Vehicle line pressing identification method and device

Abstract

The invention provides a vehicle line pressing identification method and device, and relates to the technical field of artificial intelligence, in particular to the technical fields of image processing, map navigation, automatic driving, intelligent traffic and the like. According to the specific implementation scheme, whether line pressing detection is triggered or not is determined based on driving information of a target vehicle; in response to the triggered line pressing detection, generating a first line pressing detection result and a second line pressing detection result based on the plurality of vehicle front images of the target vehicle; and determining whether the target vehicle presses the line based on the first line pressing detection result and the second line pressing detection result. According to the mode, the accuracy and the reliability of the generated line pressing identification result are improved.

Description

Technical Field

[0001] The present disclosure relates to the field of artificial intelligence technology, specifically to the technical fields of image processing, map navigation, automatic driving, intelligent transportation, etc., and more particularly to a method and device for identifying vehicle crossing a line. Background Art

[0002] The vehicle crossing lane judgment and warning technology currently widely used in the market mainly relies on the surround camera system. This technology usually uses multiple cameras installed around the vehicle to monitor the relationship between the vehicle and the lane line. Summary of the invention

[0003] Embodiments of the present disclosure provide a vehicle crossing line recognition method, device, equipment and storage medium.

[0004] In a first aspect, an embodiment of the present disclosure provides a method for identifying a vehicle crossing a line, the method comprising: determining whether to trigger a crossing line detection based on driving information of a target vehicle; in response to triggering the crossing line detection, generating a first crossing line detection result and a second crossing line detection result based on multiple vehicle front images of the target vehicle; and determining whether the target vehicle crosses the line based on the first crossing line detection result and the second crossing line detection result.

[0005] In a second aspect, an embodiment of the present disclosure provides a vehicle cross-line recognition device, the device comprising: a cross-line trigger module, a cross-line detection module, and a cross-line generation module, wherein the cross-line trigger module is configured to determine whether to trigger cross-line detection based on driving information of a target vehicle; the cross-line detection module is configured to generate a first cross-line detection result and a second cross-line detection result based on multiple vehicle front images of the target vehicle in response to triggering cross-line detection; the cross-line generation module is configured to determine whether the target vehicle crosses the line based on the first cross-line detection result and the second cross-line detection result.

[0006] In a third aspect, an embodiment of the present disclosure provides an electronic device, which includes one or more processors; a storage device on which one or more programs are stored, and when the one or more programs are executed by the one or more processors, the one or more processors implement a vehicle crossing line recognition method as in any embodiment of the first aspect.

[0007] In a fourth aspect, an embodiment of the present disclosure provides a computer-readable medium having a computer program stored thereon, which, when executed by a processor, implements a vehicle crossing line recognition method as in any embodiment of the first aspect.

[0008] In a fifth aspect, an embodiment of the present disclosure provides a computer program product, including a computer program, which, when executed by a processor, implements a vehicle crossing line recognition method as in any embodiment of the first aspect.

[0009] It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present disclosure, nor is it intended to limit the scope of the present disclosure. Other features of the present disclosure will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] Figure 1 is an exemplary system architecture diagram in which the present disclosure may be applied;

[0011] Figure 2 is a flow chart of an embodiment of a vehicle crossing line identification method according to the present disclosure;

[0012] Figure 3 is a flow chart of another embodiment of a vehicle line crossing identification method according to the present disclosure;

[0013] Figure 4a is a schematic diagram of an application scenario of the vehicle crossing line recognition method according to the present disclosure;

[0014] Figure 4b is a schematic diagram of another application scenario of the vehicle line crossing recognition method disclosed in the present invention;

[0015] Figure 5 is a schematic diagram of an embodiment of a vehicle line crossing recognition device according to the present disclosure;

[0016] Figure 6 It is a schematic diagram of the structure of a computer system suitable for implementing the electronic device of the embodiment of the present disclosure. DETAILED DESCRIPTION

[0017] The following is a description of exemplary embodiments of the present disclosure in conjunction with the accompanying drawings, including various details of the embodiments of the present disclosure to facilitate understanding, which should be considered as merely exemplary. Therefore, it should be recognized by those of ordinary skill in the art that various changes and modifications may be made to the embodiments described herein without departing from the scope and spirit of the present disclosure. Similarly, for the sake of clarity and conciseness, descriptions of well-known functions and structures are omitted in the following description.

[0018] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present disclosure may be combined with each other. The present disclosure will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0019] Figure 1 An exemplary system architecture 100 is shown to which an embodiment of the vehicle crossing line recognition method disclosed herein can be applied.

[0020] like Figure 1As shown, the system architecture 100 may include terminal devices 101, 102, 103, a network 104 and a server 105. The network 104 is used to provide a medium for communication links between the terminal devices 101, 102, 103 and the server 105. The network 104 may include various connection types, such as wired, wireless communication links or optical fiber cables, etc.

[0021] Users can use terminal devices 101 , 102 , 103 to interact with server 105 via network 104 to receive or send messages, etc.

[0022] The terminal devices 101, 102, 103 may be hardware or software. When the terminal devices 101, 102, 103 are software, they may be installed in the electronic devices listed above. They may be implemented as multiple software or software modules, or as a single software or software module. No specific limitation is made here.

[0023] Server 105 can be a server that provides various services, for example, determining whether to trigger line-crossing detection based on driving information of the target vehicle; in response to triggering line-crossing detection, generating a first line-crossing detection result and a second line-crossing detection result based on multiple vehicle front images of the target vehicle; and determining whether the target vehicle crosses the line based on the first line-crossing detection result and the second line-crossing detection result.

[0024] It should be noted that the server 105 can be hardware or software. When the server 105 is hardware, it can be implemented as a distributed server cluster consisting of multiple servers, or it can be implemented as a single server. When the server is software, it can be implemented as multiple software or software modules (for example, used to provide vehicle cross-line recognition services), or it can be implemented as a single software or software module. No specific limitation is made here.

[0025] It should be noted that the vehicle crossing line recognition method provided by the embodiment of the present disclosure can be executed by the server 105, or by the terminal devices 101, 102, 103, or by the server 105 and the terminal devices 101, 102, 103 cooperating with each other. Accordingly, the various parts (such as various units, sub-units, modules, sub-modules) included in the vehicle crossing line recognition device can be all arranged in the server 105, or all arranged in the terminal devices 101, 102, 103, or respectively arranged in the server 105 and the terminal devices 101, 102, 103.

[0026] It should be understood that Figure 1 The number of terminal devices, networks and servers in the embodiment is only for illustration. Any number of terminal devices, networks and servers may be provided according to implementation requirements.

