A vehicle driving control method and device for zebra crossing road scenarios

Abstract

This invention provides a vehicle driving control method and device for zebra crossing scenarios. During vehicle operation, it monitors road information in real time and determines whether a zebra crossing exists ahead of the vehicle. If so, it acquires the distance between the vehicle and the zebra crossing and determines if the distance is below a preset distance. If so, it identifies other vehicles in other lanes ahead of the vehicle and, based on these other vehicles, determines the target vehicle closest to the zebra crossing in each of the other lanes. It acquires vehicle information for both the current vehicle and the target vehicle, controls a camera to capture images of the zebra crossing, and determines whether pedestrians are present in the zebra crossing based on the images. If pedestrians are present, it intelligently controls the vehicle to slow down until it stops before the zebra crossing, thus preventing traffic accidents caused by pedestrians suddenly darting out from in front of adjacent vehicles.

Description

Technical Field

[0001] This invention belongs to the technical field of vehicles, specifically relating to a vehicle driving control method and device for zebra crossing road scenarios. Background Technology

[0002] With the development of society, people have higher and higher requirements for the intelligence of automobiles. More and more driverless cars are operating on specific roads. Among them, most driverless cars are equipped with monitoring equipment such as video cameras, radar sensors, laser rangefinders and positioning devices. Through these monitoring devices, some traffic accidents can be prevented and avoided. However, the testing accuracy and reaction time of the monitoring equipment are limited. If an emergency occurs, it is difficult for driverless cars to deal with it in a short time.

[0003] It is important to note that countless traffic accidents occur at crosswalks. Many of these accidents are related not only to vehicle speed but also to pedestrian behavior, especially when vehicles are crossing crosswalks without traffic lights. If a pedestrian suddenly darts out from in front of an adjacent vehicle, even though autonomous vehicles are equipped with video cameras and radar sensors, the field of view of the vehicle's video camera is obstructed by adjacent vehicles, making it impossible to predict the pedestrian's sudden appearance. Furthermore, due to the presence of adjacent vehicles, the radar sensors, which have long-range detection capabilities, can only detect nearby vehicles. Additionally, although autonomous vehicles will immediately brake after the radar sensors detect a pedestrian, they cannot stop the vehicle immediately at a certain speed. Therefore, significant safety hazards remain. Thus, there is an urgent need for a vehicle driving control method and system for crosswalk scenarios to avoid these problems. Summary of the Invention

[0004] Based on this, the present invention provides a vehicle driving control method and device for zebra crossing road scenarios, which aims to take measures in advance to eliminate safety hazards in the event that pedestrians may suddenly dart out from in front of adjacent vehicles.

[0005] A first aspect of this invention provides a vehicle driving control method for a zebra crossing road scenario, applied to a zebra crossing road scenario system equipped with an image acquisition device. The image acquisition device includes a plurality of photographing devices, each of which is sequentially installed above the zebra crossing road. Each photographing device is used to capture an image of the zebra crossing road within a preset range. The method includes:

[0006] During vehicle operation, road information is monitored in real time, and the presence of a zebra crossing is determined based on this information.

[0007] If so, the distance between the current vehicle and the zebra crossing is obtained, and it is determined whether the distance is less than a preset distance;

[0008] If so, then identify other vehicles in front of the current vehicle in other lanes, and based on the other vehicles, identify the target vehicle closest to the zebra crossing in each of the other lanes;

[0009] The system acquires vehicle information of the current vehicle and the target vehicle, controls the camera to capture road images of the zebra crossing, and determines whether there are pedestrians on the zebra crossing based on the road images.

[0010] If so, then control the current vehicle to decelerate.

[0011] Furthermore, the step of obtaining the distance between the vehicle and the zebra crossing and determining whether the distance is less than a preset distance includes:

[0012] The first position coordinates of the current vehicle and the target position coordinates of the zebra crossing are obtained, and the distance is calculated based on the first position coordinates and the target position coordinates.

[0013] Furthermore, the steps of acquiring vehicle information of the other vehicles and the target vehicle, controlling the camera to capture road images of the zebra crossing, and determining whether there are pedestrians at the zebra crossing based on the road images include:

[0014] Obtain vehicle information of the current vehicle and the target vehicle, wherein the vehicle information includes at least vehicle speed sub-information, wherein the vehicle speed sub-information includes a first vehicle speed value of the current vehicle and a second vehicle speed value of the target vehicle;

[0015] Based on the first vehicle speed value and the second vehicle speed value, determine whether the first vehicle speed value is greater than the second vehicle speed value;

[0016] If so, the steps are executed: controlling the camera to capture a road image of the zebra crossing, and determining whether there are pedestrians on the zebra crossing based on the road image.

[0017] Furthermore, the steps of acquiring vehicle information of the other vehicles and the target vehicle, controlling the camera to capture road images of the zebra crossing, and determining whether there are pedestrians at the zebra crossing based on the road images also include:

[0018] The road image is acquired and input into an image processing model to obtain each feature region and its corresponding label.

[0019] Determine whether the target identifier exists among the aforementioned identifiers;

[0020] If so, it means there are pedestrians on the zebra crossing, and the step of controlling the current vehicle to slow down is executed.

