Data processing method and device, computer equipment and storage medium

Abstract

The invention relates to the technical field of computers, and discloses a data processing method and device, computer equipment and a storage medium, and the method comprises the steps: executing at least one determination operation, and obtaining a target pitch angle of a camera for a target vehicle; according to the target pitch angle, two-dimensional points of lane lines related to the target vehicle are mapped to a three-dimensional space, and the determination operation comprises the steps that according to the pitch angle targeted by the determination operation, the two-dimensional points of each lane line in the first lane line and the second lane line are mapped to the three-dimensional space, and three-dimensional points, targeted by the determination operation, of each lane line are obtained; fitting the three-dimensional point of each lane line for the determination operation to obtain a first fitting line segment of the first lane line for the determination operation and a second fitting line segment of the second lane line for the determination operation; and according to a first point on the first fitting line segment and a second point corresponding to the first point on the second fitting line segment, determining a target pitch angle or a pitch angle for next determination operation.

Description

Technical Field

[0001] The present disclosure relates to the field of computer technology, and in particular to a data processing method, apparatus, computer equipment and storage medium. Background Art

[0002] Lane detection is a function of the Advanced Driving Assistance System (ADAS). Lane detection under the bird's eye view (BEV) has natural advantages over the front view, such as the lane detection results can be directly used to plan the route of the vehicle and control the vehicle. Therefore, three-dimensional (3D) point-based lane detection is widely used. Before obtaining the lane detection results based on the three-dimensional points, the two-dimensional (2D) points of the lane need to be mapped to the three-dimensional space to obtain the three-dimensional points of the lane for obtaining the three-dimensional lane detection results.

[0003] In the related art, the two-dimensional points of the detected lane lines are mapped to three-dimensional space through inverse perspective mapping (IPM) to obtain three-dimensional points used to obtain three-dimensional lane line results. However, inverse perspective transformation is an algorithm that assumes that the road the vehicle is traveling on is always flat. When a vehicle is driving, it is often the case that the road the vehicle is traveling on is not flat. As a result, the accuracy of the three-dimensional points used to obtain the three-dimensional lane line detection results is low, and the accuracy of the three-dimensional lane line detection results is low. How to improve the accuracy of the three-dimensional points used to obtain the three-dimensional lane line detection results has become a problem that needs to be solved. Summary of the invention

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

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

[0006] Perform at least one determination operation to obtain a target pitch angle of a camera for a target vehicle;

[0007] According to the target pitch angle, a two-dimensional point of a lane line associated with the target vehicle is mapped to a three-dimensional space to obtain a three-dimensional point for obtaining a lane line detection result, wherein the determination operation includes:

[0008] According to the pitch angle targeted by the determination operation, respectively mapping the two-dimensional points of each lane line in the first lane line and the second lane line associated with the target vehicle to the three-dimensional space to obtain the three-dimensional points of each lane line targeted by the determination operation;

[0009] Fitting the three-dimensional points of each lane line for the determination operation respectively to obtain a first fitting line segment of the first lane line for the determination operation and a second fitting line segment of the second lane line for the determination operation;

[0010] A target pitch angle or a pitch angle for a next determination operation is determined according to a first point on the first fitting line segment and a second point on the second fitting line segment corresponding to the first point, wherein the abscissa value of the first point is a sampled abscissa value obtained by sampling the abscissa value of the three-dimensional point of the first lane line for the determination operation.

[0011] In a possible implementation, determining a target pitch angle or a pitch angle for a next determination operation according to a first point on the first fitting line segment and a second point on the second fitting line segment corresponding to the first point includes:

[0012] Calculating the lateral distance difference between the lateral distance corresponding to each first point other than the target first point on the first fitting line segment and the lateral distance corresponding to the target first point on the first fitting line segment, to obtain a plurality of lateral distance differences, wherein the lateral distance corresponding to the first point on the first fitting line segment is determined according to the ordinate value of the first point on the first fitting line segment and the ordinate value of the second point on the second fitting line segment corresponding to the first point;

[0013] According to the plurality of lateral distance differences, a target pitch angle or a pitch angle for a next determination operation is determined.

[0014] In a possible implementation, determining the target pitch angle or the pitch angle for the next determination operation according to the multiple lateral distance differences includes:

[0015] determining whether an average value of absolute values ​​of the lateral distance differences among the plurality of lateral distance differences is less than an average value threshold;

[0016] If yes, determining the pitch angle targeted by the determination operation as a target pitch angle;

[0017] If not, determining candidate pitch angles for the determination operation according to the multiple lateral distance differences, and determining a target pitch angle or a pitch angle for the next determination operation according to the candidate pitch angles.

[0018] In one possible implementation, determining, based on the multiple lateral distance differences, a candidate pitch angle for which the determination operation is performed includes: determining whether a sum of the multiple lateral distance differences is less than 0; if so, determining a first pitch angle that is smaller than the pitch angle for which the determination operation is performed, and using the first pitch angle as the candidate pitch angle; if not, determining a second pitch angle that is greater than the pitch angle for which the determination operation is performed, and using the second pitch angle as the candidate pitch angle.

[0019] In one possible implementation, determining the target pitch angle or the pitch angle for the next determination operation based on the candidate pitch angle includes: when the number of times the determination operation is performed reaches a number threshold, determining the candidate pitch angle as the target pitch angle; when the number of times the determination operation is performed does not reach the number threshold, using the candidate pitch angle to back-project the three-dimensional points of each lane line for the determination operation to obtain multiple projected two-dimensional points of each lane line for the determination operation, and determining to use the candidate pitch angle as the target pitch angle or to use the candidate pitch angle as the pitch angle for the next determination operation based on the multiple projected two-dimensional points of each lane line for the determination operation.

[0020] In a possible implementation, the first lane line and the second lane line associated with the target vehicle are two adjacent lane lines belonging to the lane in which the target vehicle is traveling at the current moment.

[0021] In a second aspect, an embodiment of the present disclosure provides a data processing device, including:

[0022] a target pitch angle determination unit, configured to perform at least one determination operation to obtain a target pitch angle of a camera for a target vehicle;

[0023] A three-dimensional point acquisition unit is used to map the two-dimensional points of the lane line related to the target vehicle to the three-dimensional space according to the target pitch angle, so as to obtain the three-dimensional points used to obtain the lane line detection result, wherein the determination operation includes: mapping the two-dimensional points of each lane line of the first lane line and the second lane line related to the target vehicle to the three-dimensional space according to the pitch angle targeted by the determination operation, so as to obtain the three-dimensional points of each lane line targeted by the determination operation; fitting the three-dimensional points of each lane line targeted by the determination operation, so as to obtain the first fitting line segment of the first lane line targeted by the determination operation and the second fitting line segment of the second lane line targeted by the determination operation; determining the target pitch angle or the pitch angle targeted by the next determination operation according to the first point on the first fitting line segment and the second point on the second fitting line segment corresponding to the first point, wherein the abscissa value of the first point is the sampled abscissa value obtained by sampling the abscissa value of the three-dimensional point of the first lane line targeted by the determination operation.

[0024] In a possible implementation, the target pitch angle determination unit is further used to calculate the lateral distance difference between the lateral distance corresponding to each first point other than the target first point on the first fitting line segment and the lateral distance corresponding to the target first point on the first fitting line segment, to obtain multiple lateral distance differences, wherein the lateral distance corresponding to the first point on the first fitting line segment is determined according to the ordinate value of the first point on the first fitting line segment and the ordinate value of the second point corresponding to the first point on the second fitting line segment; and determine the target pitch angle or the pitch angle for the next determination operation according to the multiple lateral distance differences.

