Front vehicle position determination method and device, vehicle and medium

By using a pre-trained recognition model in the car to identify image information, filter the interfering light source according to the size and distance information of the light source, accurately determine the position of the headlights in the front car, solving the problem of errors in the front car position caused by the interference of the light source at night. Improves positioning accuracy and saves costs.

CN119068449APending Publication Date: 2024-12-03XIAOMI EV TECH CO LTD +1
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202310640075.5
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2023-05-31
Publication Date
2024-12-03

AI Technical Summary

Technical Problem

The adaptive high beam system in the existing automotive intelligent lighting functions is prone to errors in the identification of headlights due to interference from light sources, and it is difficult to accurately locate the headlight position of the headlights.

Method used

The image information is identified through the pre-trained recognition model, the size and distance information of the light source information are obtained, the interfering light source is filtered according to the preset conditions, and the position of the headlights is determined.

Benefits of technology

It improves the accuracy of headlight position recognition, reduces labor and time costs, and avoids the increase in model parameters and storage resource usage.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119068449A_ABST
    Figure CN119068449A_ABST
Patent Text Reader

Abstract

The invention provides a front vehicle position determination method and device, a vehicle and a medium, and relates to the technical field of automobiles, and the method comprises the steps: firstly obtaining image information including various types of light source information, then carrying out the preliminary recognition of the image information through a pre-trained recognition model, obtaining a light source recognition result corresponding to each type of light source information, and carrying out the recognition of the light source. As the interference light source existing in the image information may cause interference on the identification of the vehicle lamp, the size information and the distance information are obtained according to the light source identification result corresponding to each kind of light source information, the light source identification result is detected again by using the distance information and / or the size information, and then the vehicle lamp of the front vehicle is determined, so that the position of the front vehicle is determined. The accuracy of vehicle lamp position identification is ensured, so that the accuracy of front vehicle position positioning is ensured.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present disclosure relates to the field of automobile technology, and in particular to a method, device, vehicle and medium for determining a position of a leading vehicle. Background Art

[0002] The intelligent lighting function of automobiles usually includes an adaptive driving beam (ADB) system. ADB uses a light source detection model to sense the front lights and thus know the position of the vehicle in front. Since there are many light sources at night, street lights, indicator lights, traffic lights, reflections from water on the road, and other light sources will interfere with the light source detection model's recognition of the lights, which can easily lead to incorrect judgment of the lights, making it difficult to accurately locate the position of the vehicle in front. Summary of the invention

[0003] In order to overcome the problems existing in the related art, the present disclosure provides a method, device, vehicle and medium for determining the position of a leading vehicle.

[0004] According to a first aspect of an embodiment of the present disclosure, a method for determining the position of a leading vehicle is provided, the method comprising: obtaining image information including multiple light source information; recognizing the image information through a pre-trained recognition model to obtain a light source recognition result corresponding to each of the multiple light source information; obtaining size information of each of the multiple light sources and distance information of each of the light sources from the ground based on the multiple light source recognition results; if there is a light source among the multiple light sources whose size information satisfies a first preset condition, and / or whose distance information satisfies a second preset condition, determining that the light source is a headlight of a leading vehicle; and determining the position of the leading vehicle based on the headlight of the leading vehicle.

[0005] Optionally, the light source identification result corresponding to each light source information includes the first coordinates in the image information of two vertices on a diagonal line of a rectangular detection box for each light source selected in the image information, and obtaining the size information of each light source among the multiple light sources based on the multiple light source identification results, including: obtaining the length of the detection box and the width of the detection box based on the first coordinates of the two vertices; and obtaining the aspect ratio based on the ratio between the length and the width.

[0006] Optionally, if there is a light source among the multiple light sources whose size information satisfies a first preset condition, then the light source is determined to be the headlight of the vehicle in front, including: if there is a light source among the multiple light sources whose aspect ratio is within a preset aspect ratio range, and the size information representing the light source satisfies the first preset condition, then the light source is determined to be the headlight of the vehicle in front.

[0007] Optionally, obtaining the distance information of each of the multiple light sources from the ground based on the results of identifying multiple light sources includes: obtaining the center point coordinates of the center point of the detection frame of each light source based on the first coordinates of the two vertices corresponding to each of the multiple light sources, wherein the center point is the intersection of two diagonals of the detection frame; mapping each center point coordinate in the image information to a second coordinate in the three-dimensional space coordinate system corresponding to each center point coordinate based on a preset mapping relationship, wherein the preset mapping relationship represents the correspondence between the coordinates in the image information and the coordinates in the three-dimensional space coordinate system; determining the distance information of each light source from the ground based on each second coordinate; and determining that the light source is the headlight of the preceding vehicle if or for a light source whose distance information satisfies a second preset condition, including: if there is a light source in the multiple light source information whose distance information satisfies a preset distance range, representing that the distance information of the light source satisfies the second preset condition, then determining that the light source is the headlight of the preceding vehicle.

