by the solution operation is less than the distance error threshold.

[0116] In another possible implementation, the error between the bounding box targeted by the solution operation and the bounding box of the third set of crosswalk objects includes: a total angular error of the bounding box targeted by the solution operation, and a total distance error of the bounding box targeted by the solution operation. The error threshold condition includes: the total angular error of the bounding box targeted by the solution operation is less than an angular error threshold, and the total distance error of the bounding box targeted by the solution operation is less than a distance error threshold.

[0117] In the embodiment of the present application, the total angle error of the bounding box targeted by the solving operation may be determined based on the difference between the main direction angle of the bounding box of each third crosswalk object and the main direction angle of the bounding box targeted by the solving operation.

[0118] In an embodiment of the present application, a first distance between a first boundary edge of the edge of the bounding box of the third crosswalk object set and a first boundary edge of the bounding box targeted by the solution operation, and a second distance between a second boundary edge of the edge of the bounding box of the third crosswalk object set and a second boundary edge of the bounding box targeted by the solution operation can be determined; the sum of the first distance and the second distance is determined as the total distance error of the bounding box targeted by the solution operation.

[0119] When performing the i-th solve operation, the total angle error of the bounding box targeted by the i-th solve operation is determined based on the difference between the main direction angle of the bounding box of each third crosswalk object and the main direction angle of the bounding box targeted by the i-th solve operation.

[0120] When performing the i-th solve operation, determine a first distance between a first boundary edge of the edges of the bounding box of the third crosswalk object set and the first boundary edge of the bounding box targeted by the i-th solve operation, and a second distance between a second boundary edge of the edges of the bounding box of the third crosswalk object set and the second boundary edge of the bounding box targeted by the i-th solve operation; and determine the sum of the first distance and the second distance as the total distance error of the bounding box targeted by the i-th solve operation.

[0121] In an embodiment of the present application, when solving the bounding box of the target crosswalk object, both the angle-related error, i.e., the total angle error of the bounding box targeted by the solution operation, and the distance-related error, i.e., the total distance error of the bounding box targeted by the solution operation, are taken into consideration, and the degree of difference between the solved bounding box and the bounding box of the third set of crosswalk objects is measured more comprehensively. Only when the total angle error and the total distance error of the bounding box targeted by the solution operation are both less than the corresponding thresholds, the bounding box targeted by the solution operation is determined as the bounding box of the target crosswalk object, thereby improving the accuracy of the generated bounding box of the target crosswalk object.

[0122] In this embodiment of the present application, the sum of the differences between the main direction angle of the bounding box of each third crosswalk object and the main direction angle of the bounding box for which the solution operation is performed can be determined as the total angular error of the bounding box for which the solution operation is performed. The total angular error of the bounding box for which the solution operation is performed can also be calculated using the following formula:

[0123]

[0124] Where e1 represents the total angular error of the bounding box for which the solution is performed, θ represents the main direction angle of the bounding box for which the solution is performed, and θ i represents the main direction angle of the bounding box of the i-th third crosswalk object, and n represents the number of third crosswalk objects.

[0125] The total distance error for the bounding box being solved can be expressed as:

[0126] e2=d1+d2

[0127] Wherein, e2 represents the total distance error of the bounding box for which the solution operation is performed, d1 represents a first distance between a first bounding edge of the edges of the bounding box of the third crosswalk object set and the first bounding edge of the bounding box for which the solution operation is performed, and d2 represents a second distance between a second bounding edge of the edges of the bounding box of the third crosswalk object set and the second bounding edge of the bounding box for which the solution operation is performed.

[0128] In the embodiment of the present application, for side A and side B, the smallest distance between the vertices of the two sides can be determined as the distance between side A and side B. Alternatively, a perpendicular line can be drawn from the midpoint of side A to side B, and the length of the perpendicular line can be determined as the distance between side A and side B.

[0129] Wherein, side A and side B are any two sides.

[0130] In an embodiment of the present application, in order to determine the first boundary edge in the edge of the bounding box of the third crosswalk object set and the second boundary edge in the edge of the bounding box of the third crosswalk object set, it is possible to determine the vertex to which the smallest coordinate among the coordinates of the vertex set of the third crosswalk object on the x-axis of the target coordinate system belongs and the vertex to which the largest coordinate among the coordinates of the vertex set of the third crosswalk object on the x-axis of the target coordinate system belongs.