[0025] In one possible implementation, the target pitch angle determination unit is further used to determine whether the average of the absolute values ​​of the lateral distance differences among the multiple lateral distance differences is less than an average value threshold; if so, determine the pitch angle targeted by the determination operation as the target pitch angle; if not, determine the candidate pitch angles targeted by the determination operation based on the multiple lateral distance differences, and determine the target pitch angle or the pitch angle targeted by the next determination operation based on the candidate pitch angles.

[0026] In one possible implementation, the target pitch angle determination unit is further used to determine whether the sum of the multiple lateral distance differences is less than 0; if so, determine a first pitch angle that is smaller than the pitch angle targeted by the determination operation, and use the first pitch angle as the candidate pitch angle; if not, determine a second pitch angle that is greater than the pitch angle targeted by the determination operation, and use the second pitch angle as the candidate pitch angle.

[0027] In one possible implementation, the target pitch angle determination unit is further used to determine the candidate pitch angle as the target pitch angle when the number of times the determination operation is performed reaches a number threshold; when the number of times the determination operation is performed does not reach the number threshold, use the candidate pitch angle to back-project the three-dimensional points of each lane line for the determination operation to obtain multiple projected two-dimensional points of each lane line for the determination operation, and determine to use the candidate pitch angle as the target pitch angle or to use the candidate pitch angle as the pitch angle for the next determination operation based on the multiple projected two-dimensional points of each lane line for the determination operation.

[0028] In a possible implementation, the first lane line and the second lane line associated with the target vehicle are two adjacent lane lines belonging to the lane in which the target vehicle is traveling at the current moment.

[0029] In a third aspect, an embodiment of the present disclosure 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.

[0030] In a fourth aspect, an embodiment of the present disclosure 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 first aspect or any corresponding implementation manner thereof.

[0031] 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.

[0032] The data processing method provided by the embodiment of the present disclosure can obtain a relatively accurate target pitch angle of the camera of the target vehicle by performing at least one determination operation. The two-dimensional points of the lane line related to the target vehicle are mapped to the three-dimensional space through the relatively accurate target pitch angle of the camera of the target vehicle, and the accuracy of the three-dimensional points obtained for obtaining the lane line detection results is higher. Thus, the accuracy of the three-dimensional points used to obtain the three-dimensional lane line detection results is improved, and the accuracy of the three-dimensional lane line detection results is improved. Avoid the situation where the inverse perspective transformation assumes that the road the vehicle is driving on is always flat, resulting in a low accuracy of the three-dimensional points used to obtain the three-dimensional lane line detection results. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] In order to more clearly illustrate the specific embodiments of the present disclosure or the technical solutions in the prior art, the drawings required for use in the specific embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present disclosure. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0034] Figure 1 is a flowchart of a data processing method provided by an embodiment of the present disclosure;

[0035] Figure 2 is a schematic diagram of an example of a second point corresponding to a first point when the first fitting line segment is a straight line;

[0036] Figure 3 is a schematic diagram of an example of a second point corresponding to a first point when the first fitting line segment is a curve;

[0037] Figure 4 is a flowchart of another data processing method provided by an embodiment of the present disclosure;

[0038] Figure 5 is a flowchart of an example of obtaining a target pitch angle or a pitch angle for the next determination operation;

[0039] Figure 6 It is a schematic diagram of the structure of a computer device provided in an embodiment of the present disclosure. DETAILED DESCRIPTION

[0040] In order to make the purpose, technical solution and advantages of the embodiments of the present disclosure clearer, the technical solution in the embodiments of the present disclosure will be clearly and completely described below in conjunction with the drawings in the embodiments of the present disclosure. Obviously, the described embodiments are part of the embodiments of the present disclosure, rather than all the embodiments. Based on the embodiments in the present disclosure, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present disclosure.

[0041] refer to Figure 1 , which shows a flow chart of the data processing method provided by an embodiment of the present disclosure. It should be noted that during the lane line detection of the target vehicle, the lane line detection can be performed in sequence according to each image captured by the camera of the target vehicle. During the lane line detection according to any image captured by the camera of the target vehicle, steps S101-S102 can be executed. The target pitch angle does not specifically refer to a certain pitch angle, and the target pitch angle can be obtained each time step S101 is executed.

[0042] In step S101 , at least one determination operation is performed to obtain a target pitch angle of a camera for a target vehicle.

[0043] It can also be said that in step S101, the determination operation is repeatedly performed until the target pitch angle of the camera for the target vehicle is obtained.

[0044] The target vehicle may be any vehicle to which the data processing method provided in the embodiments of the present disclosure may be applied.

[0045] In the embodiment of the present disclosure, the first lane line and the second lane line related to the target vehicle may be: two adjacent lane lines on the road on which the target vehicle is traveling.

[0046] In a possible implementation, the first lane line and the second lane line associated with the target vehicle are two adjacent lane lines belonging to the lane in which the target vehicle is traveling at the current moment.

[0047] The current time may refer to the time when step S101 starts to be executed.

[0048] The i-th determination operation includes: step S1011 to step S1013.

[0049] The number of determination operations performed to obtain the target pitch angle is recorded as n.

[0050] The i-th determination operation is: performing any determination operation to obtain the target pitch angle.

[0051] If the pitch angle for the (i+1)th determination operation is determined by performing the (i+1)th determination operation, the (i+1)th determination operation is performed.

[0052] Step S1011 includes: step S10111-step S10112.

[0053] In step S10111, the target image captured by the camera of the target vehicle is input into a 2D lane line detection network for detecting two-dimensional points of lane lines, and the coordinates of each two-dimensional point of the first lane line and each two-dimensional point of the second lane line output by the 2D lane line detection network are obtained, thereby determining each two-dimensional point of the first lane line and each two-dimensional point of the second lane line. The coordinates of the two-dimensional point may be the coordinates of the two-dimensional point output by the 2D lane line detection network, which may be the coordinates of the two-dimensional point in the pixel coordinate system.

[0054] The target image includes: a lane line object representing each lane line of a plurality of lane lines on a lane where the target vehicle is traveling. The plurality of lane lines include: a first lane line and a second lane line related to the target vehicle.

[0055] The target image may be any image captured by a camera of the target vehicle during lane line detection of the target vehicle.

[0056] In step S10112, according to the pitch angle targeted by the i-th determination operation, the two-dimensional points of each lane line in the first lane line and the second lane line are respectively mapped to the three-dimensional space to obtain the three-dimensional points of each lane line in the first lane line and the second lane line targeted by the i-th determination operation, including: according to the pitch angle targeted by the i-th determination operation, each two-dimensional point of the first lane line is mapped to the three-dimensional space to obtain each three-dimensional point of the first lane line targeted by the i-th determination operation; according to the pitch angle targeted by the i-th determination operation, each two-dimensional point of the second lane line is mapped to the three-dimensional space to obtain each three-dimensional point of the second lane line targeted by the i-th determination operation.

[0057] For a two-dimensional point of the first lane line, the two-dimensional point is mapped to the three-dimensional space to obtain a three-dimensional point of the first lane line for the i-th determination operation, and the three-dimensional point corresponds to the two-dimensional point.

[0058] For a two-dimensional point of the second lane line, the two-dimensional point is mapped to the three-dimensional space to obtain a three-dimensional point of the second lane line for the i-th determination operation, and the three-dimensional point corresponds to the two-dimensional point.

[0059] It should be noted that a two-dimensional point may also be called a two-dimensional lane line point, i.e., a 2D lane line point, and a three-dimensional point may also be called a three-dimensional lane line point, i.e., a 3D lane line point.

[0060] Mapping a two-dimensional point to a three-dimensional space can be understood as: through the coordinate conversion relationship between the pixel coordinate system, the image coordinate system, the camera coordinate system, and the world coordinate system, the coordinates of the two-dimensional point are converted into the coordinates of the three-dimensional point in the world coordinate system corresponding to the two-dimensional point, thereby determining the three-dimensional point in the world coordinate system corresponding to the two-dimensional point, wherein the three-dimensional point in the world coordinate system is the three-dimensional point in the world coordinate system. The coordinates of the two-dimensional point can refer to the coordinates of the two-dimensional point in the pixel coordinate system.