[0008] Optionally, the preset mapping relationship is expressed by the following formula:

[0009]

[0010] in, is the second coordinate, (p cx , p cy ) is the center point coordinate, cu, cv, fu, fv and D are the internal parameters of the camera that captured the image information, M cam2dev is the external parameter of the camera.

[0011] Optionally, the preset aspect ratio ranges from 1.1 to 2.5.

[0012] Optionally, the method for determining the position of the leading vehicle further includes: controlling driving according to the position of the leading vehicle.

[0013] According to a second aspect of an embodiment of the present disclosure, a device for determining the position of a leading vehicle is provided, the device comprising: a shooting module for obtaining image information including multiple light source information; a recognition module for recognizing the image information by means of a pre-trained recognition model, and obtaining a light source recognition result corresponding to each of the multiple light source information; an acquisition module for obtaining, based on the multiple light source recognition results, size information of each of the multiple light sources and distance information of each of the light sources from the ground; a headlight determination module for determining that a light source is a headlight of a leading vehicle if there is a light source among the multiple light sources whose size information satisfies a first preset condition and / or whose distance information satisfies a second preset condition; and a position determination module for determining the position of the leading vehicle based on the headlight of the leading vehicle.

[0014] According to a third aspect of an embodiment of the present disclosure, a vehicle is provided, comprising: a processor; and a memory for storing processor executable instructions; wherein the processor is configured to implement the steps of the method described in the first aspect when executed.

[0015] According to a fourth aspect of an embodiment of the present disclosure, a computer-readable storage medium is provided, on which computer program instructions are stored. When the program instructions are executed by a processor, the steps of the method provided in the first aspect of the present disclosure are implemented.

[0016] The present disclosure provides a method, device, vehicle and medium for determining the position of a leading vehicle. First, the leading vehicle with its lights on is photographed to obtain image information, wherein the image information includes information of multiple light sources such as interference light sources and headlights. Then, the image information is preliminarily recognized using a pre-trained recognition model to obtain a light source recognition result corresponding to each light source information. Since the interference light source in the image information may interfere with the recognition of the headlights, size information and distance information are obtained based on the light source recognition result corresponding to each light source information. After the light source recognition result is re-detected using the distance information and / or size information, the headlights of the leading vehicle are determined, thereby determining the position of the leading vehicle, thereby ensuring the accuracy of the headlight position recognition and thus ensuring the accuracy of the positioning of the leading vehicle.

[0017] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present disclosure and, together with the description, serve to explain the principles of the present disclosure.

[0019] Figure 1 is a flowchart of a method for determining a position of a preceding vehicle shown in an exemplary embodiment of the present disclosure;

[0020] Figure 2 is a block diagram of a device for determining a position of a preceding vehicle according to an exemplary embodiment;

[0021] Figure 3 The invention is a block diagram of a vehicle for a method for determining a position of a leading vehicle according to an exemplary embodiment. DETAILED DESCRIPTION

[0022] Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present disclosure. Instead, they are merely examples of devices and methods consistent with some aspects of the present disclosure as detailed in the appended claims.

[0023] Before introducing a method, device, vehicle and medium for determining the position of a leading vehicle provided by the present disclosure, the prior art involved in the present disclosure is first described. ADB senses the front vehicle lights through a light source detection model to obtain the position of the leading vehicle, and can accurately obtain the position of the leading vehicle to ensure the safety of subsequent driving. Due to the large number of light sources at night, street lights, indicator lights, traffic lights, reflections of water on the road, and other light sources may interfere with the light source detection model's recognition of vehicle lights, especially when other light sources are close to the leading vehicle, blur or glare may make it more difficult to distinguish between vehicle lights and other light sources, and it is more likely to cause misjudgment of vehicle lights, making it difficult to accurately locate the position of the leading vehicle.

[0024] At present, the main way to solve the above problems is to start from the data side, mining a large number of images of other light sources as negative samples to add to the light source detection model for training, reducing the confidence of the false detection target, thereby reducing the false detection rate of other light sources. However, this method will take a lot of time to collect and clean the negative sample data, and the light source detection model also needs to be re-trained iteratively, which will cost a lot of manpower and time. Especially when other light sources are close to the vehicle in front, the similarity between other light sources and car lights is difficult to eradicate. In this case, the trained light source detection model cannot achieve the expected effect. In addition, the parameters of the light source detection model trained through negative sample iterations become more, and deploying the light source detection model requires more storage resources.