[0027] Figure 2The process 200 of an embodiment of a vehicle crossing line identification method is shown. The vehicle crossing line identification method may specifically include the following steps:

[0028] Step 201 : determining whether to trigger line crossing detection based on driving information of a target vehicle.

[0029] In this embodiment, the execution subject (for example, Figure 1 The server 105 or terminal devices 101, 102, 103 in the method can obtain the vehicle's driving position information and / or driving environment information via sensor devices and / or image acquisition devices installed on the target vehicle, such as GPS (Global Positioning System), navigator, driving recorder, etc., and determine whether the driving position information and / or driving environment information meets the conditions for triggering the line pressing detection. If so, the line pressing detection is triggered, otherwise, the line pressing detection is not triggered.

[0030] Among them, the conditions for triggering line crossing detection may include multiple conditions, for example, the vehicle is within a preset distance range of a fork in the road, the vehicle is within a preset distance range of a curve, in an environment with obstructed vision (such as obstacles such as large vehicles, billboards, trees, etc., which affect the line of sight), at night or in bad weather conditions (such as snow, heavy fog, etc.), and when road markings are blurred or missing.

[0031] Specifically, the conditions for triggering the line crossing detection include driving within a preset distance range of the curve. If the execution subject determines through the vehicle's driving position information that the target vehicle has driven within the preset distance range of the curve, the line crossing detection is triggered.

[0032] Step 202 , in response to triggering the line-crossing detection, generating a first line-crossing detection result and a second line-crossing detection result based on a plurality of vehicle front images of the target vehicle.

[0033] In this embodiment, if the executing entity determines to trigger the line-crossing detection based on the driving information of the target vehicle, multiple vehicle front images of the vehicle can be collected via an image acquisition device installed in front of the vehicle (for example, the inside of the windshield, above the hood, etc.), and the vehicle front images can be pre-processed.

[0034] Among them, the image acquisition device can be a component that uses optical imaging principles to form images in existing technologies or future development technologies, such as a camera, a video camera, etc., and this application does not limit this.

[0035] Here, the preprocessing operation is used to indicate an operation for improving image quality, which may specifically include: noise reduction, enhancement, de-distortion and other operations.

[0036] Furthermore, the execution subject may use two different line-crossing detection algorithms to perform line-crossing detection on the image in front of the vehicle after the preprocessing operation, and generate a first line-crossing detection result and a second line-crossing detection result.

[0037] Here, the line-crossing detection algorithm may include multiple types, for example, a model algorithm, that is, directly inputting the image in front of the vehicle into a preset line-crossing detection model to generate a line-crossing detection result; an intersection algorithm, that is, identifying the outline of the vehicle hood in the image in front of the vehicle, taking the midpoint of the front edge of the outline as a marking point, determining the first intersection of the left lane line adjacent to the marking point and the hood outline, and the second intersection of the right lane line adjacent to the marking point and the hood outline, respectively calculating the first and second distances in the horizontal direction between the first intersection and the second intersection and the marking point, and determining whether the vehicle crosses the line based on whether the first distance or the second distance in the horizontal direction is less than a preset distance threshold; an inverse perspective algorithm, that is, performing an inverse perspective transformation on the image in front of the vehicle according to the parameters of the image acquisition device, and determining whether the vehicle crosses the line based on whether the actual distance corresponding to the distance between the visual center point and the adjacent lane line in the image after the inverse perspective change is greater than half the width of the vehicle body, etc.

[0038] Step 203, determining whether the target vehicle crosses the line based on the first line-crossing detection result and the second line-crossing detection result.

[0039] In this embodiment, after determining the first line-crossing detection result and the second line-crossing detection result, the executing entity can directly determine whether the target vehicle crosses the line based on the first line-crossing detection result and the second line-crossing detection result, or it can determine whether the target vehicle crosses the line based on the first line-crossing detection result, the first confidence level of the first line-crossing detection result, the second line-crossing detection result, and the second confidence level of the second line-crossing detection result. This application does not limit this.

[0040] Specifically, if the first line-crossing detection result is line-crossing and the second line-crossing detection result is also line-crossing, it can be determined that the target vehicle crosses the line.

[0041] It should be noted that the first confidence level of the first cross-line detection result and the second confidence level of the second cross-line detection result can usually be determined based on the degree to which the cross-line detection algorithm corresponding to the cross-line detection result is affected by the target vehicle's driving position information and / or driving environment information.

[0042] Generally, the more the cross-line detection algorithm is affected by the target vehicle's driving position information and / or driving environment information, the lower the corresponding confidence of the cross-line detection algorithm; if the less the cross-line detection algorithm is affected by the target vehicle's driving position information and / or driving environment information, the higher the corresponding confidence of the cross-line detection algorithm. For example, the worse the driving environment, the lower the accuracy of the recognition result generated by the cross-line detection algorithm.

[0043] In some optional implementations of this embodiment, after determining that the target vehicle crosses the line, the execution entity may output line crossing warning information.

[0044] Among them, the output methods of the line pressure warning information can include multiple methods, such as sound warning, visual warning, vibration warning, etc.

[0045] The vehicle crossing line recognition method provided by the embodiments of the present disclosure determines whether to trigger the crossing line detection based on the driving information of the target vehicle; in response to triggering the crossing line detection, generates a first crossing line detection result and a second crossing line detection result based on multiple vehicle front images of the target vehicle; determines whether the target vehicle crosses the line based on the first crossing line detection result and the second crossing line detection result, thereby realizing the determination of whether the target vehicle crosses the line based on the two crossing line detection results of the vehicle front images, and improving the accuracy and reliability of the generated crossing line recognition results.

[0046] In some optional embodiments, determining whether to trigger line crossing detection based on driving information of the target vehicle includes: in response to detecting that the target vehicle has traveled within a preset distance range of a fork in the road based on driving position information, determining to trigger line crossing detection.

[0047] In this implementation, the driving information may include driving position information, and the execution subject may determine the road network structure of the area where the current position is located in the map data according to the current position.

[0048] The road network structure may include multiple road sections and the connection relationship between each road section.

[0049] Furthermore, the execution entity may determine whether the target vehicle has traveled to within a preset distance range of the fork in the road based on the current position and the road network structure.

[0050] Among them, the preset distance can be 100 meters, 200 meters, etc., and this application does not limit this.

[0051] If the target vehicle travels within a preset distance range of the fork in the road, it can be determined that the line crossing detection is triggered.

[0052] This implementation method realizes the line crossing detection in the fork intersection scenario.

[0053] Further references Figure 3 , which shows Figure 2 The process 300 of another embodiment of the vehicle crossing line identification method is shown. In this embodiment, the process 300 of the vehicle crossing line identification method may include the following steps:

[0054] Step 301, determining whether to trigger line crossing detection based on driving information of the target vehicle.

