Construction scene recognition method and device and storage medium

Abstract

The invention relates to the technical field of intelligent transportation, in particular to a construction scene recognition method and device and a storage medium. Comprising the steps of obtaining vehicle parameters, lane parameters and obstacle parameters of at least one obstacle; when the vehicle parameters and the lane parameters meet the preset conditions, determining a primary obstacle in the at least one obstacle according to the obstacle parameters; wherein the preset condition comprises that the vehicle speed is greater than or equal to a first speed threshold value; the lane width is greater than or equal to the first width threshold and less than the second width threshold; the lane curvature radius is greater than or equal to a preset curvature threshold value; and when it is determined that the number of secondary obstacles in the primary obstacles is greater than a number threshold value according to the obstacle parameters of the primary obstacles and the primary obstacles are arranged in a preset line type in the lane, determining that the current scene is a construction scene. The embodiment of the invention is used for solving the problem of low accuracy of construction scene recognition.

Description

Technical Field

[0001] The present application relates to the field of intelligent transportation technology, and in particular to a construction scene recognition method, device and storage medium. Background Art

[0002] The emergence of autonomous driving technology has brought great convenience to people's travel. For example, it has improved the user's travel experience at the user level and reduced the management cost of traffic management at the management level. The realization of autonomous driving by vehicles usually requires the combination of scene recognition technology. The scenes that autonomous driving vehicles need to recognize usually include road construction scenes, traffic signal scenes, pedestrians and non-motorized vehicles, and other scenes.

[0003] At present, the method for identifying construction scenes is usually to collect road pictures in real time, input the road pictures into the scene classification model, and use the scene classification model to determine whether it is a construction scene. This identification method often requires the collection of a large amount of data in the early stage to train the generation model, and the accuracy of the recognition is usually limited by the geographical area. For example, the model generated by the data collected in place A has low accuracy when identifying the scene in place B. At the same time, it is also difficult to obtain data for all construction scenes in practice. Therefore, how to use other methods to more effectively identify different construction scenes has become an urgent problem to be solved. Summary of the invention

[0004] In order to solve the above technical problems, the present application provides a construction scene recognition method, device and storage medium, which can improve the accuracy of construction scene recognition.

[0005] In a first aspect, the present application provides a construction scene recognition method, comprising: acquiring vehicle parameters, lane parameters and obstacle parameters of at least one obstacle; the vehicle parameters include vehicle speed, and the lane parameters include lane width and lane curvature radius; when both the vehicle parameters and the lane parameters meet preset conditions, determining a primary obstacle in at least one obstacle according to the obstacle parameters; wherein the preset conditions include: the vehicle speed is greater than or equal to a first speed threshold; the lane width is greater than or equal to a first width threshold and less than a second width threshold; the lane curvature radius is greater than or equal to a preset curvature threshold; when it is determined according to the obstacle parameters of the primary obstacle that the number of secondary obstacles in the primary obstacle is greater than the number threshold, and the primary obstacles are arranged in a preset line type in the lane, determining that the current scene is a construction scene.

[0006] In a second aspect, the present application provides a construction scene recognition device, including: an acquisition module, used to acquire vehicle parameters, lane parameters and obstacle parameters of at least one obstacle; the vehicle parameters include vehicle speed, and the lane parameters include lane width and lane curvature radius; a determination module, used to determine a primary obstacle in at least one obstacle according to the obstacle parameters when both the vehicle parameters and the lane parameters meet preset conditions; wherein the preset conditions include: the vehicle speed is greater than or equal to a first speed threshold; the lane width is greater than or equal to a first width threshold and less than a second width threshold; the lane curvature radius is greater than or equal to a preset curvature threshold; a processing module, used to determine that the current scene is a construction scene when it is determined based on the obstacle parameters of the primary obstacle that the number of secondary obstacles in the primary obstacle is greater than the number threshold and the primary obstacles are arranged in a preset line type in the lane.

[0007] In a third aspect, the present application provides an electronic device comprising: a processor, a memory, and a computer program stored in the memory and executable on the processor, wherein when the computer program is executed by the processor, the construction scene recognition method of the first aspect is implemented.

[0008] In a fourth aspect, the present application provides a computer-readable storage medium, comprising: a computer program is stored on the computer-readable storage medium, and when the computer program is executed by a processor, the construction scene recognition method as in the first aspect is implemented.

[0009] In a fifth aspect, the present application provides a computer program product, comprising: when the computer program product runs on a computer, the computer implements the construction scene recognition method as in the first aspect.