[0061] It should be noted that the pitch angle targeted by the first determination operation is: the current pitch angle (pitch) of the camera targeted at the target vehicle, that is, the pitch angle of the camera targeted at the target vehicle at the current moment.

[0062] That is to say, if the i-th determination operation is the first determination operation, the pitch angle targeted by the i-th determination operation is: the current pitch angle of the camera targeted at the target vehicle.

[0063] In step S10112, according to the pitch angle targeted by the i-th determination operation, the camera extrinsic parameters of the camera of the target vehicle targeted by the i-th determination operation are determined.

[0064] In step S10112, in order to determine the camera extrinsic parameters of the target vehicle's camera for the i-th determination operation, the current camera extrinsic parameters of the target vehicle's camera, the current pitch angle of the target vehicle's camera, the current roll angle (roll) of the target vehicle's camera, that is, the roll angle of the target vehicle's camera at the current moment, and the current yaw angle (yaw) of the target vehicle's camera, that is, the yaw angle of the target vehicle's camera at the current moment are obtained.

[0065] The current camera extrinsic parameters of the target vehicle's camera include: the current rotation matrix of the target vehicle's camera and the current translation vector of the target vehicle's camera, that is, the translation vector of the target vehicle's camera at the current moment. The rotation matrix is ​​represented by R, and the translation vector is represented by T.

[0066] The current rotation matrix of the camera of the target vehicle is determined according to the current pitch angle of the camera of the target vehicle, the current roll angle of the camera of the target vehicle, and the current yaw angle of the camera of the target vehicle.

[0067] The pitch angle of the target vehicle, the roll angle of the target vehicle, and the yaw angle of the target vehicle may respectively represent an angle of rotation around the x-axis in the world coordinate system, an angle of rotation around the y-axis in the world coordinate system, and an angle of rotation around the z-axis in the world coordinate system.

[0068] The x-axis of the world coordinate system may be the longitudinal direction of the target vehicle, i.e., the front-to-back direction. The y-axis of the world coordinate system may be the lateral direction of the target vehicle, i.e., the left-to-right direction. The z-axis of the world coordinate system is perpendicular to the x-axis of the world coordinate system and perpendicular to the y-axis of the world coordinate system.

[0069] In step S10112, if the i-th determination operation is the first determination operation, the camera extrinsic parameters of the camera of the target vehicle for the i-th determination operation are: the current camera extrinsic parameters of the camera of the target vehicle.

[0070] In step S10112, if the i-th determination operation is not the first determination operation, the rotation matrix for the i-th determination operation is calculated according to the pitch angle for the i-th determination operation, the current roll angle for the camera of the target vehicle, and the current yaw angle for the camera of the target vehicle. The current rotation matrix of the camera of the target vehicle is replaced by the rotation matrix for the i-th determination operation to obtain the camera extrinsic parameters of the camera of the target vehicle for the i-th determination operation.

[0071] The camera external parameters of the camera of the target vehicle for the i-th determination operation include: a rotation matrix for the i-th determination operation and a current translation vector of the camera of the target vehicle.

[0072] In step S10112, based on the camera extrinsic parameters of the target vehicle's camera for the i-th determination operation, the intrinsic parameters of the target vehicle's camera, and the conversion relationship between the coordinates of the same two-dimensional point in the pixel coordinate system and the coordinates of the same two-dimensional point in the image coordinate system, each two-dimensional point of the first lane line is mapped to the three-dimensional space to obtain each three-dimensional point of the first lane line for the i-th determination operation.

[0073] For a two-dimensional point of the first lane line, the two-dimensional point is mapped to the three-dimensional space through the mapping, and a three-dimensional point of the first lane line for the i-th determination operation can be obtained, and the three-dimensional point corresponds to the two-dimensional point.

[0074] In step S10112, based on the camera extrinsic parameters of the target vehicle's camera for the i-th determination operation, the intrinsic parameters of the target vehicle's camera, and the conversion relationship between the coordinates of the same two-dimensional point in the pixel coordinate system and the coordinates of the same two-dimensional point in the image coordinate system, each two-dimensional point of the second lane line is mapped to the three-dimensional space to obtain each three-dimensional point of the second lane line for the i-th determination operation.

[0075] For each two-dimensional point of the second lane line, the two-dimensional point is mapped to the three-dimensional space through the mapping, and a three-dimensional point of the second lane line for the i-th determination operation can be obtained, and the three-dimensional point corresponds to the two-dimensional point.

[0076] In step S1012, the three-dimensional points of each lane line in the first lane line and the second lane line for the i-th determination operation are fitted respectively to obtain a first fitted line segment of the first lane line for the i-th determination operation and a second fitted line segment of the second lane line for the i-th determination operation, including: fitting the three-dimensional points of the first lane line for the i-th determination operation to obtain the first fitted line segment of the first lane line for the i-th determination operation; fitting the three-dimensional points of the second lane line for the i-th determination operation to obtain the second fitted line segment of the second lane line for the i-th determination operation.

[0077] In one possible implementation, in step S1012, for one of the first lane lines and the second lane lines associated with the target vehicle, a Ransac algorithm is used to fit the three-dimensional points of the lane line for the i-th determination operation to obtain a fitting line segment of the lane line for the i-th determination operation.

[0078] In step S1013, a target pitch angle or a pitch angle targeted by a next determination operation, i.e., an (i+1)th determination operation, is determined according to a first point on the first fitting line segment and a second point on the second fitting line segment corresponding to the first point.

[0079] For a first point on the first fitting line segment, the abscissa value of the first point is: a sampled abscissa value obtained by sampling the abscissa value of the three-dimensional point of the first lane line for the i-th determination operation.

[0080] That is to say, for a first point on the first fitting line segment, the first point is: a three-dimensional point on the first fitting line segment whose abscissa value of coordinates in the world coordinate system is the sampled abscissa value.

[0081] Each first point on the first fitting line segment has a different sampled abscissa value, and the number of sampled abscissa values ​​is equal to the number of first points on the first fitting line segment.

[0082] In step S1013, the horizontal coordinate values ​​of the coordinates of all three-dimensional points of the first lane line for the i-th determination operation in the world coordinate system may be sampled to obtain a plurality of sampled horizontal coordinate values.

[0083] In a possible implementation, the horizontal coordinate values ​​of the coordinates of all three-dimensional points of the first lane line in the world coordinate system for the i-th determination operation are averagely sampled. The difference between any two adjacent horizontal coordinate values ​​of the multiple sampled horizontal coordinate values ​​may be a preset difference.

[0084] In the embodiment of the present disclosure, for a first point on a first fitting line segment, if the first fitting line segment is a straight line, the abscissa value of a second point on the second fitting line segment corresponding to the first point is: the abscissa value of the first point.

[0085] The fact that the first fitting line segment is a straight line can be determined based on the fact that the derivative of the first fitting line segment at the first point is less than a derivative threshold.

[0086] That is to say, for a first point on the first fitting line segment, if the first fitting line segment is a straight line, the second point corresponding to the first point is: a three-dimensional point on the second fitting line segment whose abscissa value in the world coordinate system is the abscissa value of the first point.

[0087] refer to Figure 2 , which is a schematic diagram showing an example of a second point corresponding to a first point when the first fitting line segment is a straight line.

[0088] The first lane line associated with the target vehicle is the right lane line, and the second lane line associated with the target vehicle is the left lane line. The right lane line is the lane line on the right side of the target vehicle, which belongs to the lane in which the target vehicle is traveling at the current moment. The left lane line is the lane line on the left side of the target vehicle, which belongs to the lane in which the target vehicle is traveling at the current moment.