[0021] Furthermore, before the step of acquiring the road image, inputting the road image into the image processing model, and obtaining the feature identifiers, the following steps are included:

[0022] Historical road images are acquired and preprocessed to obtain preprocessed images;

[0023] The feature regions in the preprocessed image are identified by manual annotation to obtain a training set, wherein the feature regions include at least the human head region, the human shoulder region, and the zebra crossing region.

[0024] The training set is used to train the neural network model to establish the image processing model.

[0025] Furthermore, if the presence of pedestrians at the zebra crossing is determined based on the road image, the step of controlling the current vehicle to decelerate includes:

[0026] The image acquisition device is controlled to continuously capture multiple images of the target road, and the multiple images of the target road are input into the image processing model to obtain the target human head region and the target zebra crossing region in each of the target road images;

[0027] Based on the target human head region and the zebra crossing region, determine the target zebra crossing sub-region that overlaps with the target human head region within the target zebra crossing region;

[0028] Based on the continuously acquired target zebra crossing sub-regions, the pedestrian's movement direction is determined, and it is determined whether the movement direction is away from the lane where the vehicle is located.

[0029] If not, then control the current vehicle to decelerate.

[0030] Furthermore, the step of determining the pedestrian's movement direction based on the continuously acquired target zebra crossing sub-regions, and determining whether the movement direction is away from the lane where the vehicle is located, includes:

[0031] Obtain the first target zebra crossing area at the current time and the second target zebra crossing area within a preset time. The first target zebra crossing area includes at least the first leftmost target zebra crossing sub-area and the first rightmost target zebra crossing sub-area that overlap with the head area of ​​the target human body. The second target zebra crossing area includes at least the second leftmost target zebra crossing sub-area and the second rightmost target zebra crossing sub-area that overlap with the head area of ​​the target human body.

[0032] Determine whether the second leftmost target zebra crossing sub-region is on the same side as the first leftmost target zebra crossing sub-region, and whether the second rightmost target zebra crossing sub-region is on the same side as the first rightmost target zebra crossing sub-region;

[0033] If so, the direction of movement is determined based on the lateral direction.

[0034] A second aspect of the present invention provides a vehicle driving control device for zebra crossing road scenarios, the device comprising:

[0035] The first judgment module is used to monitor road information in real time during vehicle operation and determine whether there is a zebra crossing ahead of the vehicle based on the road information.

[0036] The second judgment module is used to obtain the distance between the current vehicle and the zebra crossing when it is determined that there is a zebra crossing road in front of the current vehicle, and to determine whether the distance is less than a preset distance.

[0037] The target vehicle determination module is used to determine other vehicles in other lanes in front of the current vehicle when the distance is determined to be lower than a preset distance, and to determine the target vehicle in each of the other lanes that is closest to the zebra crossing based on the other vehicles.

[0038] The third judgment module is used to obtain vehicle information of the current vehicle and the target vehicle, control the camera to capture road images of the zebra crossing, and determine whether there are pedestrians on the zebra crossing based on the road images.

[0039] The control module is used to control the current vehicle to decelerate when it determines that there are pedestrians on the zebra crossing based on the road image.

[0040] A third aspect of the present invention provides a readable storage medium, comprising:

[0041] The readable storage medium stores one or more programs that, when executed by a processor, implement the vehicle driving control method described above for zebra crossing road scenarios.

[0042] A fourth aspect of the present invention provides a vehicle, characterized in that the vehicle includes a memory and a processor, wherein:

[0043] The memory is used to store computer programs;

[0044] When the processor executes the computer program stored in the memory, it implements the above-described vehicle driving control method for zebra crossing road scenarios.

[0045] In summary, this embodiment of the invention monitors road information in real time during vehicle operation and determines whether a zebra crossing exists ahead of the vehicle based on the road information. If so, it obtains the distance between the current vehicle and the zebra crossing and determines whether the distance is lower than a preset distance. If so, it identifies other vehicles in other lanes ahead of the current vehicle and, based on these other vehicles, identifies the target vehicle closest to the zebra crossing in each of the other lanes. It obtains vehicle information for both the current vehicle and the target vehicle, controls a camera to capture images of the zebra crossing, and determines whether pedestrians exist at the zebra crossing based on the images. If so, it intelligently controls the current vehicle to slow down until it stops before the zebra crossing, thus avoiding traffic accidents caused by pedestrians suddenly darting out from in front of adjacent vehicles. Attached Figure Description

[0046] Figure 1 This is a flowchart illustrating the implementation of a vehicle driving control method for a zebra crossing road scenario provided in the first embodiment of the present invention.

[0047] Figure 2 This is a flowchart illustrating the implementation of a vehicle driving control device for a zebra crossing road scenario, provided in the second embodiment of the present invention.

[0048] Figure 3 This is a structural block diagram of a vehicle provided in the third embodiment of the present invention.