[0131] The first boundary edge among the edges of the bounding box of the third crosswalk object set may be an edge having as an endpoint the vertex to which the smallest coordinate among the coordinates of the vertex set on the x-axis of the target coordinate system belongs. In other words, the vertex to which the smallest coordinate among the coordinates of the vertex set on the x-axis of the target coordinate system belongs is on the first boundary edge.

[0132] It should be noted that the first boundary edge can be called the left boundary edge.

[0133] The second boundary edge among the edges of the bounding box of the third crosswalk object set may be an edge having as its endpoint the vertex to which the largest coordinate among the vertex set's coordinates on the x-axis of the target coordinate system belongs. In other words, the vertex to which the largest coordinate among the vertex set's coordinates on the x-axis of the target coordinate system belongs is on the second boundary edge.

[0134] It should be noted that the second boundary edge can be called the right boundary edge.

[0135] The first boundary edge of the bounding box targeted by the solve operation may be an edge having the following vertex as an endpoint: the vertex having the smallest coordinate among the vertex set of the bounding box targeted by the solve operation on the x-axis of the target coordinate system.

[0136] The vertex set of the bounding box targeted by the solve operation consists of the four vertices of the bounding box targeted by the solve operation.

[0137] The second boundary edge of the bounding box targeted by the solve operation may be an edge having the following vertex as an endpoint: the vertex having the largest coordinate among the vertex set of the bounding box targeted by the solve operation on the x-axis of the target coordinate system.

[0138] refer to Figure 4 , which is a schematic diagram showing the distance between the bounding edge of the edge of the bounding box of the third set of crosswalk objects and the bounding edge of the bounding box for which the solve operation is performed.

[0139] In this example, two third crosswalk objects are determined.

[0140] Figure 4 Two third crosswalk object's bounding boxes 401 and solved bounding boxes 402 are shown.

[0141] Figure 4 The distance d1 between the left edge of the bounding box of the third crosswalk object set and the left edge of the solved bounding box 402 is shown.

[0142] Figure 4 A distance d2 between the right edge of the bounding box of the third crosswalk object set and the right edge of the solved bounding box 402 is shown.

[0143] In step S303, a nonlinear optimization solver such as G2O may be called to solve the bounding box for the next solving operation, ie, the (i+1)th solving operation, based on the constraints and the bounding box for the i-th solving operation.

[0144] In an embodiment of the present application, in step S303, the bounding box of the target crosswalk object can be determined by solving the initial bounding box. The error between the solved bounding box of the target crosswalk object and the bounding box of the third crosswalk object set satisfies an error threshold condition. Only when the error between the solved bounding box and the bounding box of the third crosswalk object set satisfies the error threshold condition will the solved bounding box be determined as the bounding box of the target crosswalk object. This improves the accuracy of the solved bounding box of the target crosswalk object.

[0145] In the embodiment of the present application, the constraint condition indicates a condition that the solved bounding box needs to meet. The constraint condition may be pre-set.

[0146] In one possible implementation, the constraint condition includes: the height of the bounding box targeted by the solving operation is the height of the initial bounding box.

[0147] If, when performing the i-th solve operation, the bounding box for the next solve operation, i.e., the i+1-th solve operation, is solved based on the constraints and the bounding box targeted by the i-th solve operation, then the height of the bounding box targeted by the i+1-th solve operation is the height of the initial bounding box.

[0148] In another possible implementation, the constraints include: the height of the solved bounding box is the height of the initial bounding box, and the center point of the solved bounding box is in the main direction of the initial bounding box.

[0149] If, when performing the i-th solve operation, the bounding box for the next solve operation, i.e., the i+1-th solve operation, is solved based on the constraints and the bounding box targeted by the i-th solve operation, then the height of the bounding box targeted by the i+1-th solve operation is the height of the initial bounding box, and the center point of the bounding box targeted by the i+1-th solve operation is in the main direction of the initial bounding box.

[0150] In step S303, if, when performing the i-th solving operation, the error between the bounding box targeted by the i-th solving operation and the bounding box of the third crosswalk object set meets the error threshold condition, the bounding box targeted by the i-th solving operation is determined as the bounding box of the target crosswalk object.

