Vehicle alarm method, device and equipment, storage medium and vehicle

Abstract

The invention discloses a vehicle warning method, device and equipment, a storage medium and a vehicle. The method comprises the following steps: acquiring an environment image of a vehicle in a parking state; the position relation between the target obstacle and the warning range of the vehicle is determined based on target pixel points in the environment image, the environment image comprises an image of the target obstacle, and the target pixel points are pixel points corresponding to the target obstacle in the environment image; and when the position relationship is that at least part of the target obstacle is within the warning range of the vehicle and the target obstacle is close to the vehicle, if the vehicle is in a vibration state, outputting first-level warning information. According to the embodiment of the invention, scratch of the vehicle can be accurately judged, and the alarm information is output to the user.

Description

Technical Field

[0001] This application belongs to the field of automotive safety technology, and in particular relates to a vehicle alarm method, device, equipment, storage medium, and vehicle. Background Technology

[0002] Today, vehicle warning systems are an important part of vehicles. Whenever a vehicle warning device detects that a parked vehicle may scrape against a surrounding obstacle, it will issue an alert to the user.

[0003] However, commonly used vehicle warning systems in related technologies generally rely on IMU (Inertial Measurement Unit) signals installed on the vehicle. They detect whether the vehicle has collided with an obstacle by judging whether the IMU signal fluctuates significantly in a short period of time. However, this type of method has low accuracy and is prone to misjudgment in scenarios such as rain, thunder, or heavy vehicles passing by, thus causing false alarms. Summary of the Invention

[0004] This application provides a vehicle alarm method, device, equipment, storage medium, and vehicle, which can solve the problem in the prior art that it is impossible to accurately determine whether a vehicle has been scratched, thus causing false alarms.

[0005] In a first aspect, embodiments of this application provide a vehicle alarm method, the method comprising:

[0006] Acquire environmental images of the vehicle while it is parked;

[0007] The positional relationship between the target obstacle and the vehicle's alarm range is determined based on the target pixels in the environmental image, wherein the environmental image includes an image of the target obstacle, and the target pixels are the pixels corresponding to the target obstacle in the environmental image;

[0008] If the target obstacle is at least partially within the vehicle's alarm range and is close to the vehicle, and the vehicle is vibrating, a Level 1 alarm message is output.

[0009] In some embodiments, determining the positional relationship between the target obstacle and the vehicle's warning range based on target pixels in the environmental image includes:

[0010] Target pixels in the environmental image are determined by semantic segmentation;

[0011] The estimated depth value corresponding to the target pixel is predicted using monocular depth estimation technology;

[0012] The estimated depth value corresponding to the target pixel and the position of the target pixel in the environmental image are converted into the actual position of the target obstacle.

[0013] The positional relationship between the target obstacle and the vehicle's alarm range is determined based on the actual location and the vehicle's current location.

[0014] In some embodiments, determining the positional relationship between the target obstacle and the vehicle's warning range based on target pixels in the environmental image includes:

[0015] Target pixels in the environmental image are determined by semantic segmentation;

[0016] The alarm range of the vehicle is projected onto the environmental image, and a projected image of the alarm range is generated in the environmental image;

[0017] The positional relationship between the target obstacle and the vehicle's alarm range is determined based on whether there is an intersection between the projected image of the alarm range in the environmental image and the set of target pixels.

[0018] In some embodiments, determining the target pixel in the environment image through semantic segmentation includes:

[0019] By performing semantic segmentation on the environmental image, the pixels in the environmental image are divided into first pixels and second pixels, wherein the first pixels are the background pixels in the environmental image, and the second pixels are the foreground elements pixels in the environmental image.

[0020] Based on the image features, a detection box for the target obstacle is generated on the environmental image;

[0021] The second pixel within the detection frame is determined as the target pixel.

[0022] In some embodiments, before outputting a level one alarm message if the vehicle is in a vibration state, the method further includes:

[0023] Obtain the dynamic parameters of the vehicle within a first time period before the current moment, the dynamic parameters including the vehicle's acceleration and angular velocity;

[0024] If the highest value of the dynamic parameter within the first time period is greater than the preset threshold corresponding to the dynamic parameter, the vehicle is determined to be in a vibration state.

[0025] In some embodiments, after determining the positional relationship between the target obstacle and the vehicle's warning range based on target pixels in the environmental image, the method further includes:

[0026] If the target obstacle is at least partially within the vehicle's alarm range and is close to the vehicle, and the vehicle is not vibrating, a secondary alarm message is output.

[0027] When the target obstacle is not within the vehicle's alarm range and the vehicle is vibrating, the secondary alarm information is output, wherein the risk level indicated by the secondary alarm information is lower than that indicated by the primary alarm information.