[0055] In this embodiment, the implementation details and technical effects of step 301 can refer to the description of step 201 and will not be repeated here.

[0056] Step 302: Input a plurality of vehicle front images of the target vehicle into a pre-trained line-crossing detection model to obtain a first line-crossing detection result.

[0057] In this embodiment, the execution entity may first determine a line-crossing detection model that matches the driving information from multiple pre-trained line-crossing detection models, and then input multiple vehicle front images of the target vehicle into the line-crossing detection model that matches the driving information to obtain a first line-crossing detection result.

[0058] The line-crossing detection model may be trained based on a vehicle front image sample annotated with a first line-crossing detection result.

[0059] Here, the pressure line detection model can be a neural network model in existing technology or future development technology, such as a convolutional neural network model, a recurrent neural network model, etc.

[0060] Step 303: determining a target lane line and a target distance between the target lane line and the target vehicle based on the plurality of vehicle front images.

[0061] In this embodiment, the execution entity can directly identify multiple vehicle front images to determine the target distance between the target lane line and the target vehicle; it can also first determine the map lane information of the current driving position of the target vehicle, and perform auxiliary recognition on multiple vehicle front images based on the map lane information to determine the target distance between the target lane line (i.e., the lane lines immediately adjacent to the left and right sides of the lane where the target vehicle is currently located) and the target vehicle. This application does not limit this.

[0062] Here, the distance between the target lane line and the target vehicle may be represented by the distance between a point on the center axis of the target vehicle and the target lane line.

[0063] The target lane line may include a left lane line and / or a right lane line adjacent to the target vehicle, and the target distance may include a first distance between the target vehicle and the left lane line and / or a second distance between the target vehicle and the right lane line.

[0064] Specifically, the execution subject can identify the hood outline of the target vehicle in the image in front of the vehicle, take the midpoint of the front edge of the outline as the marking point, determine the first intersection point between the left lane line adjacent to the marking point and the hood outline, and the second intersection point between the right lane line adjacent to the marking point and the hood outline, calculate the initial first distance between the first intersection point and the marking point in the horizontal direction, and the initial second distance between the second intersection point and the marking point in the horizontal direction. Further, according to the product of the ratio value of the actual distance to the image distance and the initial first distance, the first distance between the target vehicle and the adjacent left lane line is determined, and according to the product of the ratio value of the actual distance to the image distance and the initial second distance, the second distance between the target vehicle and the adjacent right lane line is determined.

[0065] Step 304: Generate a second lane crossing detection result based on the target distance between the target lane line and the target vehicle.

[0066] In this embodiment, the executing entity may match the first distance between the target vehicle and the adjacent left lane line (i.e., the left lane line of the lane where the target vehicle is located), and the second distance between the target vehicle and the adjacent right lane line (i.e., the right lane line of the lane where the target vehicle is located) with a preset width threshold, respectively. If the first distance or the second distance is less than or equal to the preset width threshold, it can be determined that the second line-crossing detection result is line-crossing. If the first distance and the second distance are both greater than the preset width threshold, it can be determined that the second line-crossing detection result is not line-crossing.

[0067] Here, the preset width threshold may be determined based on the half body width of the target vehicle. Specifically, the preset width threshold may be determined by multiplying the target body width by a preset ratio value (eg, 1.1, 1.2, 0.95, etc.).

[0068] Among them, the preset ratio value can be determined according to the specific model of the target vehicle.

[0069] Step 305: Determine whether the target vehicle crosses the line based on the first line-crossing detection result and the second line-crossing detection result.

[0070] In this embodiment, the implementation details and technical effects of step 305 can refer to the description of step 203 and will not be repeated here.

[0071] The above-mentioned embodiments of the present disclosure highlight the input of multiple vehicle front images of the target vehicle into a pre-trained line-crossing detection model to obtain a first line-crossing detection result, determine the target lane line and the target distance between the target lane line and the target vehicle based on the multiple vehicle front images, generate a second line-crossing detection result based on the target distance between the target lane line and the target vehicle, and then determine whether the target vehicle crosses the line based on the first line-crossing detection result and the second line-crossing detection result, thereby realizing the fusion of the first line-crossing detection result based on the model and the second line-crossing detection result based on the lane line detection result, and further improving the accuracy and reliability of the generated line-crossing detection result.

[0072] In some optional embodiments, determining the target lane line and the target distance between the target lane line and the target vehicle based on multiple images in front of the vehicle includes: determining map lane information of the current driving position; determining the target lane line and the target distance between the target lane line and the target vehicle based on the map lane information and multiple images in front of the vehicle.

[0073] In this implementation, the execution entity may first determine the current driving position of the target vehicle, and then determine the map lane information of the current driving position in the map data according to the current driving position of the target vehicle.

[0074] Among them, the map lane information may include the total number of lanes, lane line type, lane line spacing, lane line color, lane line width and other information.

[0075] Furthermore, the execution entity may perform auxiliary recognition on multiple vehicle front images based on the map lane information of the current driving position of the target vehicle to determine the target distance between the target lane line and the target vehicle.

[0076] This implementation method calibrates the target distance between the target lane line and the target vehicle by using the map lane information, thereby effectively improving the accuracy of the determined target distance.

[0077] In some optional methods, a target distance between a target lane line and a target vehicle is determined based on multiple images in front of the vehicle, including: using an inverse perspective transformation to convert the multiple images in front of the vehicle into a top view sequence; based on the position information of the target lane line and the top view sequence, determining the target distance between the target lane line and the target vehicle; based on the target distance between the target lane line and the target vehicle, generating a second line pressing detection result includes: determining the second line pressing detection result based on the target distance and half the body width of the target vehicle.

[0078] In this implementation, the execution entity may first use an inverse perspective transformation to convert a plurality of vehicle front images into a sequence of top view images.

[0079] Among them, inverse perspective transformation is an image processing technology used to convert a perspective projection image into a top view or a bird's-eye view.

[0080] Furthermore, the execution subject can determine the road network lane line spacing, i.e., the distance between the left lane line and the right lane line, based on the position information of the target lane line, i.e., the left lane line adjacent to the target vehicle and the right lane line closest to the target vehicle. The view lane line spacing, i.e., the visual left spacing between the visual center point and the left lane line in the image in front of the vehicle, and the visual right spacing between the visual center point and the right lane line in the image in front of the vehicle, are determined based on the top view sequence.

[0081] Furthermore, the execution entity can determine the target distance, that is, the first distance between the target vehicle and the left lane line, and the second distance between the target vehicle and the right lane line according to the proportional relationship between the visual lane line spacing and the road network lane line spacing, as well as the visual left spacing and the visual right spacing.