[0010] The technical solution provided by the present application has the following advantages compared with the prior art: vehicle parameters, lane parameters and obstacle parameters of at least one obstacle are obtained, and when both the vehicle parameters and the lane parameters meet the preset conditions, the primary obstacle is determined in at least one obstacle according to the obstacle parameters. Wherein, the preset conditions include: the vehicle speed is greater than or equal to the first speed threshold; the lane width is greater than or equal to the first width threshold and less than the second width threshold; the lane curvature radius is greater than or equal to the preset curvature threshold. Afterwards, when it is determined that the number of secondary obstacles in the primary obstacle is greater than the number threshold and the primary obstacles are arranged in a preset line type in the lane according to the obstacle parameters of the primary obstacle, the current scene is determined to be a construction scene. In this way, during the process of automatic driving of the vehicle, it is possible to directly determine whether the current scene of the road is a construction scene according to the vehicle parameters, lane parameters and obstacle parameters, avoiding the collection of a large amount of data to train the model, and is applicable to construction scenes of all roads, thereby improving the applicability and accuracy of construction scene recognition. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0012] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0013] Figure 1 One of the flowcharts of the construction scene recognition method provided in the embodiment of the present application;

[0014] Figure 2 The second flowchart of the construction scene recognition method provided in the embodiment of the present application;

[0015] Figure 3 The third flowchart of the construction scene recognition method provided in the embodiment of the present application;

[0016] Figure 4 A fourth flowchart of the construction scene recognition method provided in an embodiment of the present application;

[0017] Figure 5 A schematic diagram of a construction scenario provided in an embodiment of the present application;

[0018] Figure 6 A schematic diagram of another construction scenario provided in an embodiment of the present application;

[0019] Figure 7 A schematic diagram of the structure of a construction scene recognition device provided in an embodiment of the present application;

[0020] Figure 8 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0021] In order to more clearly understand the above-mentioned purposes, features and advantages of the present application, the scheme of the present application will be further described below. It should be noted that the embodiments of the present application and the features in the embodiments can be combined with each other without conflict.

[0022] In the following description, many specific details are set forth to facilitate a full understanding of the present application, but the present application may also be implemented in other ways different from those described herein; obviously, the embodiments in the specification are only part of the embodiments of the present application, rather than all of the embodiments.

[0023] When the vehicle is in autonomous driving mode, users hope to stay away from danger and avoid collisions when encountering complex road conditions, such as construction scenes, and to be warned when necessary and request manual takeover in advance. Currently, the recognition results of road construction scene recognition usually have the problem of low accuracy.

[0024] In view of the above problems, the embodiment of the present application provides a construction scene recognition method, which first obtains vehicle parameters, lane parameters and obstacle parameters of at least one obstacle, and when both the vehicle parameters and the lane parameters meet the preset conditions, determines the primary obstacle in at least one obstacle according to the obstacle parameters. Wherein, the preset conditions include: the vehicle speed is greater than or equal to the first speed threshold; the lane width is greater than or equal to the first width threshold and less than the second width threshold; the lane curvature radius is greater than or equal to the preset curvature threshold. Afterwards, when it is determined that the number of secondary obstacles in the primary obstacle is greater than the number threshold and the primary obstacles are arranged in a preset line type in the lane according to the obstacle parameters of the primary obstacle, it is determined that the current scene is a construction scene. It can directly determine whether the current scene of the road is a construction scene according to the vehicle parameters, lane parameters and obstacle parameters during the automatic driving of the vehicle, which is applicable to the construction scenes of all roads, and improves the applicability and accuracy of construction scene recognition.

[0025] The construction scene recognition method provided in the embodiment of the present application can be executed by a construction scene recognition device, which can be hardware or software. When the construction scene recognition device is hardware, it can be various electronic devices with a construction scene recognition function, including but not limited to vehicle-mounted equipment and intelligent vehicles. When the construction scene recognition device is software, it can be installed in the electronic devices listed above. It can be implemented as multiple software or software modules, or it can be implemented as a single software or software module. No specific limitation is made here.

[0026] Figure 1 A schematic diagram of a construction scene recognition method provided in an embodiment of the present application is shown in FIG. Figure 1 As shown, the construction scene recognition method may include the following steps.

[0027] S11. Acquire vehicle parameters, lane parameters, and obstacle parameters of at least one obstacle.

[0028] Among them, vehicle parameters include vehicle speed, lane parameters include lane width, lane curvature radius and lane line position, and obstacle parameters include yaw angle, size, number, identification, type, position and obstacle speed. The lane curvature radius is the average curvature radius within the preset lane distance in the vehicle's driving direction. The yaw angle includes the clockwise deflection angle and the counterclockwise deflection angle with the vehicle's driving direction as zero degrees.