[0089] Fit all three-dimensional points of the right lane line for the i-th determination operation to obtain the first fitting line segment of the right lane line for the i-th determination operation. Fit all three-dimensional points of the left lane line for the i-th determination operation to obtain the second fitting line segment of the left lane line for the i-th determination operation.

[0090] Figure 2 Multiple first points on the first fitting line segment are shown, including a first point with coordinates (x1, y1, z1), a first point with coordinates (x2, y2, z2), and a first point with coordinates (x3, y3, z3).

[0091] Among them, the horizontal coordinate values ​​such as x1, x2, and x3 are the horizontal coordinate values ​​obtained by averaging the horizontal coordinate values ​​of the coordinates of all three-dimensional points of the first lane line for the i-th determination operation in the world coordinate system.

[0092] Figure 2 Multiple second points on the second fitting line segment are shown, including a second point with coordinates (x1, y1', z1'), a second point with coordinates (x2, y2', z2'), and a second point with coordinates (x3, y3', z3').

[0093] The second point with coordinates (x1, y1', z1') is the second point corresponding to the first point with coordinates (x1, y1, z1). The second point with coordinates (x2, y2', z2') is the second point corresponding to the first point with coordinates (x2, y2, z2). The second point with coordinates (x3, y3', z3') is the second point corresponding to the first point with coordinates (x3, y3, z3).

[0094] The horizontal distance corresponding to the first point with coordinates (x1, y1, z1) is: y1-y1'. The horizontal distance corresponding to the first point with coordinates (x2, y2, z2) is: y2-y2'. The horizontal distance corresponding to the first point with coordinates (x3, y3, z3) is: y3-y3'.

[0095] refer to Figure 3 , which is a schematic diagram showing an example of a second point corresponding to a first point when the first fitting line segment is a curve.

[0096] Figure 3 The first fitting line segment of the right lane line for the i-th determination operation and the second fitting line segment of the left lane line for the i-th determination operation are shown. The first fitting line segment of the right lane line for the i-th determination operation and the second fitting line segment of the left lane line for the i-th determination operation are both curves.

[0097] Figure 3 The first point 1 on the first fitting line segment of the right lane line for the i-th determination operation and the second point 1 on the second fitting line segment of the left lane line for the i-th determination operation are shown.

[0098] The second point 1 is an intersection point of a tangent line passing through the first point 1 and the second fitting line segment, and the second point 1 is a second point corresponding to the first point 1.

[0099] The horizontal distance corresponding to the first point 1 is: the vertical coordinate value of the first point 1 minus the vertical coordinate value of the second point 1.

[0100] In a possible implementation, step S1013 includes: step S10131 - step S10132.

[0101] In step S10131, the lateral distance corresponding to each first point on the first fitting line segment is calculated.

[0102] For a first point on the first fitting line segment, in order to calculate the horizontal distance corresponding to the first point, determine the maximum vertical coordinate value of the first point, that is, the coordinate value of the y-axis of the world coordinate system and the vertical coordinate value of the second point corresponding to the first point on the second fitting line segment, and determine the minimum vertical coordinate value of the first point and the vertical coordinate value of the second point corresponding to the first point on the second fitting line segment. The horizontal distance corresponding to the first point is: the maximum vertical coordinate value minus the minimum vertical coordinate value.

[0103] In step S10132, if the absolute value of the difference between the lateral distances corresponding to any two first points on the first fitting line segment is less than a lateral distance difference threshold, the pitch angle targeted by the i-th determination operation may be determined as the target pitch angle, where the lateral distance difference threshold is greater than 0.

[0104] In step S10132, if the absolute value of the difference between the lateral distances corresponding to the two first points on the first fitting line segment is not less than the lateral distance difference threshold, the pitch angle for the next determination operation, i.e., the (i+1)th determination operation, is determined according to the pitch angle for the i-th determination operation.

[0105] In the embodiment of the present disclosure, the first point with the smallest horizontal coordinate value in the world coordinate system among all the first points on the first fitting line segment may be called the target first point on the first fitting line segment. The first point other than the target first point among all the first points on the first fitting line segment may be called the other first points on the first fitting line segment.

[0106] In step S10132, according to the pitch angle targeted by the i-th determination operation, determining the pitch angle targeted by the next determination operation, i.e., the i+1-th determination operation, includes: if the number of other first points whose corresponding lateral distances are less than the lateral distance corresponding to the target first point is greater than the number of other first points whose corresponding lateral distances are greater than the lateral distance corresponding to the target first point, it can be determined that the first fitting line segment is not parallel to the second fitting line segment and the first fitting line segment and the second fitting line segment are in an overall contraction state. If the number of other first points whose corresponding lateral distances are less than the lateral distance corresponding to the target first point is less than the number of other first points whose corresponding lateral distances are greater than the lateral distance corresponding to the target first point, it can be determined that the first fitting line segment is not parallel to the second fitting line segment and the first fitting line segment and the second fitting line segment are in an overall expansion state.

[0107] In step S10132, based on the pitch angle targeted by the i-th determination operation, determining the pitch angle targeted by the next determination operation, i.e., the (i+1)-th determination operation, includes: a binary search of the pitch angle may be used to determine the pitch angle targeted by the (i+1)-th determination operation based on the pitch angle targeted by the i-th determination operation and the first fitting line segment and the second fitting line segment being in an overall contracted state or an expanded state.

[0108] If the first fitting line segment and the second fitting line segment are in an overall contracted state, the pitch angle for the (i+1)th determination operation is: the average value of the minimum value within the preset value range of the pitch angle and the pitch angle for the i-th determination operation, that is, the sum of the minimum value within the preset value range of the pitch angle and the pitch angle for the i-th determination operation divided by 2.

[0109] If the first fitting line segment and the second fitting line segment are in an overall outward expansion state, the pitch angle targeted by the (i+1)th determination operation is: the maximum value within the preset value range of the pitch angle and the average value of the pitch angle targeted by the i-th determination operation, that is, the sum of the maximum value within the preset value range of the pitch angle and the pitch angle targeted by the i-th determination operation divided by 2.

[0110] In step S102, according to the target pitch angle, the two-dimensional points of the target lane line associated with the target vehicle are mapped to the three-dimensional space to obtain the three-dimensional points used to obtain the lane line detection result of the target lane line.

[0111] The target lane line associated with the target vehicle is any lane line on the road where the target vehicle is traveling at the current moment.

[0112] In step S102, according to the target pitch angle, the two-dimensional point of the target lane line associated with the target vehicle is mapped to the three-dimensional space to obtain the three-dimensional point for obtaining the lane line detection result of the target lane line, including: mapping each two-dimensional point of the target lane line associated with the target vehicle to the three-dimensional space respectively, to obtain each three-dimensional point for obtaining the 3D lane line detection result of the target lane line.

[0113] In step S102, a 3D lane line detection result of the target lane line may be obtained according to each three-dimensional point used to obtain the lane line detection result of the target lane line.

[0114] For a two-dimensional point of a target lane line related to the target vehicle, the two-dimensional point is mapped to a three-dimensional space according to the target pitch angle to obtain a three-dimensional point for obtaining a lane line detection result of the target lane line, and the three-dimensional point corresponds to the two-dimensional point.

[0115] In step S102, in order to map each two-dimensional point of the target lane line related to the target vehicle to the three-dimensional space respectively, the target rotation matrix of the camera of the target vehicle is calculated according to the target pitch angle, the current roll angle of the camera of the target vehicle, and the current yaw angle of the camera of the target vehicle.

[0116] The current rotation matrix of the target vehicle's camera in the current camera extrinsic parameters of the target vehicle's camera is replaced by the target rotation matrix of the target vehicle's camera to obtain the target camera extrinsic parameters of the target vehicle's camera.

[0117] The target camera external parameters of the target vehicle's camera include: a target rotation matrix of the target vehicle's camera and a current translation vector of the target vehicle's camera.