[0028] Secondly, embodiments of this application provide a vehicle warning device, the device comprising:

[0029] The acquisition module is used to acquire environmental images of the vehicle while it is parked.

[0030] The determination module is used to determine the positional relationship between the target obstacle and the vehicle's alarm range based on the target pixels in the environmental image, wherein the environmental image includes an image of the target obstacle, and the target pixels are the pixels corresponding to the target obstacle in the environmental image;

[0031] The first output module is used to output a level one alarm message when the target obstacle is at least partially within the alarm range of the vehicle and the target obstacle is close to the vehicle, and the vehicle is in a vibration state.

[0032] Thirdly, embodiments of this application provide a vehicle alarm device, the device including: a processor and a memory storing computer program instructions;

[0033] The processor implements the vehicle alarm method described above when executing computer program instructions.

[0034] Fourthly, embodiments of this application provide a computer storage medium storing computer program instructions, which, when executed by a processor, implement the vehicle alarm method described above.

[0035] Fifthly, embodiments of this application provide a vehicle that includes computer program instructions, which, when executed by a processor, implement the vehicle alarm method described above.

[0036] In this application, an environmental image of the vehicle is acquired; the positional relationship between the target obstacle and the vehicle's warning range is determined based on target pixels in the environmental image. The environmental image includes an image of the target obstacle, and the target pixels are the pixels corresponding to the target obstacle in the environmental image. When the positional relationship is such that the target obstacle is at least partially within the vehicle's warning range and is close to the vehicle, a Level 1 warning is output if the vehicle is vibrating. In this way, the positional relationship between the target obstacle and the vehicle's warning range, as well as the movement trend of the target obstacle, can be determined from the image of the target obstacle in the captured environmental image of the vehicle. The positional relationship and movement trend between the target obstacle and the vehicle's warning range are used to determine whether a nearby obstacle exists within the vehicle's warning range. If a nearby obstacle exists within the vehicle's warning range and the vehicle vibrates, an alarm is triggered. This method only issues an alarm when both an obstacle is nearby and the vehicle is vibrating simultaneously, avoiding misjudgments in scenarios such as rain, thunder, or heavy vehicle passage. It can accurately identify whether a collision has occurred, thus providing accurate warning information to the user. Attached Figure Description

[0037] To more clearly illustrate the technical solutions of the embodiments of this application, the drawings used in the embodiments of this application will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0038] Figure 1 This is a schematic flowchart of a vehicle alarm method provided in an embodiment of this application;

[0039] Figure 2 This is a schematic diagram of the structure of a vehicle alarm device provided in one embodiment of this application;

[0040] Figure 3 This is a schematic diagram of the hardware structure of a vehicle alarm device provided in one embodiment of this application. Detailed Implementation

[0041] The features and exemplary embodiments of various aspects of this application will be described in detail below. To make the objectives, technical solutions, and advantages of this application clearer, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are only intended to explain this application and not to limit it. For those skilled in the art, this application can be implemented without some of these specific details. The following description of the embodiments is merely to provide a better understanding of this application by illustrating examples of this application.

[0042] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising..." does not exclude the presence of additional identical elements in the process, method, article, or apparatus that includes the element.

[0043] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. The embodiments will now be described in detail with reference to the accompanying drawings.

[0044] Specifically, in order to solve the problems of the prior art, embodiments of this application provide a vehicle alarm method, apparatus, device, storage medium, and vehicle. The vehicle alarm method provided in this application embodiment will be described first below.

[0045] Figure 1 A flowchart illustrating a vehicle alarm method according to an embodiment of this application is shown. The vehicle alarm method is applied to a processor, which can be located in a smart terminal, an in-vehicle infotainment system, or a server. The method includes the following steps:

[0046] S110: Acquire an environmental image of the vehicle while it is parked.

[0047] In this embodiment, the vehicle's environmental image refers to an image of the environment surrounding the vehicle. These environmental images can be acquired by cameras or sensors installed on the vehicle. For example, four surround-view fisheye cameras can be installed in the four directions of the vehicle body (front, rear, left, and right). When the vehicle is parked, the surround-view fisheye cameras can take an image at fixed intervals as the vehicle's environmental image.

[0048] S120, determine the positional relationship between the target obstacle and the vehicle's alarm range based on the target pixels in the environmental image, wherein the environmental image includes the image of the target obstacle, and the target pixels are the pixels corresponding to the target obstacle in the environmental image.