[0029] Specifically, the method of obtaining vehicle parameters may be to obtain the vehicle speed by calling the data in the vehicle center console, or to collect the vehicle speed in real time through a speed sensor, which is not limited in this application.

[0030] Lane parameters can be obtained by calling data from the vehicle's center console, or by collecting road images in real time and then applying computer vision algorithms to detect and extract lane parameters, or by calling high-precision maps and obtaining lane parameters through map information. This application does not limit this.

[0031] Obstacle parameters may be obtained by collecting environmental data around the vehicle in real time, and then using a recognition algorithm to identify the environmental data to obtain obstacle parameters. For example, when the environmental data is image data, an image recognition algorithm may be used to identify obstacles in the image to obtain obstacle parameters.

[0032] In some embodiments, after collecting image data around the vehicle in real time, obstacle parameters can be obtained through the bev perception algorithm and the occupancy recognition algorithm. Among them, the bev perception algorithm is used to obtain obstacle parameters of target type obstacles, and the occupancy recognition algorithm is used to identify obstacle parameters of other types of obstacles. Target type obstacles are obstacles related to the construction scene, such as water barriers, traffic cones, crash barrels, etc. Other types of obstacles are obstacles other than the first type of obstacles.

[0033] In the above solution, obstacles related to the construction scene are identified through the bev perception algorithm, and other types of obstacles are identified through the occupancy recognition algorithm to obtain obstacle parameters. Obstacles can be identified in the same picture with different algorithms, which improves the accuracy of obstacle recognition.

[0034] S12: When both the vehicle parameters and the lane parameters satisfy preset conditions, determine a primary obstacle from at least one obstacle according to the obstacle parameters.

[0035] First, the preset conditions include: the vehicle speed is greater than or equal to the first speed threshold; the lane width is greater than or equal to the first width threshold and less than the second width threshold; the lane curvature radius is greater than or equal to the preset curvature threshold. The first speed threshold, the first width threshold, the second width threshold, and the preset curvature threshold are all preset, for example, they can be default values, or they can be values ​​set by relevant personnel according to actual conditions.

[0036] For another example, the first speed threshold is 40 kph, the first width threshold is 2.5 m, the second width threshold is 5 m, and the preset curvature threshold is 250 m. Thus, the preset conditions include: vehicle speed ≥ 40 kph, 2.5 m ≤ lane width < 5 m, and lane curvature radius ≥ 250 m.

[0037] Secondly, a primary obstacle is determined from at least one obstacle according to the obstacle parameters.

[0038] Specifically, the manner of determining the primary obstacle in at least one obstacle according to the obstacle parameters may be to perform an evaluation operation on the obstacle parameters of each obstacle to determine the primary obstacle in at least one obstacle.

[0039] The evaluation operation includes: when the yaw angle of the first obstacle is greater than or equal to the first angle and less than the second angle, the obstacle speed of the first obstacle is less than the second speed threshold, and the size of the first obstacle is greater than or equal to the first size, determining that the first obstacle is a primary obstacle.

[0040] The first obstacle is any obstacle among the at least one obstacle. The first angle, the second angle, the second speed threshold, and the first size are all preset, for example, they can be default values, or they can be values ​​set by relevant personnel according to actual conditions. For another example, the first angle is 30°, the second angle is 150°, the second speed threshold is 1kph, and the first size is 10cm. In this way, the evaluation operation includes: when the yaw angle of the first obstacle is between 30° and 150° (that is, 30°≤the yaw angle of the first obstacle<150°), the obstacle speed of the first obstacle is<1kph, and the size of the first obstacle is ≥10cm, the first obstacle is determined to be a primary obstacle.

[0041] In the above scheme, when the yaw angle of the first obstacle is greater than or equal to the first angle and less than the second angle, the obstacle speed of the first obstacle is less than the second speed threshold, and the size of the first obstacle is greater than or equal to the first size, the first obstacle is determined to be a primary obstacle. The yaw angle can be used to filter out the influence of roadside railings on the recognition results; the obstacle speed can be used to filter out the influence of moving obstacles on the recognition results; the obstacle size can be used to filter out small obstacles that will not affect the vehicle driving, that is, the influence of obstacle noise on the obstacle recognition results is filtered out, and the accuracy of obstacle recognition is improved.

[0042] S13: When it is determined, based on the obstacle parameters of the primary obstacles, that the number of secondary obstacles in the primary obstacles is greater than a number threshold, and the primary obstacles are arranged in a preset line shape in the lane, determining that the current scene is a construction scene.

[0043] In some embodiments, the preset line type may be a straight line, a curve, or a line type pre-stored in a line type library.