[0082] Furthermore, the executing entity may respectively compare the first distance and the second distance with the half body width of the target vehicle. If the first distance or the second distance is less than or equal to the half body width of the target vehicle, the second line pressing detection result may be determined to be line pressing; if the first distance and the second distance are both greater than the half body width of the target vehicle, the second line pressing detection result may be determined to be no line pressing.

[0083] This implementation method further improves the reliability of the determined pressing line result by using the top-down perspective transformation to determine the pressing line result.

[0084] In some optional embodiments, the method also includes: in response to determining that the target vehicle has not crossed the line, generating a predicted trajectory based on the current moving path of the target vehicle and vehicle control information; based on the predicted trajectory, determining the predicted distance between the target lane line and the target vehicle; based on the predicted distance and half the body width of the target vehicle, determining the probability of the target vehicle crossing the line; in response to determining that the probability of crossing the line is greater than a probability threshold, outputting an alarm message that the target vehicle is about to cross the line.

[0085] In this implementation, if the execution entity determines that the target vehicle has not crossed the line, the execution entity can generate a predicted trajectory based on the current moving path and vehicle control information of the target vehicle.

[0086] Among them, the vehicle control information may include vehicle speed control information, steering control information, braking control information, etc.

[0087] Here, the execution entity may further determine the predicted trajectory of the target vehicle based on the current moving path of the target vehicle, vehicle control information, environmental information of the current road section, and road condition information.

[0088] Specifically, the execution entity may input the current moving path of the target vehicle, vehicle control information, environmental information of the current road section, and road condition information into a preset trajectory prediction model to generate a predicted trajectory of the target vehicle.

[0089] Furthermore, the execution entity may determine the predicted distance between the target lane line and the target vehicle based on the predicted trajectory, and determine the probability of the target vehicle crossing the lane line based on the predicted distance and the half body width of the target vehicle.

[0090] Here, the probability of crossing the line is usually positively correlated with the difference between the half body width and the predicted distance (the difference is a positive value), that is, the larger the difference, the greater the probability of crossing the line, and the smaller the difference, the smaller the probability of crossing the line.

[0091] Furthermore, if the probability of crossing the line is greater than the probability threshold, an alarm message of impending crossing the line is input.

[0092] This implementation method predicts the trajectory of the target vehicle when the target vehicle has not crossed the line; based on the predicted trajectory, determines the probability of the target vehicle crossing the line, and outputs an imminent crossing line warning message when the probability of crossing the line is greater than a probability threshold, which helps to output an imminent crossing line warning in a timely manner.

[0093] In some optional embodiments, the method further includes: in response to determining that the target vehicle crosses the line and the length of time the vehicle crosses the line exceeds a preset length of time, outputting a line crossing warning message.

[0094] In this implementation, if the execution entity determines that the target vehicle crosses the line based on the first detection result and the second detection result, it can further determine whether the crossing time exceeds the preset time. If so, the execution entity can output the crossing warning information.

[0095] Among them, the pressing time can be 5s, 8s, etc.

[0096] This implementation method helps to avoid false alarms by outputting alarm information when the line pressing duration exceeds a preset duration.

[0097] In some optional embodiments, in response to triggering the cross-line detection, a first cross-line detection result and a second cross-line detection result are generated based on multiple vehicle front images of the target vehicle, including: in response to triggering the cross-line detection, a first cross-line detection result and a second cross-line detection result are generated based on multiple vehicle front images of the target vehicle captured by a driving recorder.

[0098] In this implementation, if the line-crossing detection is triggered, the execution entity may generate a first line-crossing detection result and a second line-crossing detection result using a plurality of vehicle front images of the target vehicle collected by the driving recorder.

[0099] Among them, a driving recorder is a device installed on a car (usually installed at the top center of the front windshield or on the right side near the rearview mirror) to record video and audio during the vehicle's driving process to record various situations during driving. The image in front of the vehicle captured by the driving recorder can be a color image or a grayscale image.

[0100] This implementation method takes advantage of the fact that the dashcam is usually installed at a high position in front of the vehicle, which helps to collect images in front of the vehicle in a timely and effective manner and then determine the cross-line detection results.

[0101] In some optional embodiments, a target lane line and a target distance between the target lane line and the target vehicle are determined based on multiple vehicle front images, including: in response to determining that a guide line area exists in multiple vehicle front images of the target vehicle, determining a moving direction of the guide line area relative to the target vehicle; based on the moving direction of the guide line area relative to the target vehicle, determining a target moving path of the target vehicle at a fork in the guide line area; determining an area boundary line in the guide line area corresponding to the target moving path as the target lane line; and determining a target distance between the target lane line and the target vehicle based on multiple vehicle front images.

[0102] In this implementation, if the executing entity detects the existence of a guide line area in multiple vehicle front images of the target vehicle, it can determine the relative position relationship between the guide line area and the target vehicle based on the multiple vehicle front images, and then determine the moving direction of the guide line area relative to the target vehicle based on the change in the relative position relationship corresponding to the multiple vehicle front images.

[0103] Here, the number of guide line areas in an image in front of a vehicle may be one or more, and this application does not impose any limitation on this.

[0104] Among them, the guide line area is used to indicate one or more white V-shaped lines or diagonal line areas set according to the terrain of the intersection. The guide line area is usually set at road intersections, ramp entrances, exits, etc. to guide vehicles to travel along designated routes and reduce traffic conflicts.

[0105] Furthermore, the execution entity may determine a target moving path of the target vehicle at a fork in the road corresponding to the guide line area according to the movement direction of the identified one or more guide line areas relative to the target vehicle.

[0106] Specifically, the number of guide line areas is one. If the guide line area is to the left rear relative to the moving direction of the target vehicle, it can be determined that the target vehicle enters the right side of the guide line area, that is, the target moving path of the target vehicle at the fork in the guide line area is the right road of the guide line area.

[0107] If the guide line area is moving to the right rear relative to the target vehicle, it can be determined that the target vehicle enters the left side of the guide line area, that is, the target moving path of the target vehicle at the fork corresponding to the guide line area is the left road of the guide line area.

[0108] Furthermore, the execution entity may determine the area boundary line corresponding to the target moving path in the guide line area as the target lane line; and determine the target distance between the target lane line and the target vehicle based on multiple vehicle front images.

[0109] This implementation method can identify the target lane line directly based on the image in front of the vehicle collected by the on-board imaging device in scenarios with weak GPS signals such as elevated intersections, and determine the target distance between the target lane line and the target vehicle based on multiple images in front of the vehicle.