[0049] In this embodiment, the target obstacle is a movable object near the vehicle that may scrape against it, such as other cars, electric vehicles, bicycles, etc. The vehicle's warning range can be a three-dimensional spatial area extending around the vehicle in advance, as set by the user. Obstacles within the vehicle's warning range are considered to pose a risk of scraping against the vehicle. For example, a distance of 30cm or less from the vehicle is considered within the vehicle's warning range, or a distance of 50cm or less is considered within the vehicle's warning range.

[0050] After determining the vehicle's warning range, each environmental image can be identified to determine whether it includes an image of the target obstacle. If the environmental image contains an image of the target obstacle, the target pixels corresponding to that image can be segmented from the environmental image. Since the category label of the target pixels is "target obstacle," the features of the target pixels can be extracted, and based on these features, the positional relationship between the target obstacle and the vehicle's warning range in the real world can be determined. This positional relationship includes situations where the target obstacle is partially within the warning range, entirely within the warning range, or not within the warning range.

[0051] As an optional embodiment, determining the positional relationship between the target obstacle and the vehicle's warning range based on target pixels in the environmental image includes:

[0052] Target pixels in the environmental image are determined by semantic segmentation;

[0053] The estimated depth value corresponding to the target pixel is predicted using monocular depth estimation technology;

[0054] The estimated depth value corresponding to the target pixel and the position of the target pixel in the environmental image are converted into the actual position of the target obstacle.

[0055] The positional relationship between the target obstacle and the vehicle's alarm range is determined based on the actual location and the vehicle's current location.

[0056] In this embodiment, if the environmental image includes an image of a target obstacle, the environmental image can be semantically segmented. Through semantic segmentation, each pixel in the environmental image can be assigned a category label, and the pixel with the category label of target obstacle can be determined as the target pixel.

[0057] After the target pixels are segmented, monocular depth estimation can be used to determine the estimated depth value of each target pixel. Specifically, monocular depth estimation refers to using a single camera to estimate the estimated depth value of each target pixel in the environmental image in three-dimensional space. The estimated depth value is the distance from the camera to the surface of the target obstacle corresponding to the target pixel.

[0058] Since a target pixel is a pixel in the image of a target obstacle, and an obstacle surface point is a point on the surface of the target obstacle in the real world, each target pixel corresponds to one obstacle surface point. After estimating the estimated depth value corresponding to a target pixel, the position of the target pixel in the environmental image and its estimated depth value can be converted into the first coordinate value of the obstacle surface point in the world coordinate system. Since the first coordinate value can describe the actual position of the obstacle surface point in three-dimensional space, the actual position of the target obstacle in three-dimensional space can be determined after determining the first coordinate value of each obstacle surface point.

[0059] Since the vehicle's current location in the real world is known, the vehicle's warning range is a three-dimensional spatial area extending around the vehicle based on its location. Therefore, after determining the actual location of the target obstacle, the vehicle's warning range in the real world can be determined based on its location in the real world, and the relationship between the vehicle's warning range and the target obstacle's location can be determined based on the vehicle's warning range in the real world and the actual location of the target obstacle.

[0060] Using the above method, the target obstacle can be photographed by a camera or camera installed on the vehicle to obtain an environmental image. By extracting the features of the target pixels in the environmental image, the position of the target obstacle in three-dimensional space can be accurately determined, and the positional relationship between the target obstacle and the vehicle's warning range can be determined.

[0061] As an optional embodiment, determining the positional relationship between the target obstacle and the vehicle's warning range based on target pixels in the environmental image includes:

[0062] Target pixels in the environmental image are determined by semantic segmentation;

[0063] The alarm range of the vehicle is projected onto the environmental image, and a projected image of the alarm range is generated in the environmental image;

[0064] The positional relationship between the target obstacle and the vehicle's alarm range is determined based on whether there is an intersection between the projected image of the alarm range in the environmental image and the set of target pixels.

[0065] In this embodiment, semantic segmentation can be performed on the environmental image. Each pixel in the environmental image is assigned a category label through semantic segmentation, and the pixel with the category label of target obstacle is determined as the target pixel.

[0066] After determining the target pixels through segmentation, a computer-generated scene image of the vehicle can be used. The vehicle scene image can include a top-down view of the scene in which the vehicle is located, and includes a top-down view of the vehicle's warning range.

[0067] After generating the scene image, the top-down image of the vehicle's warning range can be projected onto the vehicle's environmental image using inverse perspective mapping (IPM) to obtain the projected image of the warning range. In this way, both the projected image of the warning range and the target obstacle can be displayed in the environmental image. Therefore, it is possible to determine from the environmental image whether there is any intersection between the target pixel and the projected image of the warning range, and further determine the positional relationship between the target obstacle and the warning range.