[0049] The following detailed embodiments will be further described in conjunction with the above-mentioned accompanying drawings. Detailed Implementation

[0050] To facilitate understanding of the present invention, a more complete description will be given below with reference to the accompanying drawings. Several embodiments of the invention are illustrated in the drawings. However, the invention can be implemented in many different forms and is not limited to the embodiments described herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete.

[0051] It should be noted that when a component is said to be "fixed to" another component, it can be directly on the other component or there may be an intervening component. When a component is said to be "connected to" another component, it can be directly connected to the other component or there may be an intervening component. The terms "vertical," "horizontal," "left," "right," and similar expressions used in this document are for illustrative purposes only.

[0052] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. The terminology used herein in the description of the invention is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0053] Example 1

[0054] Please see Figure 1 , Figure 1 The present invention illustrates a vehicle driving control method for a zebra crossing road scenario provided by a first embodiment of the present invention, the method specifically including steps S01 to S05.

[0055] Step S01: During the vehicle's journey, monitor road information in real time and determine whether there is a zebra crossing ahead of the vehicle based on the road information. If so, proceed to step S02.

[0056] During operation, autonomous vehicles, lacking human intervention, require real-time monitoring of road information to analyze road conditions. Specifically, this involves configuring road information recognition modules, such as: a visual perception module (primarily consisting of a high-definition camera module mounted on the front of the vehicle or behind the windshield), which captures road images and uploads them to a road analysis and calculation module for real-time online identification of lane lines, traffic signs, target vehicles, and pedestrians; a lidar perception module (primarily consisting of lidar units mounted around the vehicle, communicating with the road analysis and calculation module via gigabit Ethernet), which measures the distance to target vehicles within the area and can also avoid obstacles or brake; and a positioning module (primarily consisting of dual antennas mounted on the roof and an inertial navigation system integrated into the rear axle), which receives differential centimeter-level positioning information and high-precision heading measurements outdoors in real time.

[0057] Specifically, the system can capture road images using a visual perception module and analyze whether the images contain a target reference object. This target reference object can be the black and white stripes of a zebra crossing. If the target reference object is identified, it indicates that there is a zebra crossing ahead of the vehicle. This is an example, not a limitation. The roads where autonomous vehicles travel are usually special roads where traffic signs can be installed to indicate the distance to the zebra crossing ahead. When the road image captured by the visual perception module contains such a traffic sign, it is identified to determine whether there is a zebra crossing ahead of the vehicle and the distance between the vehicle and the zebra crossing.

[0058] In addition, since the vehicle driving control method is applied to the zebra crossing road scene system equipped with image acquisition devices, the image acquisition devices include several photographing devices, and each photographing device is installed in sequence directly above the zebra crossing road. Each photographing device is used to capture images of the zebra crossing road within a preset range. The position coordinates of the zebra crossing road and the corresponding photographing device can also be pre-established. The photographing device can be a camera, video camera, etc., and the position coordinates of the zebra crossing road can be consistent with the position coordinates of the photographing device.

[0059] Step S02: Obtain the distance between the current vehicle and the zebra crossing road, and determine whether the distance is less than a preset distance. If so, proceed to step S03.

[0060] In this embodiment, since the location of the zebra crossing is fixed, the corresponding coordinates are also fixed, i.e., the target coordinates. These target coordinates can be pre-input into the positioning module of the autonomous vehicle. During subsequent driving, the positioning module obtains the vehicle's first position coordinates in real time and calculates the distance between the vehicle and the preset target coordinates on the road to be traversed. This determines the first distance between the vehicle and the zebra crossing. The first distance is compared with the preset distance. If the first distance is less than the preset distance, it indicates that the autonomous vehicle has entered the radiation range of the zebra crossing and should yield to pedestrians when crossing. Specifically, the preset distance can be set according to the speed limit of the zebra crossing section. The higher the speed limit, the larger the preset distance; the lower the speed limit, the smaller the preset distance. In other words, the faster the vehicle speed, the greater the reaction distance that needs to be reserved.

[0061] Step S03 involves identifying other vehicles in the other lanes of the current vehicle that are in front of the current vehicle, and based on these other vehicles, identifying the target vehicle in each of the other lanes that is closest to the zebra crossing.

[0062] Specifically, the second position coordinates of other vehicles are obtained, and the second distance is calculated based on the second position coordinates and the target position coordinates. Then, it is determined whether the second distance is less than the first distance. If so, it means that other vehicles are in front of the vehicle, that is, the distance between other vehicles and the zebra crossing is closer than the distance between the current vehicle and the zebra crossing. In this case, the other vehicle with the smallest second distance in each lane is marked as the target vehicle.

[0063] Step S04: Obtain vehicle information of the current vehicle and the target vehicle, control the camera to capture road images of the zebra crossing, and determine whether there are pedestrians on the zebra crossing based on the road images. If so, proceed to step S05.