[0110] In some optional embodiments, based on the moving direction of the guide line area relative to the target vehicle, the target moving path of the target vehicle at the fork in the road corresponding to the guide line area is determined, including: in response to determining that there are multiple guide line areas, determining the moving directions of multiple guide line areas relative to the target vehicle, and the spacing distances between the multiple guide line areas and the target vehicle; based on the moving directions and spacing distances of the multiple guide line areas, determining the target moving path of the target vehicle at the fork in the road.

[0111] In this implementation, the number of multi-streamline areas can be multiple, for example, 2, 3, etc., and the flow diversion directions of each flow-guiding area can be different. The execution subject can identify the flow-guiding area in the image in front of the vehicle by using an image recognition algorithm, determine the relative position relationship between the target vehicle and the flow-guiding area, and then determine the moving direction of the multiple flow-guiding areas relative to the target vehicle and the interval distance between the multiple flow-guiding areas and the target vehicle according to the change of the relative position relationship in the images in front of the multiple vehicles.

[0112] After determining the moving directions and spacing distances of multiple guide line areas, the execution entity can determine the target moving path of the target vehicle at the fork in the road based on the moving directions and spacing distances of multiple, for example, 3, 4, etc., guide line areas.

[0113] The moving directions of the multiple guide line areas relative to the target vehicle may be the same or different, and the interval distances between the multiple guide line areas and the target vehicle are usually different.

[0114] Specifically, there are three guide line areas, namely the first guide line area, the second guide line area and the third guide line area. If the moving direction of the first guide line area and the second guide line area relative to the target vehicle is the left rear, and the moving direction of the third guide line area relative to the target vehicle is the right rear, and among the first guide line area, the second guide line area and the third guide line area, the interval distance between the third guide line area and the target vehicle is the smallest, then it can be determined that the target vehicle enters the left lane of the third guide line area, that is, the left lane of the third guide area is the target moving path.

[0115] This implementation method realizes the identification of the target movement path in the scenario of multiple guide line areas.

[0116] In some optional embodiments, a target moving path of a target vehicle at a fork in the road is determined based on the moving directions and spacing distances of multiple guide line sub-areas, including: in response to the multiple guide line areas having the same moving directions, determining a target area closest to the target vehicle based on the spacing distances of the multiple guide line areas; and determining the target moving path of the target vehicle at the fork in the road based on the moving direction of the target area relative to the target vehicle.

[0117] In this implementation, if there are multiple guide line areas and the moving directions of the multiple guide line areas relative to the target vehicle are the same, the execution entity can determine the target area closest to the target vehicle based on the interval distances between the multiple guide line areas and the target vehicle.

[0118] Furthermore, the execution entity may determine the target moving path of the target vehicle at the fork in the road according to the moving direction of the target area relative to the target vehicle.

[0119] Specifically, there are three guide line areas, namely the first guide line area, the second guide line area and the third guide line area. If the moving directions of the first guide line area, the second guide line area and the third guide line area relative to the target vehicle are all left rear, and among the first guide line area, the second guide line area and the third guide line area, the interval distance between the second guide line area and the target vehicle is the smallest, then the second guide line area can be determined as the target area, and the target moving path is the right lane of the second guide line area.

[0120] This implementation method realizes the identification of the target moving path in the scenario where there are multiple diversion areas and the moving directions of the diversion line areas are the same.

[0121] In some optional embodiments, based on the moving direction of the guide line area relative to the target vehicle, the target moving path of the target vehicle at the fork in the road corresponding to the guide line area is determined, including: in response to determining that the moving directions of the first guide line area and the second guide line area relative to the target vehicle are both left rear, and the first spacing distance between the first guide line area and the target vehicle is greater than the second spacing distance between the second guide line area and the target vehicle, determining the target moving path to be the right side road of the second guide line sub-area; in response to determining that the moving directions of the first guide line area and the second guide line area are both right rear, and the first spacing distance between the first guide line area and the target vehicle is less than the second spacing distance between the second guide line area and the target vehicle, determining the target moving path to be the left side road of the first guide line area; in response to determining that the moving direction of the first guide line area relative to the target vehicle is left rear and the moving direction of the second guide line area relative to the target vehicle is right rear, determining the target moving path to be the middle road between the first guide line area and the second guide line area.

[0122] In this implementation, there may be multiple guide line areas, namely, a first guide line area and a second guide line area, and the first guide line area and the second guide line area have different guiding directions.

[0123] If the moving directions of the first guide line area and the second guide line area relative to the target vehicle are both left rear, and the first interval distance between the first guide line area and the target vehicle is greater than the second interval distance between the second guide line area and the target vehicle, it can be determined that the target vehicle has entered the right side of the second guide line area, that is, the target moving path of the target vehicle at the fork in the guide line area corresponding to the guide line area is the right side road of the second guide line area.

[0124] If the moving directions of the first guide line area and the second guide line area relative to the target vehicle are both right rear, and the first interval distance between the first guide line area and the target vehicle is smaller than the second interval distance between the second guide line area and the target vehicle, it can be determined that the target vehicle has entered the left side of the first guide line area, that is, the target moving path of the target vehicle at the fork in the road corresponding to the guide line area is the left road of the first guide line area.

[0125] If the first guide line area is moving to the left rear relative to the target vehicle, and the second guide line area is moving to the right rear relative to the target vehicle, it can be determined that the target vehicle enters the middle road between the first guide line area and the second guide line area, that is, the target moving path of the target vehicle at the fork in the guide line area corresponding to the middle road between the first guide line area and the second guide line area.

[0126] This implementation method realizes yaw recognition in a scenario where there are two guide line areas at a fork in the road.

[0127] Continue to see Figure 4a , Figure 4b , Figure 4a , 4b It is a schematic diagram of an application scenario of the vehicle crossing line recognition method according to this embodiment.

[0128] The execution entity can determine the current position of the target vehicle 401 via GPS, and determine the road network structure of the area where the current position is located in the pre-stored map based on the current position, and determine whether the target vehicle has traveled to a preset distance range of a fork in the road (for example, an elevated fork in the road). If so, the line crossing detection is triggered.

[0129] Furthermore, the execution entity can collect multiple vehicle front images of the target vehicle 401 through a vehicle-mounted imaging device, such as a driving recorder, and input the multiple vehicle front images of the target vehicle into a pre-trained line crossing detection model to obtain a first line crossing detection result, such as crossing the left lane line.

[0130] Here, the executing entity may also input multiple vehicle front images, local road network data of the current section of the target vehicle determined based on GPS (the accuracy of GPS at elevated vehicle intersections is poor), and EKF (Extended Kalman Filter) trajectory determined based on IMU (Global Positioning System, inertial measurement unit) into a pre-trained line crossing detection model to obtain a first line crossing detection result.