[0118] In step S102, when each two-dimensional point of the target lane line associated with the target vehicle is mapped to the three-dimensional space respectively, each two-dimensional point of the target lane line associated with the target vehicle is mapped to the three-dimensional space respectively according to the target camera extrinsic parameters of the target vehicle's camera, the intrinsic parameters of the target vehicle's camera, and the conversion relationship between the coordinates of the same two-dimensional point in the pixel coordinate system and the coordinates of the same two-dimensional point in the image coordinate system, to obtain each three-dimensional point used to obtain the 3D lane line detection result of the target lane line.

[0119] refer to Figure 4 , which shows a flow chart of another data processing method provided by an embodiment of the present disclosure.

[0120] In step S401, at least one determination operation is performed to obtain a target pitch angle of the camera for the target vehicle, the determination operation comprising: calculating a lateral distance difference between a lateral distance corresponding to each first point other than the target first point on the first fitting line segment and a lateral distance corresponding to the target first point on the first fitting line segment to obtain a plurality of lateral distance differences; and determining the target pitch angle or the pitch angle for the next determination operation based on the plurality of lateral distance differences.

[0121] The i-th determination operation includes: step S4011 to step S4013.

[0122] In step S4011, according to the pitch angle targeted by the i-th determination operation, the two-dimensional points of each lane line in the first lane line and the second lane line are respectively mapped to the three-dimensional space to obtain the three-dimensional points of each lane line in the first lane line and the second lane line targeted by the i-th determination operation.

[0123] In step S4012, the three-dimensional points of each lane line in the first lane line and the second lane line for the i-th determination operation are fitted respectively to obtain a first fitting line segment of the first lane line for the i-th determination operation and a second fitting line segment of the second lane line for the i-th determination operation.

[0124] In step S4013, the lateral distance difference between the lateral distance corresponding to each first point other than the target first point on the first fitting line segment and the lateral distance corresponding to the target first point on the first fitting line segment is calculated to obtain a plurality of lateral distance differences; and the target pitch angle or the pitch angle to be determined for the next determination operation is determined based on the plurality of lateral distance differences.

[0125] The target first point on the first fitting line segment may be: the first point whose coordinate in the world coordinate system has the smallest horizontal coordinate value among all the first points on the first fitting line segment.

[0126] The other first points on the first fitting line segment are: the first points of all the first points on the first fitting line segment except the target first point.

[0127] For a first point on the first fitting line segment, in order to calculate the horizontal distance corresponding to the first point, determine the maximum vertical coordinate value of the first point, that is, the coordinate value of the y-axis of the world coordinate system and the vertical coordinate value of the second point corresponding to the first point on the second fitting line segment, and determine the minimum vertical coordinate value of the first point and the vertical coordinate value of the second point corresponding to the first point on the second fitting line segment. The horizontal distance corresponding to the first point is: the maximum vertical coordinate value minus the minimum vertical coordinate value.

[0128] For another first point on the first fitting line segment, the lateral distance difference between the lateral distance corresponding to the other first point and the lateral distance corresponding to the target first point on the first fitting line segment is: the lateral distance corresponding to the other first point minus the lateral distance corresponding to the target first point.

[0129] In a possible implementation, in step S4013, the absolute values ​​of the multiple lateral distance differences may be added to obtain the sum of the absolute values ​​of the multiple lateral distance differences. If the sum is less than the first threshold, it can be determined that the first fitting line segment is substantially parallel to the second fitting line segment, and the pitch angle for the i-th determination operation is determined as the target pitch angle. Wherein, the first threshold is greater than 0. If the sum is not less than the first threshold, the pitch angle for the next determination operation, i.e., the i+1-th determination operation, is determined. If the number of lateral distance differences less than 0 in the multiple lateral distance differences is greater than the number of lateral distance differences greater than 0 in the multiple lateral distance differences, it can be determined that the first fitting line segment is not parallel to the second fitting line segment and the first fitting line segment is in an overall contraction state with the second fitting line segment. If the number of lateral distance differences less than 0 in the multiple lateral distance differences is less than the number of lateral distance differences greater than 0 in the multiple lateral distance differences, it can be determined that the first fitting line segment is not parallel to the second fitting line segment and the first fitting line segment is in an overall expansion state with the second fitting line segment. The pitch angle may be searched in binary search mode, and the pitch angle targeted by the (i+1)th determination operation may be determined according to whether the first fitting line segment and the second fitting line segment are in an overall contraction state or an overall expansion state and the pitch angle targeted by the i-th determination operation.

[0130] In the disclosed embodiment, one of the following items that can reflect whether the pitch angle targeted by the i-th determination operation is relatively accurate can be determined based on multiple lateral distance differences: the first fitting line segment is substantially parallel to the second fitting line segment, the first fitting line segment and the second fitting line segment are in an overall contracted state, the first fitting line segment and the second fitting line segment are in an overall contracted state or an overall expanded state. Based on the item that can reflect whether the pitch angle targeted by the i-th determination operation is relatively accurate, one of the following items can be relatively accurately determined: the pitch angle targeted by the i-th determination operation is relatively accurate and can be used as the target pitch angle, and the pitch angle targeted by the i-th determination operation is not accurate enough and needs to continue to perform the determination operation. Thereby, the accuracy of the determined target pitch angle is improved.

[0131] In one possible implementation, step S4013 includes: step S40131-step S40134.

[0132] In step S40131, the absolute value of each lateral distance difference in the multiple lateral distance differences is added to obtain the sum of the absolute values ​​of the multiple lateral distance differences, and the sum is divided by the number of lateral distance differences in the multiple lateral distance differences to obtain an average value of the absolute values ​​of the lateral distance differences in the multiple lateral distance differences.

[0133] In step S40132, it is determined whether the average value of the absolute values ​​of the lateral distance differences among the plurality of lateral distance differences is less than an average value threshold value, wherein the average value threshold value is greater than zero.

[0134] In step S40133, if the average of the absolute values ​​of the lateral distance differences among the multiple lateral distance differences is less than the average threshold, it can be determined that the first fitting line segment is substantially parallel to the second fitting line segment, and the pitch angle targeted by the i-th determination operation is determined as the target pitch angle.

[0135] In step S40134, if the average value of the absolute values ​​of the lateral distance differences in the multiple lateral distance differences is not less than the average value threshold, it can be determined that the first fitting line segment is not parallel to the second fitting line segment, and based on the multiple lateral distance differences, the candidate pitch angles for the i-th determination operation are determined, and based on the candidate pitch angles for the i-th determination operation, the target pitch angle or the pitch angle for the next determination operation, i.e., the (i+1)-th determination operation, is determined.

[0136] In the disclosed embodiment, it is possible to more accurately determine whether the first fitting line segment is substantially parallel to the second fitting line segment based on the comparison result of the average value of the absolute values ​​of the lateral distance differences among the multiple lateral distance differences and the average value threshold, and it is possible to more accurately determine one of the following items: the pitch angle targeted by the i-th determination operation is more accurate and can be used as the target pitch angle, and the pitch angle targeted by the i-th determination operation is not accurate enough and needs to continue to perform the determination operation to find a more accurate target pitch angle. Thus, the accuracy of the determined target pitch angle is improved.

[0137] In a possible implementation, in order to determine the candidate pitch angle for the i-th determination operation, in step S40134, if the number of the lateral distance differences less than 0 among the multiple lateral distance differences is greater than the number of the lateral distance differences greater than 0 among the multiple lateral distance differences, it can be determined that the first fitting line segment is not parallel to the second fitting line segment and the first fitting line segment and the second fitting line segment are in an overall contraction state, a first pitch angle less than the pitch angle for the i-th determination operation can be determined, and the first pitch angle is used as the candidate pitch angle for the i-th determination operation. If the number of the lateral distance differences less than 0 among the multiple lateral distance differences is less than the number of the lateral distance differences greater than 0 among the multiple lateral distance differences, it can be determined that the first fitting line segment is not parallel to the second fitting line segment and the first fitting line segment and the second fitting line segment are in an overall expansion state, a second pitch angle greater than the pitch angle for the i-th determination operation can be determined, and the second pitch angle is used as the candidate pitch angle for the i-th determination operation. A binary search method for the pitch angle may be used to determine the candidate pitch angle targeted by the i-th determination operation according to whether the first fitting line segment and the second fitting line segment are in an overall contraction state or an overall expansion state and the pitch angle targeted by the i-th determination operation.