[0064] In order to more intelligently control the camera to capture images of the zebra crossing, instead of keeping the camera in constant shooting mode and wasting resources, this embodiment first acquires vehicle information of the current vehicle and the target vehicle. The vehicle information includes at least speed sub-information, which includes the first speed value of the current vehicle and the second speed value of the target vehicle. Then, based on the first and second speed values, it is determined whether the first speed value is greater than the second speed value. If the first speed value is greater than the second speed value, it means that the current vehicle is about to overtake the target vehicle and has no intention of braking. If a pedestrian suddenly runs out from the adjacent lane at this time, and the current vehicle then brakes, the deceleration effect may not be ideal, posing a significant safety hazard. In this case, controlling the camera to capture images of the zebra crossing is of great significance. In addition, if it is known in advance that there are pedestrians on the zebra crossing, the current vehicle can be controlled to decelerate at a constant speed, which not only ensures the safety of pedestrians but also improves the driving experience of passengers in the vehicle.

[0065] Furthermore, it is necessary to identify the road images of zebra crossings captured by the camera to determine whether pedestrians are present. Specifically, an image processing model should first be established. The establishment process can involve acquiring historical road images and preprocessing them to obtain a preprocessed image. This preprocessed image is obtained by sequentially performing grayscale and binarization on the historical road images. The grayscale processing formula is as follows:

[0066] I(x,y)=0.31I R (x, y) + 0.6I C (x, y) + 0.11I B (x, y);

[0067] Where I(x, y) represents the intensity value of the image at coordinates (x, y), I R (x, y) represents the intensity value of the red channel in the image at coordinates (x, y). G (x, y) represents the intensity value of the green channel in the image at coordinates (x, y). B (x, y) represents the intensity value of the blue channel in the image at coordinates (x, y).

[0068] Furthermore, feature regions in the preprocessed image are manually labeled to obtain a training set. These feature regions include at least the human head region, the human shoulder region, and the zebra crossing region. Finally, the training set is used to train a neural network model to establish an image processing model. It should be noted that pedestrians can be identified through the features and combinations of the human head and shoulder regions. The human head region is approximately circular, and the human shoulder region is rectangular. Specifically, based on the circular shape of the human head region, the human shoulder region can be approximately symmetrically divided into two rectangles. This is the relationship between the human head and shoulder regions. By identifying the human head and shoulder regions and their relationship, it is possible to determine whether objects with the same shape on the zebra crossing are pedestrians, effectively improving the pedestrian recognition rate and avoiding misjudgments (e.g., the obstruction of the zebra crossing by objects other than pedestrians). At the same time, the zebra crossing pattern has a certain degree of recognizability, consisting of equidistant white stripes, facilitating the identification and recognition of the zebra crossing area.

[0069] In this embodiment, position attention and channel attention mechanisms are introduced into the image processing model to more quickly and accurately focus on the feature regions to be identified. Position attention (PAM) effectively enhances feature representation capabilities by establishing rich contextual links on local features and representing contextual information from a broader perspective; the deeper the feature channel graph in the network model, the faster its response to a specific category. The channel attention module explicitly models the dependencies between channels, highlighting interdependent feature maps, improving the feature representation of specific semantics, and enhancing the overall recognition of the same category within the network model. Specifically, position attention in the image processing model can be represented as follows:

[0070]

[0071]

[0072] Among them, S ji Let A represent the predicted probability map of the (j, i)th grid point. j B represents the feature map of size C×H×W in the j-th row. i Indicates that by A j After transformation, a feature map of the i-th column with size N×C is obtained, where C... j Indicates that by A j After transformation, the feature map of the j-th row with size C×N is obtained, D. i Indicates that by A jAfter transformation, a feature map of size C×N is obtained for the i-th column. E represents the position feature map, α represents the position feature parameter, C represents the number of paths in the feature map, H represents the height of the feature map, W represents the length of the feature map, and N represents the total number of columns of pixels in the feature map.

[0073] Furthermore, the channel attention mechanism can be represented in image processing models as follows:

[0074]

[0075]

[0076] Among them, X ji A represents the weight value of the (j, i)th channel. j B represents the feature map of size C×H×W in the j-th row. i Indicates that by A j After transformation, a feature map of the i-th column with size N×C is obtained, where C... j Indicates that by A j After transformation, the feature map of the j-th row with size C×N is obtained, D. i Indicates that by A j After transformation, a feature map of size C×N is obtained for the i-th column, where F represents the channel feature map, β represents the channel feature parameter, C represents the number of channels in the feature map, H represents the height of the feature map, and W represents the length of the feature map.

[0077] Once the image processing model is established, during actual operation, the acquired road images are input into the image processing model to obtain each feature region and its corresponding label, namely the human head region, the human shoulder region, and the zebra crossing region. The corresponding labels can be pedestrian labels and zebra crossing labels. Then, it is determined whether there is a target label among the labels, that is, whether there is a pedestrian label. If so, it means that there is a pedestrian on the zebra crossing road, and the current vehicle needs to be slowed down.

[0078] Specifically, in order to more intelligently control vehicle deceleration, multiple images of the target road can be continuously captured by the image acquisition device. The direction of pedestrian movement on the zebra crossing can be analyzed using multiple images of the target road. As an example, and not a limitation, in some other optional embodiments, if it is determined that there are no pedestrians on the zebra crossing, the lidar perception module of the autonomous vehicle can be controlled to perform obstacle testing on the road ahead to ensure that there are no pedestrians on the zebra crossing in front of the target vehicle.