[0068] As an optional embodiment, the method further includes:

[0069] Target pixels in the environmental image are determined by semantic segmentation;

[0070] Based on the target pixel, the image of the target obstacle in the environmental image is projected onto the scene image of the vehicle to generate a projected image of the target obstacle in the scene image, wherein the scene image is a top view of the scene where the vehicle is located, and the scene image includes a top view image of the vehicle's alarm range.

[0071] The relative positions of the target obstacle and the vehicle's alarm range are determined based on the projected image of the target obstacle in the scene image and the top-view image of the vehicle's alarm range.

[0072] In this embodiment, semantic segmentation can also be performed on the environmental image. Each pixel in the environmental image is assigned a category label through semantic segmentation, and the pixel with the category label of target obstacle is determined as the target pixel.

[0073] After determining the target pixels through segmentation, the vehicle scene image generated by a computer can be used. The vehicle scene image can be a top-down view of the scene in which the vehicle is located, and the vehicle scene image includes a top-down view of the vehicle.

[0074] After generating the scene image, the image of the target obstacle in the environment image represented by the target pixel can be projected onto the scene image through inverse perspective mapping (IPM). In this way, the projected image of the target obstacle and the top view image of the vehicle's warning range can be displayed in the scene image. Therefore, it can be determined from the scene image whether there is an intersection between the projected image of the target obstacle and the top view image of the vehicle's warning range, and further determine the positional relationship between the target obstacle and the warning range.

[0075] Using the above method, the alarm ranges of both the target obstacle and the vehicle can be displayed in the environmental or scene image, allowing for an accurate and clear understanding of the positional relationship between the alarm ranges of the target obstacle and the vehicle from the scene image.

[0076] S130, if the target obstacle is at least partially within the alarm range of the vehicle and the target obstacle is close to the vehicle, and the vehicle is in a vibration state, output a level one alarm message.

[0077] In this embodiment, after determining the positional relationship between the vehicle and the target obstacle, whenever a target obstacle enters the alarm range, it is further determined whether the target obstacle moved within a period of time prior to the current moment. If the target obstacle moved within a period of time prior to the current moment and continuously approached the vehicle within that period, it is considered that there is a risk of the target obstacle scraping the vehicle. At this time, the vibration status of the vehicle can be further acquired. If the vehicle vibrates while parked, it can be determined that there is a high probability of a scrape between the vehicle and the target obstacle. In this case, the vehicle can output a level one alarm message to the user to alert the user that the risk of a scrape is high. The output alarm message can be a voice prompt or an alarm text message sent to the user's mobile terminal.

[0078] In this application, an environmental image of the vehicle is acquired; the positional relationship between the target obstacle and the vehicle's warning range is determined based on target pixels in the environmental image. The environmental image includes an image of the target obstacle, and the target pixels are the pixels corresponding to the target obstacle in the environmental image. When the positional relationship is such that the target obstacle is at least partially within the vehicle's warning range and is close to the vehicle, a Level 1 warning is output if the vehicle is vibrating. In this way, the positional relationship between the target obstacle and the vehicle's warning range, as well as the movement trend of the target obstacle, can be determined from the image of the target obstacle in the captured environmental image of the vehicle. The positional relationship and movement trend between the target obstacle and the vehicle's warning range are used to determine whether a nearby obstacle exists within the vehicle's warning range. If a nearby obstacle exists within the vehicle's warning range and the vehicle vibrates, an alarm is triggered. This method only issues an alarm when both an obstacle is nearby and the vehicle is vibrating simultaneously, avoiding misjudgments in scenarios such as rain, thunder, or heavy vehicle passage. It can accurately identify whether a collision has occurred, thus providing accurate warning information to the user.

[0079] As an optional embodiment, after determining the positional relationship between the target obstacle and the vehicle's warning range based on target pixels in the environmental image, the method further includes:

[0080] If the target obstacle is at least partially within the vehicle's alarm range and is close to the vehicle, and the vehicle is not vibrating, a secondary alarm message is output.

[0081] When the target obstacle is not within the vehicle's alarm range and the vehicle is vibrating, the secondary alarm information is output, wherein the risk level indicated by the secondary alarm information is lower than that indicated by the primary alarm information.

[0082] In this embodiment, after determining the positional relationship between the vehicle and the target obstacle, whenever a target obstacle enters the alarm range, it is further determined whether the target obstacle moved during a period of time prior to the current moment. If the target obstacle moved during that period and continuously approached the vehicle, it is considered that there is a risk of collision between the target obstacle and the vehicle. At this time, the vibration status of the vehicle can be further obtained. If the vehicle does not vibrate while parked, it can be determined that the risk of collision between the vehicle and the target obstacle is low. In this case, the vehicle can output a secondary alarm message to the user. The risk level corresponding to the secondary alarm message is lower than that of the first alarm message.