[0025] To do this, see Figure 1 , Figure 1 is a flow chart of a method for determining the position of a preceding vehicle according to an exemplary embodiment of the present disclosure. The method for determining the position of a preceding vehicle can be applied to Figure 2 The preceding vehicle position determination device 100 shown, Figure 3 The vehicle 200 shown, Figure 3 The computing platform 250 and the computer readable storage medium in the present embodiment are used as an example. In order to correspond to the preceding vehicle, the vehicle mentioned in the present embodiment is the vehicle 200. Figure 1 The process shown is described in detail, and the method for determining the position of the preceding vehicle may specifically include the following steps:

[0026] Step S110: Obtain image information including multiple light source information.

[0027] Optionally, the preceding vehicle with its lights on is photographed to obtain image information containing multiple light source information. The preceding vehicle in this embodiment refers to the vehicle in front of the vehicle. The preceding vehicle may be a vehicle in front of the vehicle and traveling in the same direction as the vehicle. The preceding vehicle may also be a vehicle in front of the vehicle and traveling in the opposite direction to the vehicle.

[0028] During the driving process of this vehicle, a camera or a camera is installed on the front of this vehicle, and the camera or the camera is used to shoot the vehicle in front with the lights on to obtain image information. When this vehicle is driving at night, there are many light sources in the environment, so the image information obtained includes multiple light source information. The multiple light source information includes the lights of the vehicle in front. If there is a vehicle traveling in the same direction as the vehicle in front, the lights include the taillights of the vehicle in front; if there is a vehicle traveling in the opposite direction, the lights include the headlights of the vehicle in front. The multiple light source information also includes street lights, indicator lights, traffic lights, reflections from water on the road, reflections from reflective films, and other interfering light sources that interfere with the recognition of lights.

[0029] Optionally, the camera or camera in this embodiment can be a separate camera specially set for determining the position of the preceding vehicle, or a camera or camera on a driving recorder.

[0030] Step S120: Recognize the image information using a pre-trained recognition model to obtain a light source recognition result corresponding to each of the multiple light source information.

[0031] The recognition model is iteratively trained through paired training samples to obtain a pre-trained recognition model, wherein the paired training samples include a sample image containing a light source and a sample recognition result.

[0032] In one embodiment, the pre-trained recognition model is stored in a server that is connected to the vehicle. When image information needs to be recognized, the vehicle sends the captured image information to the server. The server recognizes the image information through the pre-trained recognition model and obtains a light source recognition result corresponding to each of the multiple light source information. In this embodiment, the pre-trained recognition model is stored in the server instead of the vehicle, which can reduce the storage resource usage of the vehicle.

[0033] In another embodiment, the pre-trained recognition model is stored locally in the vehicle, and the vehicle recognizes the captured image information through the local pre-trained recognition model to obtain a light source recognition result corresponding to each of the multiple light source information. Since the pre-trained recognition model is deployed locally in the vehicle, the pre-trained recognition model can be quickly called for recognition, and the recognition result can be quickly obtained.

[0034] Optionally, the light source recognition result corresponding to each light source information includes: the confidence corresponding to each light source information. The light source recognition result corresponding to each light source information also includes: the first coordinates in the image information of two vertices on a diagonal line of a rectangular detection box that selects each light source. It can be understood that in the process of recognition by a pre-trained recognition model, each light source is selected by a detection box. It should be understood that the size of the detection box of each light source among multiple light sources may not be the same, and the detection box is set according to the size of each light source, and the position of the detection box can be considered to be the position of the light source.

[0035] Step S130: According to the identification results of the multiple light sources, obtain the size information of each light source in the multiple light sources and the distance information of each light source from the ground.

[0036] In this embodiment, the size of each light source may include the aspect ratio of the detection frame where each light source is located, the shape of each light source, the length and width of each light source, the diameter of each light source, and the like.

[0037] The light source usually has a certain size, so different positions on the light source have different distances from the ground. Therefore, the distance from any point on the light source to the ground can be selected as distance information, and the distance information is used to represent the distance from the light source to the ground. In this embodiment, the position of the light source can be represented by selecting a detection frame of the light source, so the distance from any point on the detection frame to the ground can also be used as distance information, and the distance information is used to represent the distance from the light source to the ground.

[0038] Step S140: If there is a light source among the multiple light sources whose size information satisfies the first preset condition and / or whose distance information satisfies the second preset condition, it is determined that the light source is a headlight of the preceding vehicle.