[0044] In some embodiments, the preset line type may include requirements for the shape of the primary obstacles, and may also include requirements for the number of primary obstacles, such as a straight line consisting of at least 2 primary obstacles, a straight line consisting of at least 3 primary obstacles, or a curve consisting of at least 3 primary obstacles.

[0045] First, if Figure 2 As shown, according to the obstacle parameters of the primary obstacles, the method of determining that the number of secondary obstacles in the primary obstacles is greater than the number threshold may include the following steps.

[0046] S131: Determine an obstacle whose size is smaller than the second size among the primary obstacles as a second obstacle.

[0047] The second size is preset, for example, it can be a default value, or it can be a value set by relevant personnel according to actual conditions. For another example, the second size is 35 cm, so that obstacles with a size less than 35 cm in the primary obstacles are determined as second obstacles.

[0048] S132: Determine an obstacle among the primary obstacles whose size is greater than or equal to the second size as a third obstacle.

[0049] For example, the second size is 35 cm, so the obstacles with a size greater than or equal to 35 cm among the primary obstacles are determined as the third obstacles.

[0050] S133: When the number of the second obstacles is greater than the first number threshold and the number of the third obstacles is greater than the second number threshold, determine that the number of the secondary obstacles in the primary obstacles is greater than the number threshold.

[0051] The first quantity threshold and the second quantity threshold are both preset, for example, they can be default values, or they can be values ​​set by relevant personnel according to actual conditions. For another example, the first quantity threshold is 3 and the second quantity threshold is 1. In this way, when the number of the second obstacles is greater than 3 and the number of the third obstacles is greater than 1, it is determined that the number of secondary obstacles in the primary obstacles is greater than the quantity threshold.

[0052] In the above scheme, the obstacles in the primary obstacles whose size is smaller than the second size are determined as the second obstacles, and the obstacles in the primary obstacles whose size is greater than or equal to the second size are determined as the third obstacles. Afterwards, when the number of second obstacles is greater than the first number threshold, and the number of third obstacles is greater than the second number threshold, it is determined that the number of secondary obstacles in the primary obstacles is greater than the number threshold. It is possible to determine that the obstacles include multiple smaller obstacles and at least one larger obstacle, thereby avoiding the problem of inaccurate identification caused by identifying only one obstacle as a construction scene, and improving the accuracy of construction scene identification.

[0053] Afterwards, if Figure 3 As shown, determining the manner in which the primary obstacles are arranged into a preset line in the lane according to the obstacle parameters of the primary obstacles may include the following steps.

[0054] S134: When the distance between the position of the primary obstacle and the position of the lane line is less than a first distance threshold, determine that the primary obstacle is within the lane line.

[0055] The first distance threshold is preset, for example, it can be a default value, or it can be a value set by relevant personnel according to actual conditions. For another example, the first distance threshold is 0.8m, so that when the distance between the position of the primary obstacle and the position of the lane line is less than 0.8m, it is determined that the primary obstacle is within the lane line.

[0056] In some embodiments, the manner of determining whether the primary obstacle is within the lane according to the obstacle parameters of the primary obstacle may also be: when the position of the primary obstacle is at the zero point and the distance between the position of the primary obstacle and the position of the lane line is less than the second distance threshold, the primary obstacle is determined to be within the lane line. When the position of the primary obstacle is at the target location and the distance between the position of the primary obstacle and the position of the lane line is less than the first distance threshold, the primary obstacle is determined to be within the lane line.

[0057] The second distance threshold is less than the first distance threshold, the zero point is the current vehicle position, the target location is the position where the front of the vehicle in the lane where the current vehicle is located faces the target distance, and the target distance is the vehicle speed multiplied by the preset time. The second distance threshold and the preset time are both preset values. For example, the first distance threshold is 0.8m, the second distance threshold is 0.4m, and the preset time is 3s.

[0058] In some embodiments, the manner of determining whether the primary obstacle is within the lane according to the obstacle parameters of the primary obstacle may also be: when the position of the primary obstacle is at the zero point and the distance between the position of the primary obstacle and the position of the lane line is less than the second distance threshold, the primary obstacle is determined to be within the lane line. When the position of the primary obstacle is at the target location and the distance between the position of the primary obstacle and the position of the lane line is less than the first distance threshold, the primary obstacle is determined to be within the lane line. When the position of the primary obstacle is between the zero point and the target location and the distance between the position of the primary obstacle and the position of the lane line is less than the target distance threshold, the primary obstacle is determined to be within the lane line.

[0059] The value of the target distance threshold is related to the position of the primary obstacle. For example, when the position of the primary obstacle is between the zero point and the target location, the target distance threshold can be: the second distance threshold + (the first distance threshold - the second distance threshold) ÷ 2. For example, when the first distance threshold is 0.8m and the second distance threshold is 0.4m, the target distance threshold is 0.4 + (0.8-0.4) ÷ 2 = 0.6m.