[0079] In this embodiment, the image acquisition device continuously captures multiple images of the target road and inputs them into the image processing model to obtain the target human head region and the target zebra crossing region in each target road image. Then, based on the target human head region and the zebra crossing region, the target zebra crossing sub-region that overlaps with the target human head region is determined. Finally, based on the continuously acquired target zebra crossing sub-regions, the pedestrian's movement direction is determined, and it is determined whether the movement direction is away from the vehicle's lane. If not, the current vehicle is controlled to decelerate.

[0080] It should be noted that the target zebra crossing sub-region can be the target zebra crossing stripe. The number of zebra crossing stripes between the target zebra crossing stripe and the lane currently occupied by the vehicle is determined. Simultaneously, the zebra crossing stripe width and spacing are acquired. Based on the number of zebra crossing stripes, the waiting distance between the target zebra crossing stripe and the lane currently occupied by the vehicle is determined. Based on the waiting distance and the distance between the vehicle and the zebra crossing, the vehicle is controlled to decelerate. Taking white zebra crossing stripes as an example, the white stripes can be numbered, and the zebra crossing stripe width and spacing can be manually measured beforehand. When the road image is recognized, it is determined that... When there are pedestrians on a zebra crossing, their exact location can be determined by the white stripes on the zebra crossing, allowing for better vehicle control. Specifically, the braking ability of the vehicle can be controlled based on the distance between the pedestrian and the vehicle's lane, as well as the distance between the vehicle and the zebra crossing. When the distance between the vehicle and the zebra crossing is the same, the greater the distance between the pedestrian and the vehicle's lane, the less braking the vehicle should be, i.e., the vehicle should decelerate slowly. Conversely, the smaller the distance between the pedestrian and the vehicle's lane, the more braking the vehicle should be, i.e., the vehicle should decelerate quickly.

[0081] Furthermore, based on the continuously acquired target zebra crossing sub-regions, the movement direction of pedestrians is determined, and it is judged whether the movement direction is away from the vehicle's lane. If not, the current vehicle is controlled to slow down. Specifically, since the image acquisition device continuously captures multiple images of the target road, one road image can be acquired at preset intervals and input into the image processing model. This allows for real-time monitoring of pedestrian movement on the zebra crossing, providing feedback to following vehicles. The purpose of this feedback is to prevent traffic accidents caused by pedestrians suddenly darting out from in front of adjacent vehicles. It should be noted that the image acquisition device includes several photographic devices. Each camera is installed sequentially above the zebra crossing. Each camera captures an image of the zebra crossing within a preset area. For example, if there are 15 white stripes in the zebra crossing, three cameras can be installed above it, each covering an area with five white stripes. The cameras are pre-numbered according to the order of the five white stripe zebra crossing areas. During normal operation, the three cameras capture images simultaneously, and the images are stitched together to form the target road image. The final target road image is then identified.

[0082] Specifically, when a pedestrian is moving away from the vehicle's lane, it indicates the pedestrian will not appear in front of the vehicle, allowing the vehicle to slow down and pass without stopping at the crosswalk. This makes traffic management more orderly and smooth. Specifically, it acquires the first target crosswalk area at the current moment and the second target crosswalk area within a preset time. The first target crosswalk area includes at least the first leftmost and first rightmost target crosswalk sub-areas overlapping the head area of ​​the target person. The second target crosswalk area includes at least the second leftmost and second rightmost target crosswalk sub-areas overlapping the head area of ​​the target person. In the zebra crossing sub-area, based on the continuously acquired target zebra crossing areas, the direction of pedestrian movement is determined, and it is determined whether the direction of movement is away from the vehicle's lane. More specifically, it is determined whether the second leftmost target zebra crossing sub-area is on the same side as the first leftmost target zebra crossing sub-area, and whether the second rightmost target zebra crossing sub-area is on the same side as the first rightmost target zebra crossing sub-area. If so, the direction of movement is determined based on the side direction, and it is determined whether the direction of movement is away from the vehicle's lane. If not, it means that the pedestrian is moving towards the vehicle's lane, so the vehicle is controlled to slow down until it stops in front of the zebra crossing.

[0083] Understandably, if the pedestrians are a group, the group will obscure the zebra crossing, resulting in an image where the zebra crossing is obscured by the crowd. This allows us to identify the sub-region of the zebra crossing that overlaps with the head area of ​​the pedestrian within the target zebra crossing area. Specifically, when the crowd moves to the left, the second leftmost zebra crossing stripe in the images taken before and after will be to the left of the first leftmost zebra crossing stripe, and the second rightmost zebra crossing stripe will be to the left of the first rightmost zebra crossing stripe. Similarly, when the crowd moves to the right, the second leftmost zebra crossing stripe in the images taken before and after will be to the right of the first leftmost zebra crossing stripe, and the second rightmost zebra crossing stripe will be to the right of the first rightmost zebra crossing stripe. If the pedestrian is a single person, only one zebra crossing stripe in the first target zebra crossing area can be identified, and its position can be compared with the other zebra crossing stripe identified in the second target zebra crossing area to determine the pedestrian's direction of movement. It should be noted that if the target zebra crossing area in the target zebra crossing road pattern obtained within the preset time is obscured by pedestrians, and the direction of pedestrian movement cannot be determined by the above method, then the vehicle can be slowed down according to the requirement that there are pedestrians on the zebra crossing, and controlled until it stops in front of the zebra crossing.