[0151] refer to Figure 5 , which shows a schematic diagram of the effect of the center point of the bounding box targeted by the solving operation on the main direction of the initial bounding box.

[0152] Figure 5 The black dot 501 shown represents the center point of the initial bounding box. Figure 5 Solved bounding box 502 and solved bounding box 503 are shown. Figure 5 The lines in the solved bounding box 502 are shown as line segments in the main direction of the solved bounding box 502 . Figure 5 The line segments in the solved bounding box 503 are shown as line segments in the main direction of the solved bounding box 503. The line segments in the main direction of the solved bounding box 502 and the lines in the main direction of the solved bounding box 503 all pass through the black dot, which is the center point of the initial bounding box. Figure 5 The black dots shown are intersection points of the line segments in the main direction of the solved bounding box 502 and the line segments in the main direction of the solved bounding box 503 .

[0153] In an embodiment of the present application, if the constraints include: the height of the bounding box targeted by the solution operation is the height of the initial bounding box, and the center point of the bounding box targeted by the solution operation is in the main direction of the initial bounding box, then the height of the bounding box targeted by the solution operation is a fixed height, namely the height of the initial bounding box. Solving the coordinates of the four vertices of a bounding box is equivalent to solving the width of the bounding box, solving the main direction angle of the bounding box, and solving the coordinates of the position of the center point of the bounding box. Solving the coordinates of the position of the center point of a bounding box is equivalent to solving the distance between the position of the center point of the bounding box and the position of the center point of the initial bounding box. Therefore, the number of parameters of the bounding box that need to be solved is only three, namely, only the width of the bounding box, the main direction angle of the bounding box, and the distance between the position of the center point of the bounding box and the position of the center point of the initial bounding box are solved, which saves the amount of calculation for solving and improves the speed of solving the bounding box of the target crosswalk object.

[0154] refer to Figure 6 , which shows a schematic diagram of the parameters of the bounding box for the solve operation.

[0155] Figure 6 A representation of the four vertices of the bounding box for which the solve operation is performed is shown, namely 1, 2, 3, 4.

[0156] Figure 6 It is shown that the height of the bounding box for which the solving operation is performed, ie, the height h of the initial bounding box is fixed. Figure 6 The width w of the bounding box targeted by the solving operation, the main direction angle θ of the bounding box targeted by the solving operation, and the coordinates (c x , c y ), solve the r between the position of the center point of the bounding box and the position of the center point of the initial bounding box. Figure 6 The coordinates of the center point of the initial bounding box are also shown (s x , s y ).

[0157] The main direction angle θ of the bounding box targeted by the solution operation may refer to an angle between a side in the main direction of the bounding box targeted by the solution operation and the abscissa axis of the target coordinate system, ie, the x-axis.

[0158] In the embodiment of the present application, the coordinates of the four vertices of the solved bounding box can be calculated using the following formula:

[0159] In this embodiment of the present application, the coordinates of the center point of the bounding box in the target coordinate system can be expressed as:

[0160] c x =s x +rcosθ.

[0161] In this embodiment of the present application, the coordinates of the center point of the bounding box on the y-axis in the target coordinate system can be expressed as:

[0162] c y =s y +rcosθ.

[0163] In the embodiment of the present application, the rotation matrix corresponding to the bounding box for which the solution operation is performed can be expressed as:

[0164]

[0165] In the embodiment of the present application, the translation vector corresponding to the bounding box targeted by the solution operation can be expressed as:

[0166]

[0167] In the embodiment of the present application, the four vertices of the bounding box targeted by the solving operation may be recorded as p1, p2, p3, and p4.

[0168] The coordinates of p1 can be expressed as:

[0169]

[0170] The coordinates of p2 can be expressed as:

[0171]

[0172] The coordinates of p3 can be expressed as:

[0173]

[0174] The coordinates of p4 can be expressed as:

[0175]

[0176] In a possible implementation of generating the initial bounding box, step S303 includes: step S3031.

[0177] In step S3031, the position mean of the center point of the bounding box of the third crosswalk object set is determined as the position of the center point of the initial bounding box; the main direction mean of the bounding box of the third crosswalk object set is determined as the main direction of the initial bounding box; the height mean of the bounding box of the third crosswalk object set is determined as the height of the initial bounding box; for each vertex in the vertex set of the third crosswalk object set, the vertex is projected onto the target horizontal coordinate axis to obtain the projection point of the vertex, wherein the center point of the initial bounding box is on the target horizontal coordinate axis and the target horizontal coordinate axis is in the main direction of the initial bounding box; the difference between the maximum value and the minimum value of the coordinates of all the obtained projection points on the target horizontal coordinate axis is determined as the width of the initial bounding box.