[0060] In the above scheme, when the distance between the position of the primary obstacle and the position of the lane line is less than the first distance threshold, it is determined that the primary obstacle is within the lane line. An offset can be set between the obstacle and the lane, avoiding the problem of inaccurate long-distance recognition and improving the accuracy of construction scene recognition.

[0061] S135: Arrange the primary obstacles into a preset line type according to the identification or type of the primary obstacles.

[0062] Specifically, Figure 4 As shown, determining the manner of arranging the primary obstacles into a preset line type according to the identification or type of the primary obstacles may include the following steps.

[0063] S1351, acquiring at least two frames of images corresponding to primary obstacles that are continuous in time, and executing S1352 or S1353 to determine that the primary obstacles are arranged in a preset line type.

[0064] Specifically, the timing for acquiring images corresponding to at least two temporally continuous frames of primary obstacles may be when acquiring obstacle parameters of at least one obstacle in step S11; or may be acquired in real time when it is necessary to determine that the primary obstacles are arranged in a preset line type, which is not limited in the present application.

[0065] The method for obtaining at least two frames of images corresponding to the primary obstacle that are continuous in time may be to capture a video corresponding to the primary obstacle and split the video into at least two frames of images that are continuous in time; or it may be to directly capture at least two frames of images corresponding to the primary obstacle that are continuous in time, which is not limited in this application.

[0066] S1352: When obstacles of the target type are included in all the consecutive n frames of images and the number of obstacles of the target type in each frame of image is greater than a first type threshold, determine that the primary obstacles are arranged in a preset line type.

[0067] Among them, primary obstacles include target type obstacles. Target type obstacles are obstacles related to the construction scene, such as water barriers, traffic cones, crash barrels, etc. The values ​​of the first type threshold and n are both preset values. For example, the first type threshold is 4, and n is 3. Figure 5 The image shown includes a lane line 21 and a target type obstacle 22. When three consecutive frames of images all include a target type obstacle 22, and the number of target type obstacles 22 in each frame of image is 5, it is determined that the primary obstacles are arranged in a preset line shape.

[0068] S1353: When m consecutive image frames all include obstacles of the target type, and the number of obstacles of the target type in each image frame is greater than the second type threshold, and the identifiers of every two obstacles of the target type in each image frame are different, determine that the primary obstacles are arranged in a preset line type.

[0069] The second type threshold and m are both preset values. For example, the second type threshold is 9, and m is 10. Figure 6 The image shown includes a lane line 21 and a target type obstacle 22. When the target type obstacle 22 is included in 10 consecutive frames of images, and the number of the target type obstacles 22 in each frame of image is 10, and when the identifications of every two target type obstacles 22 in each frame of image are different, it is determined that the primary obstacles are arranged in a preset line type.

[0070] In the above scheme, at least two frames of images corresponding to primary obstacles that are continuous in time are obtained. When the target type of obstacles are included in the continuous n frames of images, and the number of target type of obstacles in each frame of images is greater than the first type threshold, it is determined that the primary obstacles are arranged in a preset line type. Alternatively, when the target type of obstacles are included in the continuous m frames of images, and the number of target type of obstacles in each frame of images is greater than the second type threshold, and the identification of every two target type of obstacles in each frame of images is different, it is determined that the primary obstacles are arranged in a preset line type. Since relevant equipment (such as water barriers, traffic cones, crash buckets, etc.) are usually used in road construction scenes to isolate the construction area from the driving area to ensure driving safety, and the setting of the relevant equipment is to arrange the obstacles in a preset line type on the lane, the above scheme can determine whether the obstacles are arranged in a preset line type according to the type or identification of the obstacles, further improving the accuracy of construction scene recognition.

[0071] The present application can obtain vehicle parameters, lane parameters and obstacle parameters of at least one obstacle, and when both the vehicle parameters and the lane parameters meet the preset conditions, determine the primary obstacle in at least one obstacle according to the obstacle parameters. Wherein, the preset conditions include: the vehicle speed is greater than or equal to the first speed threshold; the lane width is greater than or equal to the first width threshold, and less than the second width threshold; the lane curvature radius is greater than or equal to the preset curvature threshold. Afterwards, when it is determined according to the obstacle parameters of the primary obstacle that the number of secondary obstacles in the primary obstacle is greater than the number threshold, and the primary obstacles are arranged in a preset line type in the lane, it is determined that the current scene is a construction scene. In this way, during the process of automatic driving of the vehicle, it is possible to directly determine whether the current scene of the road is a construction scene according to the vehicle parameters, lane parameters and obstacle parameters, avoiding the collection of a large amount of data to train the model, and is applicable to construction scenes of all roads, improving the applicability and accuracy of construction scene recognition.