[0084] In step S05, the current vehicle is controlled to decelerate.

[0085] When it is determined that there are pedestrians on the zebra crossing, the vehicle will slow down until it stops upon reaching the zebra crossing.

[0086] In summary, this embodiment of the invention monitors road information in real time during vehicle operation and determines whether a zebra crossing exists ahead of the vehicle based on the road information. If so, it obtains the distance between the current vehicle and the zebra crossing and determines whether the distance is lower than a preset distance. If so, it identifies other vehicles in other lanes ahead of the current vehicle and, based on these other vehicles, identifies the target vehicle closest to the zebra crossing in each of the other lanes. It obtains vehicle information for both the current vehicle and the target vehicle, controls a camera to capture images of the zebra crossing, and determines whether pedestrians exist at the zebra crossing based on the images. If so, it intelligently controls the current vehicle to slow down until it stops before the zebra crossing, thus avoiding traffic accidents caused by pedestrians suddenly darting out from in front of adjacent vehicles.

[0087] Example 2

[0088] Please see Figure 2 , Figure 2 This is a structural block diagram of a vehicle driving control device for a zebra crossing road scenario provided in an embodiment of the present invention. The vehicle driving control device 200 for a zebra crossing road scenario includes: a first judgment module 21, a second judgment module 22, a target vehicle determination module 23, a third judgment module 24, and a control module 25, wherein:

[0089] The first judgment module 21 is used to monitor road information in real time during vehicle operation and determine whether there is a zebra crossing ahead of the vehicle based on the road information.

[0090] The second judgment module 22 is used to obtain the distance between the current vehicle and the zebra crossing when it is determined that there is a zebra crossing road in front of the current vehicle, and to determine whether the distance is less than a preset distance.

[0091] The target vehicle determination module 23 is used to determine other vehicles in other lanes in front of the current vehicle when the distance is determined to be lower than a preset distance, and to determine the target vehicle in each of the other lanes that is closest to the zebra crossing based on the other vehicles.

[0092] The third judgment module 24 is used to obtain vehicle information of the current vehicle and the target vehicle, control the camera to take a picture of the zebra crossing road, and determine whether there are pedestrians on the zebra crossing road based on the road image;

[0093] The control module 25 is used to control the current vehicle to decelerate when it is determined from the road image that there are pedestrians on the zebra crossing.

[0094] Furthermore, in some optional embodiments of the present invention, the second determination module 22 includes:

[0095] The first distance calculation unit is used to obtain the first position coordinates of the current vehicle and the target position coordinates of the zebra crossing road, and calculate the distance based on the first position coordinates and the target position coordinates.

[0096] Furthermore, in some optional embodiments of the present invention, the third determination module 24:

[0097] The first acquisition unit is used to acquire vehicle information of the current vehicle and the target vehicle, wherein the vehicle information includes at least vehicle speed sub-information, wherein the vehicle speed sub-information includes a first vehicle speed value of the current vehicle and a second vehicle speed value of the target vehicle.

[0098] The first judgment unit is used to determine whether the first vehicle speed value is greater than the second vehicle speed value based on the first vehicle speed value and the second vehicle speed value.

[0099] Furthermore, in some optional embodiments of the present invention, the third determination module 24 further includes:

[0100] The second acquisition unit is used to acquire the road image, input the road image into the image processing model, and obtain each feature region and its corresponding identifier.

[0101] The second judgment unit is used to determine whether a target identifier exists among the identifiers.

[0102] Furthermore, in some optional embodiments of the present invention, the vehicle driving control device 200 for zebra crossing road scenarios includes:

[0103] The preprocessing module is used to acquire historical road images and preprocess the historical road images to obtain preprocessed images;

[0104] The labeling module is used to label the feature regions in the preprocessed image by manual annotation to obtain a training set, wherein the feature regions include at least the human head region, the human shoulder region, and the zebra crossing region.

[0105] The training module is used to train the neural network model with the training set to establish the image processing model.

[0106] Furthermore, in some optional embodiments of the present invention, the control module 25 includes:

[0107] The input unit is used to control the image acquisition device to continuously capture multiple target road images and input the multiple target road images into the image processing model to obtain the target human head region and the target zebra crossing region in each target road image;

[0108] The determining unit is configured to determine, based on the target human head region and the zebra crossing region, a target zebra crossing sub-region that overlaps with the target human head region;

[0109] The third judgment unit is used to determine the pedestrian's movement direction based on the continuously acquired target zebra crossing sub-region, and to determine whether the movement direction is away from the lane where the vehicle is located.