[0178] In this embodiment of the present application, in step S3031, the parameters of the bounding boxes of all third crosswalk objects are used to determine an initial bounding box that reflects the overall parameters. This initial bounding box is made closer to the bounding box of the target crosswalk object, which can reduce the number of solving operations.

[0179] For each third crosswalk object, the vertex set of the third crosswalk object set includes the four vertices of the bounding box of the third crosswalk object. Alternatively, the vertex set of the third crosswalk object set includes the vertex set of each third crosswalk object. The vertex set of the third crosswalk object consists of the four vertices of the bounding box of the third crosswalk object.

[0180] The bounding box of the third crosswalk object set includes: a bounding box of each third crosswalk object.

[0181] In step S3031, the position mean of the center point of the bounding box of the third crosswalk object set is determined. The position mean of the center point of the bounding box of the third crosswalk object set is represented by position mean coordinates. The position mean coordinates include: the mean coordinates on the horizontal axis, i.e., the x-axis, and the mean coordinates on the vertical axis, i.e., the y-axis. The mean coordinates on the x-axis are: the mean of the coordinates of the center point of the bounding box of the third crosswalk object set on the x-axis. The mean coordinates on the y-axis are: the mean of the coordinates of the center point of the bounding box of the third crosswalk object set on the y-axis.

[0182] In step S3031, the mean of the main directions of the bounding boxes of the third crosswalk object set is determined, and the mean of the main directions of the bounding boxes of the third crosswalk object set is determined as the main direction of the initial bounding box.

[0183] In one possible implementation of determining the mean principal direction of the bounding box of the third crosswalk object set, in step S3031, the principal direction angle of each third crosswalk object is determined, and a mean principal direction angle of the third crosswalk object set is determined. The mean principal direction angle is calculated by dividing the sum of the principal direction angles of the third crosswalk object set by the number of third crosswalk objects. The principal direction having the mean principal direction angle is determined as the mean principal direction of the bounding box of the third crosswalk object set.

[0184] In step S3031, the mean height of the bounding boxes of the third crosswalk object set is determined. The mean height of the bounding boxes of the third crosswalk object set is: the mean height of the bounding boxes of the third crosswalk object set. The sum of the heights of the bounding boxes of the third crosswalk object set is determined, and the sum is divided by the number of third crosswalk objects to obtain the mean height of the bounding boxes of the third crosswalk object set.

[0185] In step S3031, the maximum value and the minimum value of the coordinates of all the obtained projected points on the target abscissa axis are determined. The obtained projected points are composed of the projection points of each vertex in the vertex set of the third crosswalk object set. The difference between the maximum value and the minimum value is determined. The difference between the maximum value and the minimum value is: the maximum value minus the minimum value. The difference between the maximum value and the minimum value is determined as the width of the initial bounding box.

[0186] In another possible implementation of generating the initial bounding box, step S303 includes: step S3032. In step S3032, a minimum bounding rectangle surrounding each third crosswalk object is generated, and the minimum bounding rectangle surrounding each third crosswalk object is determined as the initial bounding box.

[0187] The embodiments of the present application provide an information processing device. The device is used to implement the above-mentioned embodiments and preferred embodiments, and the details already described will not be repeated. As used below, the term "unit" can refer to a combination of software and / or hardware that implements a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, implementation using hardware, or a combination of software and hardware, is also possible and contemplated.

[0188] The information processing device includes:

[0189] a first determining unit configured to determine a second crosswalk object set from the first crosswalk object set, wherein each first crosswalk object is detected by performing object detection on an environment image captured by a camera of the vehicle, the target second crosswalk object matches at least one first crosswalk object other than the target second crosswalk object, and the target second crosswalk object is any one of the second crosswalk objects;

[0190] a second determining unit, configured to determine a main direction corresponding to the second crosswalk object set based on the main direction of the bounding box of each second crosswalk object, and to determine a third crosswalk object set from the second crosswalk object set based on the main direction corresponding to the second crosswalk object set;

[0191] The generating unit is configured to generate a bounding box of a target crosswalk object according to the bounding boxes of the third crosswalk object set.