[0072] In some embodiments, before determining that the current scene is a construction scene, the construction scene identification method also includes: determining, based on obstacle parameters of the primary obstacles, that the number of secondary obstacles in the primary obstacles is greater than a quantity threshold, and that the primary obstacles are arranged in a preset line type within the lane and last for a preset time period.

[0073] The preset time period may be the time length for displaying a preset number of video frames, for example, the preset time period is the time length for displaying 4 video frames; the preset time period may also be a default value, for example, the preset time period is 0.2s.

[0074] In the above scheme, according to the obstacle parameters of the primary obstacles, it is determined that the number of secondary obstacles in the primary obstacles is greater than the number threshold, and the primary obstacles are arranged in a preset linear shape in the lane and continue for a preset period of time, and then the current scene is determined to be a construction scene. This ensures that the obstacle parameters of the primary obstacles can continue to meet the relevant conditions, further improving the accuracy of construction scene recognition.

[0075] In some embodiments, before determining that the current scene is a construction scene, the method further includes: determining that there are no other obstacles between the vehicle and the primary obstacle.

[0076] In the above scheme, when it is determined based on the obstacle parameters of the primary obstacle that the number of secondary obstacles in the primary obstacle is greater than the number threshold, the primary obstacles are arranged in a preset line in the lane, and it is determined that there are no other obstacles between the vehicle and the primary obstacle, the current scene is determined to be a construction scene. This avoids the influence of other obstacles between the vehicle and the primary obstacle on the recognition result, and improves the accuracy of construction scene recognition.

[0077] In some embodiments, after determining that the current scene is a construction scene, the method further includes: sending an early warning, the early warning being used to request taking over the vehicle.

[0078] In the above solution, after determining that the current scene is a construction scene, the method further includes: sending an early warning, the early warning is used to request to take over the vehicle. After determining that the current scene is a construction scene, the person on the vehicle can be requested to take over the vehicle in advance, avoiding the problem that the automatic driving cannot identify the possible collision due to the complex driving scene.

[0079] In some embodiments, when the vehicle speed is less than a first speed threshold, the construction warning is shielded.

[0080] In the above scheme, when the vehicle speed is less than the first speed threshold, the construction warning is shielded. When the vehicle speed is less than the first speed threshold, the speed is usually relatively low, and the current autonomous driving technology has enough time to identify obstacles in the current environment and avoid them. Therefore, there is no need to request the on-board personnel to take over the vehicle, which avoids the need for manual takeover of the vehicle when the vehicle speed is low, thereby improving the user experience.

[0081] The embodiment of the present application can divide the functional modules of the construction scene recognition device according to the above method example. For example, each functional module can be divided according to each function, or two or more functions can be integrated into one processing unit. The above integrated modules can be implemented in the form of hardware or in the form of software functional modules. It should be noted that the division of modules in the embodiment of the present application is schematic and is only a logical functional division. There may be other division methods in actual implementation.

[0082] like Figure 7 , which is a schematic diagram of the structure of a construction scene recognition device provided in an embodiment of the present application, wherein the construction scene recognition device includes an acquisition module 701 , a determination module 702 , and a processing module 703 .

[0083] The acquisition module 701 is used to acquire vehicle parameters, lane parameters and obstacle parameters of at least one obstacle; the vehicle parameters include vehicle speed, and the lane parameters include lane width and lane curvature radius; the determination module 702 is used to determine a primary obstacle from at least one obstacle according to the obstacle parameters when both the vehicle parameters and the lane parameters meet preset conditions; wherein the preset conditions include: the vehicle speed is greater than or equal to a first speed threshold; the lane width is greater than or equal to a first width threshold and less than a second width threshold; the lane curvature radius is greater than or equal to a preset curvature threshold; the processing module 703 is used to determine that the current scene is a construction scene when it is determined according to the obstacle parameters of the primary obstacle that the number of secondary obstacles in the primary obstacle is greater than the number threshold and the primary obstacles are arranged in a preset line type in the lane.

[0084] In some embodiments, the obstacle parameters include a yaw angle, a size, and an obstacle speed; the determination module 702 is specifically used to: perform an evaluation operation on the obstacle parameters of each obstacle to determine a primary obstacle in at least one obstacle; wherein the evaluation operation includes: when the yaw angle of the first obstacle is greater than or equal to a first angle and less than a second angle, the obstacle speed of the first obstacle is less than a second speed threshold, and the size of the first obstacle is greater than or equal to the first size, determining that the first obstacle is a primary obstacle; the first obstacle is any obstacle among the at least one obstacle.