[0039] Interference light sources may interfere with the recognition of vehicle lights, and therefore, the light source recognition result obtained by the pre-trained recognition model in step S120 may be inaccurate. Therefore, the light source recognition result is re-recognized by the first preset condition and / or the second preset condition to filter out the interference light sources in the light source recognition result. For example, if there is a light source whose size information satisfies the first preset condition and / or whose distance information satisfies the second preset condition among the multiple light sources, it is determined that the light source is the headlight of the vehicle in front. On the contrary, if there is a light source whose size information does not meet the first preset condition and / or whose distance information does not meet the second preset condition among the multiple light sources, it is determined that the light source is an interference light source.

[0040] The first preset condition may be set according to the size information of currently common vehicles, and the second preset condition may be set according to the distance between the headlights of currently common vehicles and the ground.

[0041] Step S150: determining the position of the preceding vehicle according to the headlights of the preceding vehicle.

[0042] The position of the vehicle in front can be inferred from the position of the headlights.

[0043] The method for determining the position of the preceding vehicle provided in this embodiment first obtains image information by photographing the preceding vehicle with its lights on, wherein the image information includes information of various light sources such as interference light sources and vehicle lights, and then performs preliminary recognition of the image information through a pre-trained recognition model to obtain a light source recognition result corresponding to each light source information. Since the interference light sources in the image information may interfere with the recognition of vehicle lights, the size information and distance information are obtained according to the light source recognition result corresponding to each light source information, and the light source recognition result is detected again with the distance information and / or the size information to determine the vehicle lights of the preceding vehicle, thereby determining the position of the preceding vehicle, ensuring the accuracy of the vehicle lights position recognition, and thus ensuring the accuracy of the preceding vehicle position positioning. Moreover, this embodiment further judges through the distance information and / or the size information, and does not need to mine negative sample data to train the recognition model, thus saving manpower and time costs, and will not increase the parameters of the model, thereby not increasing the memory occupied by the model. In addition, judging through the distance information and / or the size information can remove most of the interference results in the light source recognition results.

[0044] In one embodiment, the size information includes an aspect ratio, and the light source identification result corresponding to each light source information includes the first coordinates in the image information of two vertices on a diagonal line of a rectangular detection box that selects each light source in the image information. In step S130, the size information of each light source among the multiple light sources is obtained based on the multiple light source identification results, including: obtaining the length of the detection box and the width of the detection box based on the first coordinates of the two vertices, the length of the detection box being greater than the width of the detection box; for example, the first coordinates of the two vertices are (x1, y1) and (x2, y2), respectively, the length of the detection box can be |x1-x2|, and the width of the detection box can be |y1-y2|. The aspect ratio is obtained based on the ratio between the length and the width. For example, the aspect ratio can be

[0045] In another embodiment, in step S130, obtaining the distance information of each of the multiple light sources from the ground according to the multiple light source identification results includes:

[0046] First, according to the first coordinates of the two vertices corresponding to each of the multiple light sources, the center point coordinates of the center point of the detection frame of each light source are obtained, wherein the center point is the intersection of the two diagonals of the detection frame; for example, the first coordinates of the two vertices are (x1, y1) and (x2, y2), respectively, and the center point coordinates are

[0047] Then, according to a preset mapping relationship, each center point coordinate in the image information is mapped to a second coordinate in the three-dimensional space coordinate system corresponding to each center point coordinate, wherein the preset mapping relationship represents the correspondence between the coordinates in the image information and the coordinates in the three-dimensional space coordinate system.

[0048] The preset mapping relationship is expressed by the following formula:

[0049]

[0050] in, is the second coordinate, (p cx , p cy ) is the center point coordinate, cu, cv, fu, fv and D are the internal parameters of the camera that captured the image information, M cam2dev is the external parameter of the camera. for They are respectively the x-axis, y-axis and z-axis of the second coordinate in the three-dimensional space coordinate system.

[0051] Then, according to each second coordinate, the distance information of each light source from the ground is determined. Exemplarily, the coordinate point representing the above distance is directly read from the second coordinate. as distance information.

[0052] In one embodiment, after obtaining the aspect ratio, in step S140, if there is a light source whose size information satisfies the first preset condition among the multiple light sources, then the light source is determined to be the headlight of the vehicle in front, including: if there is a light source whose aspect ratio is within the preset aspect ratio range among the multiple light sources, and the size information representing the light source satisfies the first preset condition, then the light source is determined to be the headlight of the vehicle in front. If there is a light source whose aspect ratio is not within the preset aspect ratio range among the multiple light sources, the light source may be an interfering light source such as a street lamp, a puddle reflection, etc. that is close to the real headlight in the image information. Among them, the preset aspect ratio is set according to the actual aspect ratio of the headlight, for example, the preset aspect ratio range is 1.1 to 2.5.