[0138] In a possible implementation, in order to determine the candidate pitch angle for the i-th determination operation, in step S40134, the multiple lateral distance differences are added to obtain the sum of the multiple lateral distance differences. If the sum of the multiple lateral distance differences is less than 0, it can be determined that the first fitting line segment is not parallel to the second fitting line segment and the first fitting line segment and the second fitting line segment are in an overall contraction state, a first pitch angle less than the pitch angle for the i-th determination operation can be determined, and the first pitch angle is used as the candidate pitch angle for the i-th determination operation. If the sum of the multiple lateral distance differences is not less than 0, it can be determined that the first fitting line segment is not parallel to the second fitting line segment and the first fitting line segment and the second fitting line segment are in an overall expansion state, a second pitch angle greater than the pitch angle for the i-th determination operation can be determined, and the second pitch angle is used as the candidate pitch angle for the i-th determination operation. A binary search method for the pitch angle may be used to determine the candidate pitch angle targeted by the i-th determination operation according to whether the first fitting line segment and the second fitting line segment are in an overall contraction state or an overall expansion state and the pitch angle targeted by the i-th determination operation.

[0139] The above-mentioned method of using binary search for the pitch angle, according to whether the first fitting line segment and the second fitting line segment are in an overall contraction state or an overall expansion state and the pitch angle targeted by the i-th determination operation, determining the candidate pitch angle targeted by the i-th determination operation includes: if the first fitting line segment and the second fitting line segment are in an overall contraction state, calculating the first pitch angle, the first pitch angle being: the sum of the minimum value within the preset value range of the pitch angle and the average value of the pitch angle targeted by the i-th determination operation, that is, the minimum value within the preset value range of the pitch angle and the pitch angle targeted by the i-th determination operation, divided by 2. If the first fitting line segment and the second fitting line segment are in an overall expansion state, calculating the second pitch angle, the second pitch angle being: the sum of the maximum value within the preset value range of the pitch angle and the average value of the pitch angle targeted by the i-th determination operation, that is, the maximum value within the preset value range of the pitch angle and the pitch angle targeted by the i-th determination operation, divided by 2.

[0140] In one possible implementation, step S40134 includes: step S40134a-step S40134b.

[0141] In step S40134a, when the number of times the determination operation is performed reaches the number threshold, the candidate pitch angle targeted by the i-th determination operation is determined as the target pitch angle.

[0142] In step S40134b, when the number of times the determination operation is performed does not reach the number threshold, the candidate pitch angle for the i-th determination operation is used to back-project the three-dimensional points of each lane line in the first lane line and the second lane line for the i-th determination operation to obtain multiple projected two-dimensional points of each lane line for the i-th determination operation, and based on the multiple projected two-dimensional points of each lane line for the i-th determination operation, determine to use the candidate pitch angle for the i-th determination operation as the target pitch angle or to use the candidate pitch angle for the i-th determination operation as the pitch angle for the next determination operation, i.e., the i+1-th determination operation.

[0143] In step S40134b, the three-dimensional points of each of the first lane line and the second lane line for the i-th determination operation are back-projected to obtain multiple projected two-dimensional points of each lane line for the i-th determination operation, including: back-projecting each three-dimensional point of the first lane line for the i-th determination operation to obtain multiple projected two-dimensional points of the first lane line for the i-th determination operation; back-projecting each three-dimensional point of the second lane line for the i-th determination operation to obtain multiple projected two-dimensional points of the second lane line for the i-th determination operation.

[0144] For one of the first lane line and the second lane line, based on the camera extrinsic parameters of the target vehicle's camera for the i-th determination operation, the intrinsic parameters of the target vehicle's camera, and the conversion relationship between the coordinates of the same two-dimensional point in the pixel coordinate system and the coordinates of the same two-dimensional point in the image coordinate system, back-projection is performed on each three-dimensional point of the lane line for the i-th determination operation to obtain multiple projected two-dimensional points of the lane line for the i-th determination operation.

[0145] It should be noted that the back projection of a three-dimensional point can be understood as: through the coordinate conversion relationship between the pixel coordinate system, the image coordinate system, the camera coordinate system, and the world coordinate system, the coordinates of the three-dimensional point in the world coordinate system are converted into the coordinates of the projected two-dimensional point corresponding to the three-dimensional point, thereby determining the projected two-dimensional point corresponding to the three-dimensional point.

[0146] In step S40134b, multiple projected two-dimensional points of the first lane line for the i-th determination operation are fitted to obtain a third fitted line segment of the first lane line for the i-th determination operation, and multiple projected two-dimensional points of the second lane line for the i-th determination operation are fitted to obtain a fourth fitted line segment of the second lane line for the i-th determination operation.

[0147] In step S40134b, the target pitch angle or the pitch angle for the next determination operation is determined based on the third point on the third fitting line segment of the first lane line for the i-th determination operation and the fourth point on the fourth fitting line segment of the second lane line for the i-th determination operation corresponding to the third point.

[0148] The abscissa of the third point on the third fitting line segment is: a sampled abscissa value obtained by sampling, for example, averaging, the abscissa values ​​of the coordinates of multiple projected two-dimensional points of the first lane line for the i-th determination operation.

[0149] That is, the third point on the third fitting line segment is a point on the third fitting line segment whose abscissa value is a sampled abscissa value obtained by sampling the abscissa values ​​of multiple projected two-dimensional points of the first lane line for the i-th determination operation.

[0150] For a third point on the third fitting line segment, if the third fitting line segment is a straight line, the abscissa value of the fourth point on the fourth fitting line segment corresponding to the third point is: the abscissa value of the coordinate of the third point.

[0151] For a third point on the third fitting line segment, if the third fitting line segment is a curve, the fourth point corresponding to the third point is: an intersection point of a tangent line passing through the third point and the fourth fitting line segment.

[0152] In the embodiment of the present disclosure, the third point with the smallest horizontal coordinate value among all the third points on the third fitting line segment can be called the target third point, and the third points other than the target third point among all the third points on the third fitting line segment can be called other third points.

[0153] In step S40134b, for each other third point on the third fitting line segment, the lateral distance corresponding to the other third point is subtracted from the lateral distance corresponding to the target third point on the third fitting line segment to obtain the lateral distance difference corresponding to the other third point.

[0154] For a third point on the third fitting line segment, in order to calculate the lateral distance corresponding to the third point, determine the maximum lateral value between the ordinate value of the third point and the ordinate value of the fourth point on the fourth fitting line segment corresponding to the third point, and determine the minimum lateral value between the ordinate value of the third point and the ordinate value of the fourth point on the fourth fitting line segment corresponding to the third point. The lateral distance corresponding to the third point is: the maximum lateral value minus the minimum lateral value.

[0155] In step S40134b, the absolute values ​​of the lateral distance differences corresponding to each other third point are added together to obtain the sum of the absolute values ​​of the lateral distance differences corresponding to all other third points. The sum is divided by the number of other third points to obtain the average value of the absolute values ​​of the lateral distance differences corresponding to the other third points. If the average value is less than the second threshold, the candidate pitch angle targeted by the i-th determination operation is used as the target pitch angle. The second threshold is greater than 0. If the average value is not less than the second threshold, the candidate pitch angle targeted by the i-th determination operation is used as the pitch angle targeted by the next determination operation, i.e., the i+1-th determination operation.