[0085] In some embodiments, the obstacle parameters include size and quantity; the processing module 703 is specifically used to: determine an obstacle whose size is smaller than the second size in the primary obstacles as a second obstacle; determine an obstacle whose size is greater than or equal to the second size in the primary obstacles as a third obstacle; when the number of second obstacles is greater than a first quantity threshold and the number of third obstacles is greater than a second quantity threshold, determine that the number of secondary obstacles in the primary obstacles is greater than the quantity threshold.

[0086] In some embodiments, the lane parameters also include the lane line position, and the obstacle parameters also include the identification, type and position; the processing module 703 is specifically used to: when the distance between the position of the primary obstacle and the lane line position is less than a first distance threshold, determine that the primary obstacle is within the lane line; determine that the primary obstacles are arranged into a preset line type according to the identification or type of the primary obstacle.

[0087] In some embodiments, the processing module 703 is specifically used to: obtain at least two frames of images corresponding to primary obstacles that are consecutive in time; when consecutive n frames of images include obstacles of the target type and the number of obstacles of the target type in each frame of image is greater than a first type threshold, determine that the primary obstacles are arranged in a preset line type; the primary obstacles include obstacles of the target type; or, when consecutive m frames of images include obstacles of the target type and the number of obstacles of the target type in each frame of image is greater than a second type threshold, and the identifiers of every two obstacles of the target type in each frame of image are different, determine that the primary obstacles are arranged in a preset line type.

[0088] In some embodiments, the processing module 703 is further used to: determine, based on the obstacle parameters of the primary obstacle, that the number of secondary obstacles in the primary obstacle is greater than a quantity threshold, and the primary obstacles are arranged in a preset line in the lane and last for a preset time period.

[0089] In some embodiments, the obstacle parameter includes a type; the determination module 702 is further configured to: determine that there are no other obstacles between the vehicle and the primary obstacle.

[0090] In some embodiments, the construction scene recognition device further includes a sending module 704, and the sending module 704 is used to send an early warning, where the early warning is used to request to take over the vehicle.

[0091] The construction scene recognition device provided in this embodiment can execute the construction scene recognition method provided in the above method embodiment. Its implementation principle and technical effect are similar to those of the above method and will not be repeated here.

[0092] Figure 8 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application.

[0093] like Figure 8 As shown, the embodiment of the present application provides an electronic device, which includes: a processor 801, a memory 802, and a computer program stored in the memory 802 and executable on the processor 801, and when the computer program is executed by the processor 801, each process of the construction scene recognition method in the above method embodiment is implemented. And the same technical effect can be achieved, so it will not be repeated here to avoid repetition.

[0094] An embodiment of the present application provides a computer-readable storage medium, characterized in that a computer program is stored on the computer-readable storage medium. When the computer program is executed by a processor, the various processes of the construction scene recognition method in the above method embodiment are implemented, and the same technical effect can be achieved. To avoid repetition, it will not be repeated here.

[0095] The computer-readable storage medium may be a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, etc.

[0096] An embodiment of the present application provides a computer program product, which stores a computer program. When the computer program is executed by a processor, the various processes of the construction scene recognition method in the above method embodiment are implemented, and the same technical effect can be achieved. To avoid repetition, it will not be repeated here.

[0097] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media that include computer-usable program code.

[0098] In the present application, the processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0099] In this application, memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of a computer-readable medium.

[0100] In this application, computer-readable media includes permanent and non-permanent, removable and non-removable storage media. Storage media can be implemented by any method or technology to store information, and the information can be computer-readable instructions, data structures, modules of programs or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, read-only compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. According to the definition in this article, computer-readable media does not include temporary computer-readable media (transitory media), such as modulated data and carrier waves.

[0101] It should be noted that, in this article, relational terms such as "first" and "second" are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

[0102] The above description is only a specific implementation of the present application, so that those skilled in the art can understand or implement the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments described herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A construction scene recognition method, characterized in that: include: Obtaining vehicle parameters, lane parameters, and obstacle parameters of at least one obstacle; The vehicle parameters include vehicle speed, and the lane parameters include lane width and lane curvature radius; When both the vehicle parameter and the lane parameter meet preset conditions, a primary obstacle is determined from the at least one obstacle according to the obstacle parameter; wherein the preset conditions include: the vehicle speed is greater than or equal to a first speed threshold; the lane width is greater than or equal to a first width threshold and less than a second width threshold; the lane curvature radius is greater than or equal to a preset curvature threshold; When it is determined according to the obstacle parameters of the primary obstacles that the number of secondary obstacles in the primary obstacles is greater than a number threshold, and the primary obstacles are arranged in a preset line type in the lane, it is determined that the current scene is a construction scene.