[0083] Similarly, if a vehicle vibrates while parked, but no target obstacle enters the vehicle's alarm range, a secondary alarm message can still be sent to the user.

[0084] In addition, if a target obstacle enters the alarm range, but the target obstacle has not moved in the period of time prior to the current moment, there is no need to output an alarm message.

[0085] By combining the vehicle's vibration status with visual recognition to assess different risk levels, users can be given different levels of alerts.

[0086] As an optional embodiment, determining the target pixel in the environment image through semantic segmentation includes:

[0087] By performing semantic segmentation on the environmental image, the pixels in the environmental image are divided into first pixels and second pixels, wherein the first pixels are the background pixels in the environmental image, and the second pixels are the foreground elements pixels in the environmental image.

[0088] Based on the image features, a detection box for the target obstacle is generated on the environmental image;

[0089] The second pixel within the detection frame is determined as the target pixel.

[0090] In this embodiment, a trained semantic segmentation model can be used to perform semantic segmentation on the environment image. The environment image can be input into the trained semantic segmentation model, which can classify each pixel in the environment image into a predefined category. Then, the semantic segmentation outputs a segmentation result of the same size as the input environment image, in which each pixel is labeled as belonging to a certain category.

[0091] For example, a semantic segmentation model may include a feature extraction module, a pixel classification module, and a bounding box prediction module. After the environmental image is input into the semantic segmentation model, the feature extraction module uses methods such as convolutional neural networks (CNNs) or visual attention transformers (VITs) to extract features from the input image. Based on the extracted features, the pixel classification module can use methods such as multilayer neural networks (NNs) to classify each pixel in the environmental image as either a second pixel belonging to the foreground element category or a first pixel belonging to the background category. In the bounding box prediction module, methods such as multilayer neural networks (NNs) can be used to generate bounding boxes for target obstacles in the environmental image. These bounding boxes are used to identify the location and boundaries of target obstacles in the environmental image.

[0092] Using the above method, the second pixel within the detection box in the image output by the semantic segmentation model can be used to determine the target pixel corresponding to the target obstacle in the image. In this way, the location of the target obstacle can be accurately identified from the environmental image.

[0093] As an optional embodiment, before outputting a level one alarm message if the vehicle is in a vibration state, the method further includes:

[0094] Obtain the dynamic parameters of the vehicle within a first time period before the current moment, the dynamic parameters including the vehicle's acceleration and angular velocity;

[0095] If the highest value of the dynamic parameter within the first time period is greater than the preset threshold corresponding to the dynamic parameter, the vehicle is determined to be in a vibration state.

[0096] In this embodiment, the vehicle's acceleration and angular velocity can be collectively referred to as dynamic parameters. These dynamic parameters can be detected by the vehicle's IMU (Inertial Measurement Unit) over a first time period prior to the current moment.

[0097] When a vehicle is parked, its dynamic parameters such as acceleration and angular velocity should normally be zero or very small. Only when a collision occurs will a parked vehicle experience significant vibrations, causing substantial changes in dynamic parameters such as acceleration and angular velocity.

[0098] Therefore, if the highest value of the dynamic parameter in the first time period before the current moment is greater than the preset threshold corresponding to the dynamic parameter, and the car owner has not actively started the vehicle, it can be considered that the vehicle is in a state of vibration, that is, the vehicle may have collided with other vehicles. The preset threshold corresponding to the angular velocity is the angular velocity threshold, and the dynamic threshold corresponding to the acceleration is the acceleration threshold.

[0099] In this way, dynamic parameters can be used to determine more accurately whether a vehicle is vibrating while parked.

[0100] Based on the vehicle alarm method provided in the above embodiments, this application also provides specific implementation methods of a vehicle alarm device. Please refer to the following embodiments.

[0101] First see Figure 2 The vehicle alarm device 200 provided in this application embodiment includes the following modules:

[0102] The acquisition module 201 is used to acquire environmental images of the vehicle in a parked state;

[0103] The determining module 202 is used to determine the positional relationship between the target obstacle and the alarm range of the vehicle based on the target pixels in the environmental image, wherein the environmental image includes the image of the target obstacle, and the target pixels are the pixels corresponding to the target obstacle in the environmental image;

[0104] The first output module 203 is used to output a first-level alarm message when the target obstacle is at least partially within the alarm range of the vehicle and the target obstacle is close to the vehicle, and the vehicle is in a vibration state.