[0110] Furthermore, in some optional embodiments of the present invention, the third determining unit includes:

[0111] The target zebra crossing area acquisition subunit is used to acquire the first target zebra crossing area at the current time and the second target zebra crossing area within a preset time. The first target zebra crossing area includes at least the first leftmost target zebra crossing sub-area and the first rightmost target zebra crossing sub-area that overlap with the head area of ​​the target human body. The second target zebra crossing area includes at least the second leftmost target zebra crossing sub-area and the second rightmost target zebra crossing sub-area that overlap with the head area of ​​the target human body.

[0112] The first judgment subunit is used to determine whether the second leftmost target zebra crossing sub-region is on the same side as the first leftmost target zebra crossing sub-region, and whether the second rightmost target zebra crossing sub-region is on the same side as the first rightmost target zebra crossing sub-region.

[0113] The movement direction determination subunit is used to determine the movement direction based on the lateral direction when it is determined that the second leftmost target zebra crossing sub-region is in the same lateral direction as the first leftmost target zebra crossing sub-region and the second rightmost target zebra crossing sub-region is in the same lateral direction as the first rightmost target zebra crossing sub-region.

[0114] Example 3

[0115] In another aspect, the present invention also proposes a vehicle, see [link to relevant documentation]. Figure 3 The diagram shows a structural block diagram of a vehicle in the third embodiment of the present invention, including a memory 20, a processor 10, and a computer program 30 stored in the memory and executable on the processor. When the processor 10 executes the computer program 30, it implements the vehicle driving control method for zebra crossing road scenarios as described above.

[0116] In some embodiments, the processor 10 may be a central processing unit (CPU), controller, microcontroller, microprocessor or other data processing chip, used to run program code stored in memory 20 or process data, such as executing access restriction programs.

[0117] The memory 20 includes at least one type of readable storage medium, including flash memory, hard disk, multimedia card, card-type memory (e.g., SD or DX memory), magnetic memory, magnetic disk, optical disk, etc. In some embodiments, the memory 20 may be an internal storage unit of the vehicle, such as the vehicle's hard disk. In other embodiments, the memory 20 may be an external storage device of the vehicle, such as a plug-in hard disk, SmartMedia Card (SMC), Secure Digital (SD) card, Flash Card, etc. Furthermore, the memory 20 may include both internal and external storage devices. The memory 20 can be used not only to store vehicle application software and various types of data, but also to temporarily store data that has been output or will be output.

[0118] It should be pointed out that, Figure 3 The structure shown does not constitute a limitation on the vehicle. In other embodiments, the vehicle may include fewer or more components than shown, or combine certain components, or have different component arrangements.

[0119] This invention also proposes a computer-readable storage medium storing a computer program that, when executed by a processor, implements the vehicle driving control method described above for zebra crossing road scenarios.

[0120] Those skilled in the art will understand that the logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequential list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-included system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can mean any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device.

[0121] More specific examples of computer-readable media (a non-exhaustive list) include: electrical connections (electronic devices) having one or more wires, portable computer disk drives (magnetic devices), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Furthermore, computer-readable media can even be paper or other suitable media on which programs can be printed, because programs can be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in computer memory.

[0122] It should be understood that various parts of the present invention can be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions for data states, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0123] In the description of this specification, references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0124] The above embodiments merely illustrate several implementation methods of the present invention, and their descriptions are relatively specific and detailed, but they should not be construed as limiting the scope of the present invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these all fall within the protection scope of the present invention. Therefore, the protection scope of this patent should be determined by the appended claims.

Claims

1. A vehicle driving control method for zebra crossing scenarios, applied to a zebra crossing scenario system equipped with an image acquisition device, wherein the image acquisition device includes a plurality of photographing devices, and each of the photographing devices is sequentially installed above the zebra crossing, each of the photographing devices being used to capture an image of the zebra crossing within a preset range, characterized in that, The method includes: During vehicle operation, road information is monitored in real time, and the presence of a zebra crossing is determined based on the road information. The vehicle in question is an unmanned vehicle. If so, the distance between the current vehicle and the zebra crossing is obtained, and it is determined whether the distance is less than a preset distance, wherein the larger the speed limit value of the zebra crossing, the larger the preset distance; If so, then identify other vehicles in front of the current vehicle in other lanes, and based on the other vehicles, identify the target vehicle closest to the zebra crossing in each of the other lanes; The system acquires vehicle information of the current vehicle and the target vehicle, controls the camera to capture road images of the zebra crossing, and determines whether there are pedestrians on the zebra crossing based on the road images. If so, then control the current vehicle to decelerate; The steps of acquiring vehicle information of the other vehicles and the target vehicle, controlling the camera to capture road images of the zebra crossing, and determining whether there are pedestrians at the zebra crossing based on the road images include: Obtain vehicle information of the current vehicle and the target vehicle, wherein the vehicle information includes at least vehicle speed sub-information, wherein the vehicle speed sub-information includes a first vehicle speed value of the current vehicle and a second vehicle speed value of the target vehicle; Based on the first vehicle speed value and the second vehicle speed value, determine whether the first vehicle speed value is greater than the second vehicle speed value; If so, the steps are executed: controlling the camera to capture a road image of the zebra crossing, and determining whether there are pedestrians on the zebra crossing based on the road image.