[0192] In one possible implementation, the first determination unit is further configured to determine the target first crosswalk object as the second crosswalk object when the bounding box of the target first crosswalk object and the bounding box of at least one other first crosswalk object other than the target first crosswalk object surround the same area, wherein the target first crosswalk object is any one of the first crosswalk objects; and to determine the target first crosswalk object as the second crosswalk object when there is a bounding box of other first crosswalk objects whose distance from at least one of the following bounding boxes is less than a distance threshold: the bounding box of the target first crosswalk object, and the bounding box obtained by extending the bounding box of the target first crosswalk object in the main direction of the bounding box of the target first crosswalk object.

[0193] In one possible implementation, the generation unit is further configured to generate an initial bounding box based on the bounding box of the third crosswalk object set; perform at least one solving operation based on the initial bounding box to determine the bounding box of the target crosswalk object, wherein the bounding box targeted by the first solving operation is the initial bounding box, and the solving operation includes: when the error between the bounding box targeted by the solving operation and the bounding box of the third crosswalk object set meets an error threshold condition, determining the bounding box targeted by the solving operation as the bounding box of the target crosswalk object; when the error does not meet the error threshold condition, solving the bounding box targeted by the next solving operation based on the constraint condition and the bounding box targeted by the solving operation.

[0194] In a possible implementation, the constraints include: the height of the solved bounding box is the height of the initial bounding box, and the center point of the solved bounding box is in the main direction of the initial bounding box.

[0195] In one possible implementation, the error threshold condition includes: a total angle error of the bounding box for which the solution operation is performed is less than an angle error threshold, and a total distance error of the bounding box for which the solution operation is performed is less than a distance error threshold; and the information processing apparatus further includes:

[0196] An error determination unit is configured to determine a total angular error of the bounding box targeted by the solution operation based on a difference between a main direction angle of the bounding box of each third crosswalk object and a main direction angle of the bounding box targeted by the solution operation; determine a first distance between a first boundary edge among the edges of the bounding box of the third crosswalk object set and a first boundary edge of the bounding box targeted by the solution operation, and a second distance between a second boundary edge among the edges of the bounding box of the third crosswalk object set and a second boundary edge of the bounding box targeted by the solution operation; and determine the sum of the first distance and the second distance as the total distance error of the bounding box targeted by the solution operation.

[0197] In one possible implementation, the generation unit is further used to determine the position mean of the center point of the bounding box of the third crosswalk object set as the position of the center point of the initial bounding box; determine the main direction mean of the bounding box of the third crosswalk object set as the main direction of the initial bounding box; determine the height mean of the bounding box of the third crosswalk object set as the height of the initial bounding box; for each vertex in the vertex set of the third crosswalk object set, project the vertex onto the target horizontal coordinate axis to obtain the projection point of the vertex, wherein the center point of the initial bounding box is on the target horizontal coordinate axis and the target horizontal coordinate axis is in the main direction of the initial bounding box; and determine the difference between the maximum and minimum values ​​of the coordinates of all the obtained projection points on the target horizontal coordinate axis as the width of the initial bounding box.

[0198] In this embodiment, the device is presented in the form of a functional unit, where the unit refers to an ASIC circuit, a processor and memory that executes one or more software or fixed programs, and / or other devices that can provide the above functions.

[0199] The further functional description of each of the above units is the same as that of the above corresponding embodiments and will not be repeated here.

[0200] refer to Figure 7 , which shows a schematic diagram of the structure of a computer device provided in an embodiment of the present application, the computer device includes: one or more processors 10, a memory 20, and interfaces for connecting various components, including high-speed interfaces and low-speed interfaces. The various components are connected to each other using different buses for communication, and can be installed on a common motherboard or installed in other ways as needed. The processor can process instructions executed in the computer device, including instructions stored in or on the memory to display graphical information of the GUI on an external input / output device (such as a display device coupled to the interface). In some optional embodiments, if necessary, multiple processors and / or multiple buses can be used together with multiple memories and multiple memories. Similarly, multiple computer devices can be connected, and each device provides part of the necessary operations (for example, as a server array, a group of blade servers, or a multi-processor system).