[0105] The device can acquire environmental images of the vehicle; determine the positional relationship between the target obstacle and the vehicle's alarm range based on target pixels in the environmental images, where the environmental images include images of the target obstacle, and the target pixels are the pixels corresponding to the target obstacle in the environmental images; when the positional relationship is such that the target obstacle is at least partially within the vehicle's alarm range and is close to the vehicle, if the vehicle is vibrating, a level one alarm message is output. In this way, the positional relationship between the target obstacle and the vehicle's alarm range, as well as the movement trend of the target obstacle, can be determined by the image of the target obstacle in the captured environmental images of the vehicle. By analyzing the positional relationship and movement trend between the target obstacle and the vehicle's alarm range, it can determine whether there is a nearby obstacle within the vehicle's alarm range. If there is a nearby obstacle within the vehicle's alarm range and the vehicle is vibrating, an alarm is triggered. This method only issues an alarm when both an obstacle is nearby and the vehicle is vibrating simultaneously, avoiding misjudgments in scenarios such as rain, thunder, or heavy vehicle passage, accurately identifying whether a collision has occurred, and thus outputting accurate alarm information to the user.

[0106] As one implementation of this application, the determining module 202 may further include:

[0107] A segmentation unit is used to determine target pixels in the environmental image through semantic segmentation;

[0108] The prediction unit is used to predict the estimated depth value corresponding to the target pixel using monocular depth estimation technology;

[0109] A conversion unit is used to convert the estimated depth value corresponding to the target pixel and the position of the target pixel in the environmental image into the actual position of the target obstacle.

[0110] The first determining unit is used to determine the positional relationship between the target obstacle and the vehicle's alarm range based on the actual location and the vehicle's current positioning location.

[0111] As one implementation of this application, the determining module 202 may further include:

[0112] A segmentation unit is used to determine target pixels in the environmental image through semantic segmentation;

[0113] A projection unit is used to project the alarm range of the vehicle onto the environmental image, and generate a projected image of the alarm range in the environmental image;

[0114] The second determining unit is used to determine the positional relationship between the target obstacle and the vehicle's alarm range based on whether there is an intersection between the projected image of the alarm range in the environmental image and the set of target pixels.

[0115] As one implementation of this application, the above-mentioned segmentation unit can also be used for:

[0116] By performing semantic segmentation on the environmental image, the pixels in the environmental image are divided into first pixels and second pixels, wherein the first pixels are the background pixels in the environmental image, and the second pixels are the foreground elements pixels in the environmental image.

[0117] Based on the image features, a detection box for the target obstacle is generated on the environmental image;

[0118] The second pixel within the detection frame is determined as the target pixel.

[0119] As one implementation of this application, the vehicle alarm device 200 may further include:

[0120] The acquisition module is used to acquire the dynamic parameters of the vehicle within a first time period before the current moment, the dynamic parameters including the vehicle's acceleration and angular velocity;

[0121] The vibration module is used to determine that the vehicle is in a vibration state when the highest value of the dynamic parameter within a first time period is greater than a preset threshold corresponding to the dynamic parameter.

[0122] As one implementation of this application, the vehicle alarm device 200 may further include:

[0123] The second output module is used to output a secondary alarm message when the target obstacle is at least partially within the alarm range of the vehicle and the target obstacle is close to the vehicle, and the vehicle is not in a vibration state.

[0124] The third output module is used to output the secondary alarm information when the target obstacle is not within the alarm range of the vehicle and the vehicle is in a vibration state, wherein the risk level indicated by the secondary alarm information is lower than that of the primary alarm information.

[0125] The vehicle alarm device provided in this embodiment of the invention can implement the steps in the above method embodiments, and will not be repeated here to avoid repetition.

[0126] Figure 3 A schematic diagram of the hardware structure of the vehicle alarm device provided in an embodiment of this application is shown.

[0127] The vehicle's alarm device may include a processor 1001 and a memory 1002 storing computer program instructions.

[0128] Specifically, the processor 1001 may include a central processing unit (CPU), an application-specific integrated circuit (ASIC), or one or more integrated circuits that can be configured to implement the embodiments of this application.

[0129] Memory 1002 may include mass storage for data or instructions. For example, and not limitingly, memory 1002 may include a hard disk drive (HDD), floppy disk drive, flash memory, optical disk, magneto-optical disk, magnetic tape, or Universal Serial Bus (USB) drive, or a combination of two or more of these. Where appropriate, memory 1002 may include removable or non-removable (or fixed) media. Where appropriate, memory 1002 may be internal or external to the integrated gateway disaster recovery device. In a particular embodiment, memory 1002 is non-volatile solid-state memory.