[0156] In the disclosed embodiment, the target pitch angle or the pitch angle for the next determination operation can be determined based on whether the number of times the determination operation is performed reaches the number threshold. When the number of times the determination operation is performed reaches the number threshold, the candidate pitch angle for the i-th determination operation is determined as the target pitch angle, which can avoid the problem of too much overhead in determining the target pitch angle due to too many times of performing the determination operation. When the number of times the determination operation is performed does not reach the number threshold, it can be checked based on the projected two-dimensional point whether the candidate pitch angle for the i-th determination operation is relatively accurate, thereby, one of the following items can be determined more accurately: the candidate pitch angle for the i-th determination operation is relatively accurate and can be used as the target pitch angle, and the candidate pitch angle for the i-th determination operation is not accurate enough and needs to continue to perform the determination operation to find a more accurate target pitch angle. Thereby, the accuracy of the determined target pitch angle is improved.

[0157] In step S402, according to the target pitch angle, the two-dimensional point of the target lane line associated with the target vehicle is mapped to the three-dimensional space to obtain the three-dimensional point for obtaining the lane line detection result of the target lane line.

[0158] Step S402 is the same as step S102, and step S402 refers to step S102.

[0159] refer to Figure 5 , which shows a flowchart of an example of obtaining a target pitch angle or a pitch angle for the next determination operation.

[0160] In this example, the first lane line associated with the target vehicle is the right lane line, and the second lane line associated with the target vehicle is the left lane line. The right lane line is: a lane line on the right side of the target vehicle that belongs to the lane in which the target vehicle is traveling at the current moment. The left lane line is: a lane line on the left side of the target vehicle that belongs to the lane in which the target vehicle is traveling at the current moment.

[0161] In this example, according to the pitch angle targeted by the i-th determination operation, the two-dimensional points of each lane line in the right lane line and the left lane line are respectively mapped to the three-dimensional space to obtain the three-dimensional points of each lane line in the right lane line and the left lane line targeted by the i-th determination operation.

[0162] In this example, the three-dimensional points of the right lane line for the i-th determination operation are fitted to obtain the first fitted line segment of the right lane line for the i-th determination operation, and the three-dimensional points of the left lane line for the i-th determination operation are fitted to obtain the second fitted line segment of the left lane line for the i-th determination operation.

[0163] In this example, the horizontal coordinate value of the three-dimensional point of the right lane line is sampled to determine the first point on the right lane line curve, and based on the first point on the right lane line curve, the second point on the left lane line curve corresponding to the first point is determined.

[0164] In this example, calculating the average value of the absolute values ​​of the lateral distance differences among multiple lateral distance differences includes: calculating the lateral distance difference between the lateral distance corresponding to each first point other than the target first point on the first fitting line segment and the lateral distance corresponding to the target first point on the first fitting line segment to obtain multiple lateral distance differences; calculating the average value of the absolute values ​​of the lateral distance differences among the multiple lateral distance differences.

[0165] In this example, if the average value of the absolute values ​​of the lateral distance differences among the multiple lateral distance differences is less than the average value threshold, the pitch angle targeted by the i-th determination operation is determined as the target pitch angle.

[0166] In this example, if the average value of the absolute values ​​of the lateral distance differences among the multiple lateral distance differences is not less than the average value threshold, the sum of the multiple lateral distance differences is calculated.

[0167] In this example, if the sum of the multiple lateral distance differences is less than 0, then based on the pitch angle targeted by the i-th determination operation, the pitch angle is reduced to obtain the candidate pitch angle targeted by the i-th determination operation. If the sum of the multiple lateral distance differences is not less than 0, then based on the pitch angle targeted by the i-th determination operation, the pitch angle is increased to obtain the candidate pitch angle targeted by the i-th determination operation.

[0168] In this example, after the candidate pitch angle for which the i-th determination operation is performed is obtained, it is determined whether the number of times the determination operation is performed reaches a number threshold.

[0169] In this example, if the number of times the determination operation is performed reaches the number threshold, the candidate pitch angle targeted by the i-th determination operation is determined as the target pitch angle.

[0170] In this example, if the number of times the determination operation is performed does not reach the number threshold, each three-dimensional point of the first lane line for the i-th determination operation is back-projected to obtain multiple projected two-dimensional points of the first lane line for the i-th determination operation; each three-dimensional point of the second lane line for the i-th determination operation is back-projected to obtain multiple projected two-dimensional points of the second lane line for the i-th determination operation.

[0171] In this example, the sum of the absolute values ​​of the lateral distance differences is calculated based on the projected 2D points, including: fitting multiple projected two-dimensional points of the first lane line for the i-th determination operation to obtain the third fitted line segment of the first lane line for the i-th determination operation, and fitting multiple projected two-dimensional points of the second lane line for the i-th determination operation to obtain the fourth fitted line segment of the second lane line for the i-th determination operation.

[0172] In this example, the average of the absolute values ​​of the lateral distance differences corresponding to other third points is calculated based on the projected 2D point.

[0173] In this example, if the average value is less than the second threshold, the candidate pitch angle targeted by the i-th determination operation is used as the target pitch angle, wherein the second threshold is greater than 0. If the average value is not less than the second threshold, the candidate pitch angle targeted by the i-th determination operation is used as the pitch angle targeted by the next determination operation, i.e., the i+1-th determination operation.

[0174] The disclosed embodiments provide a data processing device. The data processing device is used to implement the above-mentioned embodiments and preferred implementation modes, and the descriptions that have been made will not be repeated. As used below, the term "unit" can implement a combination of software and / or hardware for a predetermined function. Although the systems described in the following embodiments are preferably implemented in software, the implementation of hardware, or a combination of software and hardware, is also possible and conceivable.

[0175] The data processing device comprises:

[0176] a target pitch angle determination unit, configured to perform at least one determination operation to obtain a target pitch angle of a camera for a target vehicle;

[0177] A three-dimensional point acquisition unit is used to map the two-dimensional points of the lane line related to the target vehicle to the three-dimensional space according to the target pitch angle, so as to obtain the three-dimensional points used to obtain the lane line detection result, wherein the determination operation includes: mapping the two-dimensional points of each lane line of the first lane line and the second lane line related to the target vehicle to the three-dimensional space according to the pitch angle targeted by the determination operation, so as to obtain the three-dimensional points of each lane line targeted by the determination operation; fitting the three-dimensional points of each lane line targeted by the determination operation, so as to obtain the first fitting line segment of the first lane line targeted by the determination operation and the second fitting line segment of the second lane line targeted by the determination operation; determining the target pitch angle or the pitch angle targeted by the next determination operation according to the first point on the first fitting line segment and the second point on the second fitting line segment corresponding to the first point, wherein the abscissa value of the first point is the sampled abscissa value obtained by sampling the abscissa value of the three-dimensional point of the first lane line targeted by the determination operation.

[0178] In a possible implementation, the target pitch angle determination unit is further used to calculate the lateral distance difference between the lateral distance corresponding to each first point other than the target first point on the first fitting line segment and the lateral distance corresponding to the target first point on the first fitting line segment, to obtain multiple lateral distance differences, wherein the lateral distance corresponding to the first point on the first fitting line segment is determined according to the ordinate value of the first point on the first fitting line segment and the ordinate value of the second point corresponding to the first point on the second fitting line segment; and determine the target pitch angle or the pitch angle for the next determination operation according to the multiple lateral distance differences.

[0179] In one possible implementation, the target pitch angle determination unit is further used to determine whether the average of the absolute values ​​of the lateral distance differences among the multiple lateral distance differences is less than an average value threshold; if so, determine the pitch angle targeted by the determination operation as the target pitch angle; if not, determine the candidate pitch angles targeted by the determination operation based on the multiple lateral distance differences, and determine the target pitch angle or the pitch angle targeted by the next determination operation based on the candidate pitch angles.