[0201] The processor 10 may be a central processing unit, a network processor, or a combination thereof. The processor 10 may further include a hardware chip. The hardware chip may be an application-specific integrated circuit, a programmable logic device, or a combination thereof. The programmable logic device may be a complex programmable logic device, a field programmable gate array, a general purpose array logic, or any combination thereof.

[0202] The memory 20 stores instructions that can be executed by at least one processor 10, so that the at least one processor 10 executes the method shown in the above embodiment.

[0203] The memory 20 may include a program storage area and a data storage area, wherein the program storage area may store an operating system and application programs required for at least one function; the data storage area may store data created based on the use of the computer device, etc. In addition, the memory 20 may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some optional embodiments, the memory 20 may optionally include a memory remotely located relative to the processor 10, and these remote memories may be connected to the computer device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0204] The memory 20 may include a volatile memory, such as a random access memory; the memory may also include a non-volatile memory, such as a flash memory, a hard disk or a solid-state drive; the memory 20 may also include a combination of the above types of memory.

[0205] The computer device further includes an input device 30 and an output device 40. The processor 10, the memory 20, the input device 30 and the output device 40 may be connected via a bus or other means.

[0206] The input device 30 can receive input digital or character information and generate key signal input related to user settings and function control of the computer device, such as a touch screen, a keypad, a mouse, a trackpad, a touch pad, an indicator stick, one or more mouse buttons, a trackball, a joystick, etc. The output device 40 can include a display device, an auxiliary lighting device (e.g., an LED), and a tactile feedback device (e.g., a vibration motor). The above-mentioned display device includes but is not limited to a liquid crystal display, a light emitting diode, a display, and a plasma display. In some optional embodiments, the display device can be a touch screen.

[0207] The embodiments of the present application also provide a computer-readable storage medium. The above-mentioned method according to the embodiment of the present application can be implemented in hardware, firmware, or implemented as a computer code that can be recorded in a storage medium, or implemented as a computer code that is originally stored in a remote storage medium or a non-temporary machine-readable storage medium and downloaded through a network and will be stored in a local storage medium, so that the method described herein can be stored in such software processing on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. Among them, the storage medium can be a magnetic disk, an optical disk, a read-only storage memory, a random access memory, a flash memory, a hard disk or a solid-state drive, etc.; further, the storage medium can also include a combination of the above-mentioned types of memory. It can be understood that a computer, a processor, a microprocessor controller or programmable hardware includes a storage component that can store or receive software or computer code. When the software or computer code is accessed and executed by a computer, a processor or hardware, the method shown in the above embodiment is implemented.

[0208] A portion of the embodiments of the present application may be applied as a computer program product, such as a computer program instruction, which, when executed by a computer, can call or provide the method and / or technical solution according to the present invention through the operation of the computer. Those skilled in the art should understand that the form in which the computer program instruction exists in a computer-readable medium includes but is not limited to a source file, an executable file, an installation package file, etc., and accordingly, the way in which the computer program instruction is executed by the computer includes but is not limited to: the computer directly executes the instruction, or the computer compiles the instruction and then executes the corresponding compiled program, or the computer reads and executes the instruction, or the computer reads and installs the instruction and then executes the corresponding installed program. Here, the computer-readable medium can be any available computer-readable storage medium or communication medium accessible to the computer.

[0209] Although the embodiments of the present application have been described with reference to the accompanying drawings, those skilled in the art may make various modifications and variations without departing from the spirit and scope of the present application, and such modifications and variations shall fall within the scope defined by the appended claims.

Claims

1. An information processing method, characterized in that: The method comprises: Determining a second crosswalk object set from the first crosswalk object set, wherein each first crosswalk object is detected by performing object detection on an environment image captured by a camera of the vehicle, the target second crosswalk object matches at least one first crosswalk object other than the target second crosswalk object, and the target second crosswalk object is any one of the second crosswalk objects; determining a main direction corresponding to the second set of crosswalk objects based on a main direction of a bounding box of each second crosswalk object, and determining a third set of crosswalk objects from the second set of crosswalk objects based on the main direction corresponding to the second set of crosswalk objects; A bounding box of a target crosswalk object is generated based on the bounding boxes of the third crosswalk object set.