[0130] Memory may include read-only memory (ROM), random access memory (RAM), disk storage media devices, optical storage media devices, flash memory devices, and electrical, optical, or other physical / tangible memory storage devices. Therefore, typically, memory includes one or more tangible (non-transitory) computer-readable storage media (e.g., memory devices) encoded with software including computer-executable instructions, and when the software is executed (e.g., by one or more processors), it is operable to perform the operations described with reference to the methods according to one aspect of this disclosure.

[0131] The processor 1001 reads and executes computer program instructions stored in the memory 1002 to implement any of the vehicle alarm methods in the above embodiments.

[0132] In one example, the vehicle's alarm device may also include a communication interface 1003 and a bus 1010. Wherein, for example... Figure 3 As shown, the processor 1001, memory 1002, and communication interface 1003 are connected through bus 1010 and complete communication with each other.

[0133] The communication interface 1003 is mainly used to realize communication between various modules, devices, units and / or equipment in the embodiments of this application.

[0134] Bus 1010 includes hardware, software, or both, that couples components of a vehicle's alarm devices together. For example, and not limitingly, the bus may include an Accelerated Graphics Port (AGP) or other graphics bus, an Enhanced Industry Standard Architecture (EISA) bus, a Front Side Bus (FSB), HyperTransport (HT) interconnect, an Industry Standard Architecture (ISA) bus, an Infinite Bandwidth Interconnect, a Low Pin Count (LPC) bus, a memory bus, a Microchannel Architecture (MCA) bus, a Peripheral Component Interconnect (PCI) bus, a PCI-Express (PCI-X) bus, a Serial Advanced Technology Attachment (SATA) bus, a Video Electronics Standards Association Local (VLB) bus, or other suitable buses, or combinations of two or more of these. Where appropriate, bus 1010 may include one or more buses. Although specific buses are described and illustrated in embodiments of this application, any suitable bus or interconnect is contemplated herein.

[0135] The vehicle's alarm device can be based on the above embodiments, thereby realizing the vehicle alarm method and device combined with the above.

[0136] Furthermore, in conjunction with the vehicle alarm methods in the above embodiments, this application embodiment can provide a computer storage medium for implementation. This computer storage medium stores computer program instructions; when these computer program instructions are executed by a processor, they implement any of the vehicle alarm methods in the above embodiments and achieve the same technical effect. To avoid repetition, further details are omitted here. The aforementioned computer-readable storage medium may include non-transitory computer-readable storage media, such as read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks, etc., and is not limited thereto.

[0137] In addition, this application also provides a vehicle including computer program instructions, which, when executed by a processor, can implement the steps and corresponding content of the aforementioned method embodiments.

[0138] It should be clarified that this application is not limited to the specific configurations and processes described above and shown in the figures. For the sake of brevity, detailed descriptions of known methods are omitted here. In the above embodiments, several specific steps are described and shown as examples. However, the method process of this application is not limited to the specific steps described and shown. Those skilled in the art can make various changes, modifications, and additions, or change the order of steps, after understanding the spirit of this application.

[0139] The functional blocks shown in the above block diagram can be implemented as hardware, software, firmware, or a combination thereof. When implemented in hardware, they can be, for example, electronic circuits, application-specific integrated circuits (ASICs), appropriate firmware, plug-ins, function cards, etc. When implemented in software, the elements of this application are programs or code segments used to perform the required tasks. Programs or code segments can be stored on a machine-readable medium or transmitted over a transmission medium or communication link via data signals carried on a carrier wave. "Machine-readable medium" can include any medium capable of storing or transmitting information. Examples of machine-readable media include electronic circuits, semiconductor memory devices, ROM, flash memory, erasable ROM (EROM), floppy disks, CD-ROMs, optical disks, hard disks, fiber optic media, radio frequency (RF) links, etc. Code segments can be downloaded via computer networks such as the Internet, intranets, etc.

[0140] It should also be noted that the exemplary embodiments mentioned in this application describe methods or systems based on a series of steps or apparatus. However, this application is not limited to the order of the above steps; that is, the steps can be performed in the order mentioned in the embodiments, or in a different order, or several steps can be performed simultaneously.

[0141] The aspects of this disclosure have been described above with reference to flowchart illustrations and / or block diagrams of methods, apparatus, and vehicles according to embodiments of this disclosure. It should be understood that each block in the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing apparatus to create a machine such that these instructions, executable via the processor of the computer or other programmable data processing apparatus, enable the implementation of the functions / actions specified in one or more blocks of the flowchart illustrations and / or block diagrams. Such a processor can be, but is not limited to, a general-purpose processor, a special-purpose processor, a special application processor, or a field-programmable logic circuit. It is also understood that each block in the block diagrams and / or flowchart illustrations, and combinations of blocks in the block diagrams and / or flowchart illustrations, can also be implemented by special-purpose hardware performing the specified functions or actions, or can be implemented by a combination of special-purpose hardware and computer instructions.