2. The vehicle driving control method for zebra crossing scenarios according to claim 1, characterized in that, The step of obtaining the distance between the current vehicle and the zebra crossing and determining whether the distance is less than a preset distance includes: The first position coordinates of the current vehicle and the target position coordinates of the zebra crossing are obtained, and the distance is calculated based on the first position coordinates and the target position coordinates.

3. The vehicle driving control method for zebra crossing scenarios according to claim 1, characterized in that, The steps of acquiring vehicle information of the other vehicles and the target vehicle, controlling the camera to capture road images of the zebra crossing, and determining whether there are pedestrians at the zebra crossing based on the road images further include: The road image is acquired and input into an image processing model to obtain each feature region and its corresponding label. Determine whether the target identifier exists among the aforementioned identifiers; If so, it means there are pedestrians on the zebra crossing, and the step of controlling the current vehicle to slow down is executed.

4. The vehicle driving control method for zebra crossing scenarios according to claim 3, characterized in that, Before the step of acquiring the road image, inputting the road image into the image processing model, and obtaining the feature identifiers, the following steps are included: Historical road images are acquired and preprocessed to obtain preprocessed images; The feature regions in the preprocessed image are identified by manual annotation to obtain a training set, wherein the feature regions include at least the human head region, the human shoulder region, and the zebra crossing region. The training set is used to train the neural network model to establish the image processing model.

5. The vehicle driving control method for zebra crossing scenarios according to claim 4, characterized in that, If, based on the road image, it is determined that there are pedestrians at the zebra crossing, the step of controlling the current vehicle to decelerate includes: The image acquisition device is controlled to continuously capture multiple images of the target road, and the multiple images of the target road are input into the image processing model to obtain the target human head region and the target zebra crossing region in each of the target road images; Based on the target human head region and the zebra crossing region, determine the target zebra crossing sub-region that overlaps with the target human head region within the target zebra crossing region; Based on the continuously acquired target zebra crossing sub-regions, the pedestrian's movement direction is determined, and it is determined whether the movement direction is away from the lane where the vehicle is located. If not, then control the current vehicle to decelerate.

6. The vehicle driving control method for zebra crossing scenarios according to claim 5, characterized in that, The step of determining the pedestrian's movement direction based on the continuously acquired target zebra crossing sub-regions, and determining whether the movement direction is away from the lane where the vehicle is located, includes: Obtain the first target zebra crossing area at the current time and the second target zebra crossing area within a preset time. The first target zebra crossing area includes at least the first leftmost target zebra crossing sub-area and the first rightmost target zebra crossing sub-area that overlap with the head area of ​​the target human body. The second target zebra crossing area includes at least the second leftmost target zebra crossing sub-area and the second rightmost target zebra crossing sub-area that overlap with the head area of ​​the target human body. Determine whether the second leftmost target zebra crossing sub-region is on the same side as the first leftmost target zebra crossing sub-region, and whether the second rightmost target zebra crossing sub-region is on the same side as the first rightmost target zebra crossing sub-region; If so, the direction of movement is determined based on the lateral direction.

7. A vehicle driving control device for zebra crossing scenarios, characterized in that, The device includes: The first judgment module is used to monitor road information in real time during vehicle operation and determine whether there is a zebra crossing ahead of the current vehicle based on the road information, wherein the current vehicle is an unmanned vehicle. The second judgment module is used to obtain the distance between the current vehicle and the zebra crossing when it is determined that there is a zebra crossing road in front of the current vehicle, and to determine whether the distance is less than a preset distance, wherein the larger the speed limit value of the zebra crossing road, the larger the preset distance. The target vehicle determination module is used to determine other vehicles in other lanes in front of the current vehicle when the distance is determined to be lower than a preset distance, and to determine the target vehicle in each of the other lanes that is closest to the zebra crossing based on the other vehicles. The third judgment module is used to obtain vehicle information of the current vehicle and the target vehicle, control the camera to take a picture of the zebra crossing road, and determine whether there are pedestrians on the zebra crossing road based on the road image. The control module is used to control the current vehicle to decelerate when it determines that there are pedestrians on the zebra crossing based on the road image. The third judgment module is further used for: Obtain vehicle information of the current vehicle and the target vehicle, wherein the vehicle information includes at least vehicle speed sub-information, wherein the vehicle speed sub-information includes a first vehicle speed value of the current vehicle and a second vehicle speed value of the target vehicle; Based on the first vehicle speed value and the second vehicle speed value, determine whether the first vehicle speed value is greater than the second vehicle speed value; If so, the steps are executed: controlling the camera to capture a road image of the zebra crossing, and determining whether there are pedestrians on the zebra crossing based on the road image.

8. A readable storage medium, characterized in that, include: The readable storage medium stores one or more programs that, when executed by a processor, implement the vehicle driving control method for zebra crossing road scenarios as described in any one of claims 1-6.

9. A vehicle, characterized in that, The vehicle includes a memory and a processor, wherein: The memory is used to store computer programs; When the processor executes the computer program stored in the memory, it implements the claims. The vehicle driving control method described in any one of 1-6 for zebra crossing road scenarios.

Images