2. The construction scene recognition method according to claim 1, characterized in that: The obstacle parameters include a yaw angle, a size, and an obstacle speed; and determining a primary obstacle from the at least one obstacle according to the obstacle parameters includes: performing an evaluation operation on obstacle parameters of each obstacle to determine a primary obstacle among the at least one obstacle; The evaluation operation includes: when the yaw angle of the first obstacle is greater than or equal to the first angle and less than the second angle, the obstacle speed of the first obstacle is less than the second speed threshold, and the size of the first obstacle is greater than or equal to the first size, determining that the first obstacle is a primary obstacle; the first obstacle is any one of the at least one obstacle.

3. The construction scene recognition method according to claim 1, characterized in that: The obstacle parameters include size and quantity; and determining, based on the obstacle parameters of the primary obstacle, that the quantity of the secondary obstacles in the primary obstacle is greater than a quantity threshold, comprises: Determine an obstacle of a size smaller than the second size among the primary obstacles as a second obstacle; Determine an obstacle among the primary obstacles whose size is greater than or equal to the second size as a third obstacle; When the number of the second obstacles is greater than the first number threshold and the number of the third obstacles is greater than the second number threshold, it is determined that the number of secondary obstacles in the primary obstacles is greater than the number threshold.

4. The construction scene recognition method according to claim 1, characterized in that: The lane parameters also include the lane line position, and the obstacle parameters also include the identification, type, and position; determining that the primary obstacles are arranged into a preset line type in the lane according to the obstacle parameters of the primary obstacles includes: When the distance between the position of the primary obstacle and the position of the lane line is less than a first distance threshold, determining that the primary obstacle is within the lane line; The primary obstacles are arranged into a preset line type according to the identification or type of the primary obstacles.

5. The construction scene recognition method according to claim 4, characterized in that: The step of determining, according to the identification or type of the primary obstacles, that the primary obstacles are arranged into a preset line type comprises: Acquire at least two temporally consecutive frames of images corresponding to the primary obstacle; When obstacles of the target type are included in all the consecutive n frames of images, and the number of obstacles of the target type in each frame of image is greater than the first type threshold, it is determined that the primary obstacles are arranged in a preset line type; the primary obstacles include obstacles of the target type; Alternatively, when obstacles of the target type are included in m consecutive image frames, the number of obstacles of the target type in each image frame is greater than the second type threshold, and the identifications of every two obstacles of the target type in each image frame are different, it is determined that the primary obstacles are arranged in a preset line type.

6. The construction scene recognition method according to claim 1, characterized in that: Before determining that the current scene is a construction scene, the method further includes: According to the obstacle parameters of the primary obstacles, it is determined that the number of secondary obstacles in the primary obstacles is greater than a number threshold, and the primary obstacles are arranged in a preset line type in the lane and last for a preset time period.

7. The construction scene recognition method according to claim 1, characterized in that: The obstacle parameter includes a type; before determining that the current scene is a construction scene, the method further includes: It is determined that there are no other obstacles between the vehicle and the primary obstacle.

8. The construction scene recognition method according to any one of claims 1 to 7, characterized in that: After determining that the current scene is a construction scene, the method further includes: Send an early warning, the early warning is used to request to take over the vehicle.

9. A construction scene recognition device, characterized in that: include: An acquisition module, used to acquire vehicle parameters, lane parameters and obstacle parameters of at least one obstacle; The vehicle parameters include vehicle speed, and the lane parameters include lane width and lane curvature radius; a determination module, configured to determine a primary obstacle from the at least one obstacle according to the obstacle parameter when both the vehicle parameter and the lane parameter satisfy a preset condition; wherein the preset condition includes: the vehicle speed is greater than or equal to a first speed threshold; the lane width is greater than or equal to a first width threshold and less than a second width threshold; the lane curvature radius is greater than or equal to a preset curvature threshold; The processing module is used to determine that the current scene is a construction scene when it is determined based on the obstacle parameters of the primary obstacle that the number of secondary obstacles in the primary obstacle is greater than a number threshold and the primary obstacles are arranged in a preset line shape in the lane.

10. An electronic device, characterized in that: include: A processor, a memory, and a computer program stored in the memory and executable on the processor, wherein the computer program, when executed by the processor, implements the construction scene recognition method as described in any one of claims 1 to 8.

11. A computer-readable storage medium, characterized in that: include: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the construction scene recognition method according to any one of claims 1 to 8 is implemented.

Images