[0131] Furthermore, if there is a guide line area in the multiple vehicle front images of the target vehicle 401, such as the guide line area 402, the moving direction of the guide line area 402 relative to the target vehicle can be determined; and based on the moving direction of the guide line area 402 relative to the target vehicle, the target moving path of the target vehicle at the fork in the guide line area corresponding to the guide line area is determined, such as the right side road of the guide line area 402; the area boundary line corresponding to the target moving path in the guide line area 402 is determined as the target lane line; the target distance between the target lane line and the target vehicle is determined based on the multiple vehicle front images, and a second line pressing detection result of whether the target vehicle presses the line is determined based on the target distance, such as pressing the boundary line of the left guide line area.

[0132] Furthermore, the execution entity may determine whether the target vehicle crosses a line, such as a boundary line of a left guide line area, based on the first line-crossing detection result and the second line-crossing detection result.

[0133] Further references Figure 5 As an implementation of the methods shown in the above figures, the present disclosure provides an embodiment of a vehicle crossing line recognition device. Figure 2 The method embodiments shown correspond.

[0134] like Figure 5 As shown, the vehicle cross-line recognition device 500 of this embodiment includes: a cross-line triggering module 501 , a cross-line detection module 502 and a cross-line generation module 503 .

[0135] The line-crossing triggering module 501 is configured to determine whether to trigger the line-crossing detection based on the driving information of the target vehicle.

[0136] The line-crossing detection module 502 is configured to generate a first line-crossing detection result and a second line-crossing detection result based on a plurality of vehicle front images of the target vehicle in response to triggering the line-crossing detection.

[0137] The line-crossing generation module 503 is configured to determine whether the target vehicle crosses the line based on the first line-crossing detection result and the second line-crossing detection result.

[0138] In some optional modes of this embodiment, the cross-line detection module further includes: a first detection unit, configured to input multiple vehicle front images of the target vehicle into a pre-trained cross-line detection model to obtain a first cross-line detection result; a distance determination unit, configured to determine the target lane line and the target distance between the target lane line and the target vehicle based on the multiple vehicle front images; and a second detection unit, configured to generate a second cross-line detection result based on the target distance between the target lane line and the target vehicle.

[0139] In some optional embodiments of this embodiment, the distance determination unit is further configured to: determine the map lane information of the current driving position; and determine the target lane line and the target distance between the target lane line and the target vehicle based on the map lane information and multiple vehicle front images.

[0140] In some optional modes of this embodiment, the distance determination unit is further configured to: utilize inverse perspective transformation to convert multiple vehicle front images into a top view sequence; determine the target distance between the target lane line and the target vehicle based on the position information of the target lane line and the top view sequence; and the second detection unit is further configured to: determine the second pressure line detection result based on the target distance and the size of half the body width of the target vehicle.

[0141] In some optional modes of the present embodiment, the distance determination unit is further configured to: in response to determining that there is a guide line area in multiple vehicle front images of the target vehicle, determine the moving direction of the guide line area relative to the target vehicle; based on the moving direction of the guide line area relative to the target vehicle, determine the target moving path of the target vehicle at the fork in the road corresponding to the guide line area; determine the area boundary line in the guide line area corresponding to the target moving path as the target lane line; and determine the target distance between the target lane line and the target vehicle based on multiple vehicle front images.

[0142] In some optional modes of the present embodiment, based on the moving direction of the guide line area relative to the target vehicle, the target moving path of the target vehicle at the fork in the road corresponding to the guide line area is determined, including: in response to determining that the number of guide line areas is multiple, determining the moving directions of multiple guide line areas relative to the target vehicle, and the interval distances between the multiple guide line areas and the target vehicle; based on the moving directions and interval distances of the multiple guide line areas, determining the target moving path of the target vehicle at the fork in the road.

[0143] In some optional modes of the present embodiment, the target moving path of the target vehicle at the fork in the road is determined based on the moving directions and spacing distances of multiple guide line sub-areas, including: in response to the same moving directions of multiple guide line areas, determining the target area closest to the target vehicle based on the spacing distances of multiple guide line areas; determining the target moving path of the target vehicle at the fork in the road based on the moving direction of the target area relative to the target vehicle.

[0144] In some optional modes of the present embodiment, based on the moving direction of the guide line area relative to the target vehicle, the target moving path of the target vehicle at the fork in the guide line area is determined, including: in response to determining that the moving directions of the first guide line area and the second guide line area relative to the target vehicle are both left rear, and the first spacing distance between the first guide line area and the target vehicle is greater than the second spacing distance between the second guide line area and the target vehicle, determining the target moving path to be the right side road of the second guide line sub-area; in response to determining that the moving directions of the first guide line area and the second guide line area relative to the target vehicle are both right rear, and the first spacing distance between the first guide line area and the target vehicle is less than the second spacing distance between the second guide line area and the target vehicle, determining the target moving path to be the left side road of the first guide line area; in response to determining that the moving direction of the first guide line area relative to the target vehicle is left rear and the moving direction of the second guide line area relative to the target vehicle is right rear, determining the target moving path to be the middle road between the first guide line area and the second guide line area.

[0145] In some optional aspects of this embodiment, the cross-line detection module is further configured to: in response to triggering cross-line detection, generate a first cross-line detection result and a second cross-line detection result based on multiple vehicle front images of the target vehicle captured by the driving recorder.

[0146] In some optional modes of the present embodiment, the device also includes a line crossing warning module, which is configured to generate a predicted trajectory based on the current moving path and vehicle control information of the target vehicle in response to determining that the target vehicle has not crossed the line; determine the predicted distance between the target lane line and the target vehicle based on the predicted trajectory; determine the probability of the target vehicle crossing the line based on the predicted distance and half the body width of the target vehicle; and output a warning message of the imminent crossing of the line in response to determining that the probability of crossing the line is greater than a probability threshold.

[0147] In some optional aspects of this embodiment, the device further includes a line crossing alarm module configured to output line crossing alarm information in response to determining that the target vehicle crosses the line and the length of time the line is crossed exceeds a preset length of time.

[0148] In the technical solution disclosed herein, the acquisition, storage and application of user personal information involved are in compliance with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0149] According to an embodiment of the present disclosure, the present disclosure also provides an electronic device, a readable storage medium and a computer program product.

[0150] Figure 6 A schematic block diagram of an example electronic device 600 that can be used to implement an embodiment of the present disclosure is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present disclosure described and / or required herein.