[0142] The above are merely specific embodiments of this application. Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, modules, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here. It should be understood that the protection scope of this application is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in this application, and these modifications or substitutions should all be covered within the protection scope of this application.

Claims

1. A vehicle alarm method, characterized in that, The method includes: Acquire environmental images of the vehicle while it is parked; The positional relationship between the target obstacle and the vehicle's alarm range is determined based on the target pixels in the environmental image, wherein the environmental image includes an image of the target obstacle, and the target pixels are the pixels corresponding to the target obstacle in the environmental image; If the target obstacle is at least partially within the vehicle's alarm range and is close to the vehicle, and the vehicle is vibrating, a Level 1 alarm message is output.

2. The vehicle alarm method according to claim 1, characterized in that, Determining the positional relationship between the target obstacle and the vehicle's warning range based on target pixels in the environmental image includes: Target pixels in the environmental image are determined by semantic segmentation; The estimated depth value corresponding to the target pixel is predicted using monocular depth estimation technology; The estimated depth value corresponding to the target pixel and the position of the target pixel in the environmental image are converted into the actual position of the target obstacle. The positional relationship between the target obstacle and the vehicle's alarm range is determined based on the actual location and the vehicle's current location.

3. The vehicle alarm method according to claim 1, characterized in that, Determining the positional relationship between the target obstacle and the vehicle's warning range based on target pixels in the environmental image includes: Target pixels in the environmental image are determined by semantic segmentation; The alarm range of the vehicle is projected onto the environmental image, and a projected image of the alarm range is generated in the environmental image; The positional relationship between the warning range of the target obstacle and the vehicle is determined based on whether there is an intersection between the projected image of the warning range in the environmental image and the set of target pixels.

4. The vehicle alarm method according to claim 2 or 3, characterized in that, The step of determining the target pixels in the environmental image through semantic segmentation includes: By performing semantic segmentation on the environmental image, the pixels in the environmental image are divided into first pixels and second pixels, wherein the first pixels are the background pixels in the environmental image, and the second pixels are the foreground elements pixels in the environmental image. Based on the image features, a detection box for the target obstacle is generated on the environmental image; The second pixel within the detection frame is determined as the target pixel.

5. The vehicle alarm method according to claim 1, characterized in that, Before outputting a Level 1 alarm message if the vehicle is in a vibration state, the method further includes: Obtain the dynamic parameters of the vehicle within a first time period before the current moment, the dynamic parameters including the vehicle's acceleration and angular velocity; If the highest value of the dynamic parameter within the first time period is greater than the preset threshold corresponding to the dynamic parameter, the vehicle is determined to be in a vibration state.

6. The vehicle alarm method according to claim 1, characterized in that, After determining the positional relationship between the target obstacle and the vehicle's warning range based on target pixels in the environmental image, the method further includes: If the target obstacle is at least partially within the vehicle's alarm range and is close to the vehicle, and the vehicle is not vibrating, a secondary alarm message is output. When the target obstacle is not within the vehicle's alarm range and the vehicle is vibrating, the secondary alarm information is output, wherein the risk level indicated by the secondary alarm information is lower than that indicated by the primary alarm information.

7. A vehicle warning device, characterized in that, The device includes: The acquisition module is used to acquire environmental images of the vehicle while it is parked. The determination module is used to determine the positional relationship between the target obstacle and the vehicle's alarm range based on the target pixels in the environmental image, wherein the environmental image includes an image of the target obstacle, and the target pixels are the pixels corresponding to the target obstacle in the environmental image; The first output module is used to output a level one alarm message when the target obstacle is at least partially within the alarm range of the vehicle and the target obstacle is close to the vehicle, and the vehicle is in a vibration state.

8. A vehicle alarm device, characterized in that, The vehicle's alarm device includes: a processor and a memory storing computer program instructions; When the processor executes the computer program instructions, it implements the vehicle alarm method as described in any one of claims 1-6.

9. A computer storage medium, characterized in that, The computer storage medium stores computer program instructions, which, when executed by a processor, implement the vehicle alarm method as described in any one of claims 1-6.

10. A vehicle, characterized in that, The vehicle includes at least one of the following: The vehicle alarm device as described in claim 7, the vehicle alarm equipment as described in claim 8, and the computer storage medium as described in claim 9.

Images