[0180] In one possible implementation, the target pitch angle determination unit is further used to determine whether the sum of the multiple lateral distance differences is less than 0; if so, determine a first pitch angle that is smaller than the pitch angle targeted by the determination operation, and use the first pitch angle as the candidate pitch angle; if not, determine a second pitch angle that is greater than the pitch angle targeted by the determination operation, and use the second pitch angle as the candidate pitch angle.

[0181] In one possible implementation, the target pitch angle determination unit is further used to determine the candidate pitch angle as the target pitch angle when the number of times the determination operation is performed reaches a number threshold; when the number of times the determination operation is performed does not reach the number threshold, use the candidate pitch angle to back-project the three-dimensional points of each lane line for the determination operation to obtain multiple projected two-dimensional points of each lane line for the determination operation, and determine to use the candidate pitch angle as the target pitch angle or to use the candidate pitch angle as the pitch angle for the next determination operation based on the multiple projected two-dimensional points of each lane line for the determination operation.

[0182] In a possible implementation, the first lane line and the second lane line associated with the target vehicle are two adjacent lane lines belonging to the lane in which the target vehicle is traveling at the current moment.

[0183] 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 a memory that executes one or more software or fixed programs, and / or other devices that can provide the above functions.

[0184] 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.

[0185] refer to Figure 6 , which shows a schematic diagram of the structure of a computer device provided by an embodiment of the present disclosure, 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).

[0186] 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 a dedicated 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.

[0187] The memory 20 stores instructions executable by at least one processor 10, so that the at least one processor 10 executes the method shown in the above embodiment.

[0188] The memory 20 may include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function; the data storage area may store data created according to 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 arranged 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.

[0189] 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.

[0190] 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.

[0191] The input device 30 can receive input digital or character information, and generate key signal input related to the user settings and function control of the computer device, such as a touch screen, a keypad, a mouse, a track pad, a touch pad, an indicator bar, one or more mouse buttons, a trackball, a joystick, etc. The output device 40 may include a display device, an auxiliary lighting device (e.g., an LED) and a tactile feedback device (e.g., a vibration motor), etc. 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.

[0192] The embodiments of the present disclosure also provide a computer-readable storage medium. The above-mentioned method according to the embodiments of the present disclosure can be implemented in hardware, firmware, or can be implemented as a computer code that can be recorded in a storage medium, or can be implemented as a computer code that is originally stored in a remote storage medium or a non-temporary machine-readable storage medium and will be stored in a local storage medium and downloaded through a network, 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.

[0193] A portion of the embodiments of the present disclosure may be applied as a computer program product, such as a computer program instruction, which, when executed by a computer, can execute 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 existence of computer program instructions in computer-readable media includes, but is not limited to, source files, executable files, installation package files, etc., and accordingly, the way in which computer program instructions are executed by a 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 may be any available computer-readable storage medium or communication medium accessible to the computer.

[0194] Although the embodiments of the present disclosure have been described in conjunction with the accompanying drawings, those skilled in the art may make various modifications and variations without departing from the spirit and scope of the present disclosure, and such modifications and variations are all within the scope defined by the appended claims.

Claims

1. A data processing method, characterized in that: The method comprises: Perform at least one determination operation to obtain a target pitch angle of a camera for a target vehicle; According to the target pitch angle, a two-dimensional point of a lane line associated with the target vehicle is mapped to a three-dimensional space to obtain a three-dimensional point for obtaining a lane line detection result, wherein the determination operation includes: According to the pitch angle targeted by the determination operation, respectively mapping the two-dimensional points of each lane line in the first lane line and the second lane line associated with the target vehicle to the three-dimensional space to obtain the three-dimensional points of each lane line targeted by the determination operation; Fitting the three-dimensional points of each lane line for the determination operation respectively to obtain a first fitting line segment of the first lane line for the determination operation and a second fitting line segment of the second lane line for the determination operation; A target pitch angle or a pitch angle for a next determination operation is determined according to a first point on the first fitting line segment and a second point on the second fitting line segment corresponding to the first point, wherein the abscissa value of the first point is a sampled abscissa value obtained by sampling the abscissa value of the three-dimensional point of the first lane line for the determination operation.

2. The method according to claim 1, characterized in that: Determining a target pitch angle or a pitch angle for a next determination operation according to a first point on the first fitting line segment and a second point on the second fitting line segment corresponding to the first point includes: Calculating the lateral distance difference between the lateral distance corresponding to each first point other than the target first point on the first fitting line segment and the lateral distance corresponding to the target first point on the first fitting line segment, to obtain a plurality of lateral distance differences, wherein the lateral distance corresponding to the first point on the first fitting line segment is determined according to the ordinate value of the first point on the first fitting line segment and the ordinate value of the second point on the second fitting line segment corresponding to the first point; According to the plurality of lateral distance differences, a target pitch angle or a pitch angle for a next determination operation is determined.

3. The method according to claim 2, characterized in that Determining the target pitch angle or the pitch angle for the next determination operation according to the multiple lateral distance differences comprises: determining whether an average value of absolute values ​​of the lateral distance differences among the plurality of lateral distance differences is less than an average value threshold; If yes, determining the pitch angle targeted by the determination operation as a target pitch angle; If not, determining candidate pitch angles for the determination operation according to the multiple lateral distance differences, and determining a target pitch angle or a pitch angle for the next determination operation according to the candidate pitch angles.

4. The method according to claim 3, characterized in that Determining, according to the plurality of lateral distance differences, a candidate pitch angle for which the determining operation is performed comprises: Determining whether the sum of the plurality of lateral distance differences is less than 0; If so, determining a first pitch angle that is smaller than the pitch angle for which the determination operation is performed, and using the first pitch angle as the candidate pitch angle; If not, a second pitch angle greater than the pitch angle for which the determination operation is performed is determined, and the second pitch angle is used as the candidate pitch angle.

5. The method according to claim 3, characterized in that: Determining a target pitch angle or a pitch angle for a next determination operation according to the candidate pitch angles includes: When the number of times the determination operation is performed reaches a number threshold, determining the candidate pitch angle as a target pitch angle; When the number of times the determination operation is performed does not reach a number threshold, the candidate pitch angles are used to back-project the three-dimensional points of each lane line for the determination operation to obtain a plurality of projected two-dimensional points of each lane line for the determination operation, and based on the plurality of projected two-dimensional points of each lane line for the determination operation, it is determined to use the candidate pitch angle as the target pitch angle or to use the candidate pitch angle as the pitch angle for the next determination operation.

6. The method according to any one of claims 1 to 5, characterized in that The first lane line and the second lane line associated with the target vehicle are two adjacent lane lines belonging to the lane in which the target vehicle is traveling at the current moment.

7. A data processing device, characterized in that: include: a target pitch angle determination unit, configured to perform at least one determination operation to obtain a target pitch angle of a camera for a target vehicle; A three-dimensional point acquisition unit is used to map the two-dimensional points of the lane line related to the target vehicle to the three-dimensional space according to the target pitch angle, so as to obtain the three-dimensional points used to obtain the lane line detection result, wherein the determination operation includes: mapping the two-dimensional points of each lane line of the first lane line and the second lane line related to the target vehicle to the three-dimensional space according to the pitch angle targeted by the determination operation, so as to obtain the three-dimensional points of each lane line targeted by the determination operation; fitting the three-dimensional points of each lane line targeted by the determination operation, so as to obtain the first fitting line segment of the first lane line targeted by the determination operation and the second fitting line segment of the second lane line targeted by the determination operation; determining the target pitch angle or the pitch angle targeted by the next determination operation according to the first point on the first fitting line segment and the second point on the second fitting line segment corresponding to the first point, wherein the abscissa value of the first point is the sampled abscissa value obtained by sampling the abscissa value of the three-dimensional point of the first lane line targeted by the determination operation.

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