[0151] like Figure 6 As shown, the device 600 includes a computing unit 601, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 602 or a computer program loaded from a storage unit 608 into a random access memory (RAM) 603. In the RAM 603, various programs and data required for the operation of the device 600 can also be stored. The computing unit 601, the ROM 602, and the RAM 603 are connected to each other via a bus 604. An input / output (I / O) interface 605 is also connected to the bus 604.

[0152] A number of components in the device 600 are connected to the I / O interface 605, including: an input unit 606, such as a keyboard, a mouse, etc.; an output unit 607, such as various types of displays, speakers, etc.; a storage unit 608, such as a disk, an optical disk, etc.; and a communication unit 609, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 609 allows the device 600 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0153] The computing unit 601 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the computing unit 601 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, digital signal processors (DSPs), and any appropriate processors, controllers, microcontrollers, etc. The computing unit 601 performs the various methods and processes described above, such as a method for vehicle pressure line identification. For example, in some embodiments, the method for vehicle pressure line identification may be implemented as a computer software program, which is tangibly contained in a machine-readable medium, such as a storage unit 608. In some embodiments, part or all of the computer program may be loaded and / or installed on the device 600 via ROM 602 and / or communication unit 609. When the computer program is loaded into RAM 603 and executed by the computing unit 601, one or more steps of the method for vehicle pressure line identification described above may be performed. Alternatively, in other embodiments, the computing unit 601 may be configured to perform a method for vehicle pressure line identification in any other appropriate manner (e.g., by means of firmware).

[0154] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0155] The program code for implementing the method of the present disclosure may be written in any combination of one or more programming languages. These program codes may be provided to a processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that the program code, when executed by the processor or controller, enables the functions / operations specified in the flow chart and / or block diagram to be implemented. The program code may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0156] In the context of the present disclosure, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or equipment. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or equipment, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0157] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user can provide input to the computer. Other types of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0158] The systems and techniques described herein may be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such back-end components, middleware components, or front-end components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), and the Internet.

[0159] A computer system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The relationship between the client and the server is generated by computer programs running on the corresponding computers and having a client-server relationship with each other. The server may be a cloud server, also known as a cloud computing server or a cloud host, which is a host product in the cloud computing service system to solve the defects of difficult management and weak business scalability in traditional physical hosts and virtual private servers (VPS) services. The server can also be divided into a server of a distributed system, or a server combined with a blockchain.

[0160] According to the technical solution of the embodiment of the present disclosure, the accuracy and reliability of the generated pressure line recognition results are effectively improved.

[0161] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps recorded in this disclosure can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solutions provided by this disclosure can be achieved, and this document does not limit this.

[0162] The above specific implementations do not constitute a limitation on the protection scope of the present disclosure. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present disclosure shall be included in the protection scope of the present disclosure.

Claims

1. A vehicle crossing line recognition method, comprising: Determining whether to trigger line crossing detection based on driving information of the target vehicle, wherein the driving information includes driving position information and / or driving environment information; In response to triggering the line pressing detection, generating a first line pressing detection result and a second line pressing detection result based on a plurality of vehicle front images of the target vehicle; Based on the first line-crossing detection result and the second line-crossing detection result, it is determined whether the target vehicle crosses the line.

2. The method according to claim 1, wherein: The generating of the first pressure line detection result and the second pressure line detection result based on the plurality of vehicle front images of the target vehicle comprises: Inputting a plurality of vehicle front images of the target vehicle into a pre-trained line-crossing detection model to obtain the first line-crossing detection result; Determine a target lane line and a target distance between the target lane line and the target vehicle based on the plurality of vehicle front images; The second lane crossing detection result is generated based on the target distance between the target lane line and the target vehicle.

3. The method according to claim 2, wherein: The determining, based on the plurality of vehicle front images, a target lane line and a target distance between the target lane line and the target vehicle comprises: Determine the map lane information of the current driving position; Based on the map lane information and the plurality of vehicle front images, a target lane line and a target distance between the target lane line and a target vehicle are determined.

4. The method according to claim 2, wherein: The determining a target distance between a target lane line and a target vehicle based on the plurality of vehicle front images comprises: Using an inverse perspective transformation, converting the plurality of vehicle front images into a sequence of top-view images; Determine a target distance between the target lane line and the target vehicle based on the position information of the target lane line and the top view sequence; and The generating the second lane crossing detection result based on the target distance between the target lane line and the target vehicle includes: The second line pressing detection result is determined based on the target distance and the half body width of the target vehicle.

5. The method according to claim 2, wherein: The determining, based on the plurality of vehicle front images, a target lane line and a target distance between the target lane line and the target vehicle comprises: In response to determining that a guide line area exists in the plurality of vehicle front images of the target vehicle, determining a moving direction of the guide line area relative to the target vehicle; Determining a target moving path of the target vehicle at a fork in the guide line area based on the moving direction of the guide line area relative to the target vehicle; Determine a region boundary line in the guide line region corresponding to the target moving path as a target lane line; A target distance between a target lane line and a target vehicle is determined based on the plurality of vehicle front images.

6. The method according to claim 5, wherein: The step of determining a target moving path of the target vehicle at a fork in the guide line area corresponding to the guide line area based on the moving direction of the target vehicle relative to the guide line area includes: In response to determining that there are a plurality of guide line areas, determining moving directions of the plurality of guide line areas relative to the target vehicle and interval distances between the plurality of guide line areas and the target vehicle; Based on the moving directions and the interval distances of the plurality of guide line areas, a target moving path of the target vehicle at the fork in the road is determined.

7. The method according to claim 6, wherein: The step of determining a target moving path of the target vehicle at the fork in the road based on the moving directions and the interval distances of the plurality of guide line sub-areas comprises: In response to the moving directions of the plurality of guide line areas being the same, determining a target area closest to the target vehicle based on the interval distances of the plurality of guide line areas; Based on the moving direction of the target area relative to the target vehicle, a target moving path of the target vehicle at the fork in the road is determined.

8. The method according to claim 5, wherein: There are multiple guide line areas, the multiple guide line areas include a first guide line area and a second guide line area, and the first guide line area and the second guide line area have different guiding directions; The step of determining a target moving path of the target vehicle at a fork in the guide line area corresponding to the guide line area based on the moving direction of the target vehicle relative to the guide line area includes: In response to determining that the moving directions of the first guide line area and the second guide line area relative to the target vehicle are both left rear, and a first spacing distance between the first guide line area and the target vehicle is greater than a second spacing distance between the second guide line area and the target vehicle, determining that the target moving path is a right side road of the second guide line sub-area; In response to determining that the moving directions of the first guide line area and the second guide line area relative to the target vehicle are both right rearward, and a first spacing distance between the first guide line area and the target vehicle is less than a second spacing distance between the second guide line area and the target vehicle, determining that the target moving path is a left road of the first guide line area; In response to determining that the first guide line area moves to the left rear relative to the target vehicle and the second guide line area moves to the right rear relative to the target vehicle, the target moving path is determined to be a middle road between the first guide line area and the second guide line area.