2. The method according to claim 1, characterized in that Determining the second pedestrian crossing object set from the first pedestrian crossing object set includes: When a bounding box of the target first crosswalk object and a bounding box of at least one other first crosswalk object except the target first crosswalk object enclose the same area, determining the target first crosswalk object as the second crosswalk object, wherein the target first crosswalk object is any one of the first crosswalk objects; The target first crosswalk object is determined as the second crosswalk object when there is a bounding box of another first crosswalk object whose distance from at least one of the following bounding boxes is less than a distance threshold: the bounding box of the target first crosswalk object, and the bounding box obtained by extending the bounding box of the target first crosswalk object in the main direction of the bounding box of the target first crosswalk object.

3. The method according to claim 1 or 2, characterized in that Generating a bounding box of a target crosswalk object based on the bounding box of the third crosswalk object set includes: Generate an initial bounding box according to the bounding box of the third crosswalk object set; Based on the initial bounding box, at least one solving operation is performed to determine the bounding box of the target crosswalk object, wherein the bounding box targeted by the first solving operation is the initial bounding box, and the solving operation includes: when an error between the bounding box targeted by the solving operation and the bounding box of the third set of crosswalk objects satisfies an error threshold condition, determining the bounding box targeted by the solving operation as the bounding box of the target crosswalk object; when the error does not satisfy the error threshold condition, solving the bounding box targeted by the next solving operation based on the constraint condition and the bounding box targeted by the solving operation.

4. The method according to claim 3, characterized in that The constraints include: the height of the solved bounding box is the height of the initial bounding box, and the center point of the solved bounding box is in the main direction of the initial bounding box.

5. The method according to claim 3, characterized in that The error threshold condition includes: a total angle error of the bounding box for which the solution operation is performed is less than an angle error threshold, and a total distance error of the bounding box for which the solution operation is performed is less than a distance error threshold; and the method further includes: Determining a total angular error of the bounding box for the solve operation based on a difference between a principal direction angle of the bounding box of each third crosswalk object and a principal direction angle of the bounding box for the solve operation; determining a first distance between a first bounding edge of the edges of the bounding box of the third set of crosswalk objects and a first bounding edge of the bounding box for which the solve operation is performed, and a second distance between a second bounding edge of the edges of the bounding box of the third set of crosswalk objects and a second bounding edge of the bounding box for which the solve operation is performed; The sum of the first distance and the second distance is determined as a total distance error for the bounding box for which the solve operation is performed.

6. The method according to claim 3, characterized in that According to the bounding box of the third pedestrian crossing object set, generating an initial bounding box includes: Determine the position mean of the center points of the bounding boxes of the third set of crosswalk objects as the position of the center point of the initial bounding box; Determine the mean of the main directions of the bounding boxes of the third set of crosswalk objects as the main direction of the initial bounding box; Determine the mean height of the bounding boxes of the third set of crosswalk objects as the height of the initial bounding box; For each vertex in the vertex set of the third crosswalk object set, project the vertex onto the target horizontal axis to obtain a projection point of the vertex, wherein the center point of the initial bounding box is on the target horizontal axis and the target horizontal axis is in the main direction of the initial bounding box; The difference between the maximum and minimum values ​​of the coordinates of all the projection points on the target abscissa axis is determined as the width of the initial bounding box.

7. An information processing device, characterized in that The device comprises: a first determining unit configured to determine a second crosswalk object set from the first crosswalk object set, wherein each first crosswalk object is detected by performing object detection on an environment image captured by a camera of the vehicle, the target second crosswalk object matches at least one first crosswalk object other than the target second crosswalk object, and the target second crosswalk object is any one of the second crosswalk objects; a second determining unit, configured to determine a main direction corresponding to the second crosswalk object set based on the main direction of the bounding box of each second crosswalk object, and to determine a third crosswalk object set from the second crosswalk object set based on the main direction corresponding to the second crosswalk object set; The generating unit is configured to generate a bounding box of a target crosswalk object according to the bounding boxes of the third crosswalk object set.

8. A computer device, characterized in that: include: A memory and a processor, wherein the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the method according to any one of claims 1 to 6 by executing the computer instructions.

9. A computer-readable storage medium, characterized in that The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a computer to execute the method according to any one of claims 1 to 6.

10. A computer program product, characterized in that The method comprises computer instructions for causing a computer to execute the method according to any one of claims 1 to 6.

Images