9. The method according to claim 1, wherein: In response to triggering the line pressing detection, generating a first line pressing detection result and a second line pressing detection result based on a plurality of vehicle front images of the target vehicle, including: In response to triggering the line-crossing detection, a first line-crossing detection result and a second line-crossing detection result are generated based on a plurality of vehicle front images of the target vehicle collected by the driving recorder.

10. The method according to claim 1, further comprising: In response to determining that the target vehicle has not crossed the line, generating a predicted trajectory based on a current moving path of the target vehicle and vehicle control information; Based on the predicted trajectory, determine the predicted distance between the target lane line and the target vehicle; Determining the probability of the target vehicle crossing the line based on the predicted distance and the half body width of the target vehicle; In response to determining that the probability of crossing the line is greater than the probability threshold, outputting an impending crossing line warning message.

11. The method according to any one of claims 1 to 10, further comprising: In response to determining that the target vehicle crosses the line and the length of time the vehicle crosses the line exceeds a preset length of time, outputting a line-crossing warning message.

12. A vehicle crossing line recognition device, comprising: A line-crossing triggering module is configured to determine whether to trigger line-crossing detection based on driving information of the target vehicle, wherein the driving information includes driving position information and / or driving environment information; A line-crossing detection module is configured to generate a first line-crossing detection result and a second line-crossing detection result based on a plurality of vehicle front images of the target vehicle in response to triggering line-crossing detection; The line-crossing generation module is configured to determine whether the target vehicle crosses the line based on the first line-crossing detection result and the second line-crossing detection result.

13. The device according to claim 12, wherein: The wire pressing detection module further comprises: A first detection unit is configured to input a plurality of vehicle front images of the target vehicle into a pre-trained line-crossing detection model to obtain the first line-crossing detection result; a distance determination unit configured to determine a target lane line and a target distance between the target lane line and a target vehicle based on the plurality of vehicle front images; The second detection unit is configured to generate the second lane crossing detection result based on the target distance between the target lane line and the target vehicle.

14. The device according to claim 13, wherein: The distance determination unit is further configured to: Determine the map lane information of the current driving position; Based on the map lane information and the plurality of vehicle front images, a target lane line and a target distance between the target lane line and a target vehicle are determined.

15. The device according to claim 13, wherein: The distance determination unit is further configured to: Using an inverse perspective transformation, converting the plurality of vehicle front images into a sequence of top-view images; Determine a target distance between the target lane line and the target vehicle based on the position information of the target lane line and the top view sequence; and The second detection unit is further configured to: The second line pressing detection result is determined based on the target distance and the half body width of the target vehicle.

16. The device according to claim 13, wherein: The distance determination unit is further configured to: In response to determining that a guide line area exists in the plurality of vehicle front images of the target vehicle, determining a moving direction of the guide line area relative to the target vehicle; Determining a target moving path of the target vehicle at a fork in the guide line area based on the moving direction of the guide line area relative to the target vehicle; Determine a region boundary line in the guide line region corresponding to the target moving path as a target lane line; A target distance between a target lane line and a target vehicle is determined based on the plurality of vehicle front images.

17. The device according to claim 16, wherein: The step of determining a target moving path of the target vehicle at a fork in the guide line area corresponding to the guide line area based on the moving direction of the target vehicle relative to the guide line area includes: In response to determining that there are a plurality of guide line areas, determining moving directions of the plurality of guide line areas relative to the target vehicle and interval distances between the plurality of guide line areas and the target vehicle; Based on the moving directions and the interval distances of the plurality of guide line areas, a target moving path of the target vehicle at the fork in the road is determined.

18. The device according to claim 17, wherein: The step of determining a target moving path of the target vehicle at the fork in the road based on the moving directions and the interval distances of the plurality of guide line sub-areas comprises: In response to the moving directions of the plurality of guide line areas being the same, determining a target area closest to the target vehicle based on the interval distances of the plurality of guide line areas; Based on the moving direction of the target area relative to the target vehicle, a target moving path of the target vehicle at the fork in the road is determined.

19. The device according to claim 16, wherein: There are multiple guide line areas, the multiple guide line areas include a first guide line area and a second guide line area, and the first guide line area and the second guide line area have different guiding directions; The step of determining a target moving path of the target vehicle at a fork in the guide line area corresponding to the guide line area based on the moving direction of the target vehicle relative to the guide line area includes: In response to determining that the moving directions of the first guide line area and the second guide line area relative to the target vehicle are both left rear, and a first spacing distance between the first guide line area and the target vehicle is greater than a second spacing distance between the second guide line area and the target vehicle, determining that the target moving path is a right side road of the second guide line sub-area; In response to determining that the moving directions of the first guide line area and the second guide line area relative to the target vehicle are both right rearward, and a first spacing distance between the first guide line area and the target vehicle is less than a second spacing distance between the second guide line area and the target vehicle, determining that the target moving path is a left road of the first guide line area; In response to determining that the first guide line area moves to the left rear relative to the target vehicle and the second guide line area moves to the right rear relative to the target vehicle, the target moving path is determined to be a middle road between the first guide line area and the second guide line area.

20. The device according to claim 12, wherein: The wire pressing detection module is further configured to: In response to triggering the line-crossing detection, a first line-crossing detection result and a second line-crossing detection result are generated based on a plurality of vehicle front images of the target vehicle collected by the driving recorder.

21. The apparatus according to claim 12, further comprising: A line crossing warning module is configured to generate a predicted trajectory based on a current moving path of the target vehicle and vehicle control information in response to determining that the target vehicle has not crossed the line; Based on the predicted trajectory, the predicted distance between the target lane line and the target vehicle is determined; based on the predicted distance and the size of half the body width of the target vehicle, the probability of the target vehicle crossing the line is determined; in response to determining that the probability of crossing the line is greater than the probability threshold, an alarm message indicating that the line is about to be crossed is output.

22. The apparatus according to claim 12, further comprising: The line crossing warning module is configured to output line crossing warning information in response to determining that the target vehicle crosses the line and the length of time the line is crossed exceeds a preset length of time.

23. An electronic device, characterized in that: include: at least one processor; as well as a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method according to any one of claims 1 to 11.

24. A non-transitory computer-readable storage medium storing computer instructions, characterized in that: The computer instructions are used to cause the computer to execute the method according to any one of claims 1 to 11.

25. A computer program product comprising a computer program, which, when executed by a processor, implements the method according to any one of claims 1 to 11.

Images