[0053] In another embodiment, after obtaining the distance information, in step S140, if the distance information satisfies the second preset condition of the light source, then the light source is determined to be the headlight of the preceding vehicle, including: if there is a light source whose distance information satisfies the preset distance range among the multiple light source information, and the distance information representing the light source satisfies the second preset condition, then the light source is determined to be the headlight of the preceding vehicle. If there is a light source whose distance does not meet the preset distance range among the multiple light source information, then it is determined that the light source may be an interfering light source such as a street lamp, a puddle reflection, etc. that is close to the real headlight in the image information. Among them, the preset distance is set according to the distance between the headlight of the actual vehicle and the ground, and the preset distance range is 0.8m to 2m.

[0054] In another embodiment, after obtaining the aspect ratio and distance information, in step S140, if there is a light source among the multiple light sources whose size information satisfies the first preset condition, and / or whose distance information satisfies the second preset condition, then the light source is determined to be the headlight of the preceding vehicle, including: if there is a light source among the multiple light sources whose aspect ratio is within the preset aspect ratio range, indicating that the size information of the light source satisfies the first preset condition, and if there is a light source among the multiple light source information whose distance information satisfies the preset distance range, indicating that the distance information of the light source satisfies the second preset condition, then the light source is determined to be the headlight of the preceding vehicle. In this embodiment, both the aspect ratio and the distance information are used to further determine the position of the headlight, further improving the accuracy of the headlight determination.

[0055] After step S150, the method for determining the position of the preceding vehicle further includes: controlling driving according to the position of the preceding vehicle.

[0056] Control driving, for example, control the following speed of the vehicle. Or, during the driving process of the vehicle, control the brightness of some LEDs in the high beam matrix of the vehicle or turn off some LEDs to achieve the purpose of weakening the high beam of the vehicle, avoid dazzling the driver of the vehicle in front, and ensure traffic safety.

[0057] Based on the same concept, Figure 2 is a block diagram of a device for determining the position of a preceding vehicle according to an exemplary embodiment. Figure 2 The device 100 for determining the position of a preceding vehicle includes: a photographing module 110, a recognition module 120, an acquisition module 130, a vehicle light determination module 140 and a position determination module 150;

[0058] A shooting module 110 is used to obtain image information including information of various light sources;

[0059] The recognition module 120 is used to recognize the image information through a pre-trained recognition model to obtain a light source recognition result corresponding to each of the multiple light source information;

[0060] An acquisition module 130, configured to obtain size information of each of the multiple light sources and distance information of each of the multiple light sources from the ground according to the multiple light source identification results;

[0061] A vehicle light determination module 140, configured to determine that a light source is a vehicle light of a preceding vehicle if there is a light source among the plurality of light sources whose size information satisfies a first preset condition and / or whose distance information satisfies a second preset condition;

[0062] The position determination module 150 is used to determine the position of the preceding vehicle according to the headlights of the preceding vehicle.

[0063] Optionally, the light source recognition result corresponding to each light source information includes first coordinates of two vertices on a diagonal line of a rectangular detection box that selects each light source in the image information in the image information, and the acquisition module 130 includes: a length and width acquisition module and an aspect ratio acquisition module;

[0064] A length and width acquisition module, used to obtain the length of the detection frame and the width of the detection frame according to the first coordinates of the two vertices;

[0065] The aspect ratio acquisition module is used to obtain the aspect ratio according to the ratio between the length and the width.

[0066] Optionally, the vehicle light determination module 140 includes: a first vehicle light determination module;

[0067] The first vehicle light determining module is configured to determine that, if there is a light source with an aspect ratio within a preset aspect ratio range among the multiple light sources, and size information representing the light source meets a first preset condition, the light source is determined to be the vehicle light of the preceding vehicle.

[0068] Optionally, the acquisition module 130 includes: a center point coordinate acquisition module, a mapping module and a distance acquisition module;

[0069] A center point coordinate acquisition module, used to obtain the center point coordinates of the center point of the detection frame of each light source among the multiple light sources according to the first coordinates of the two vertices corresponding to each light source, wherein the center point is the intersection of two diagonals of the detection frame;

[0070] A mapping module, configured to map each center point coordinate in the image information to a second coordinate in a three-dimensional space coordinate system corresponding to each center point coordinate according to a preset mapping relationship, wherein the preset mapping relationship represents a correspondence between the coordinates in the image information and the coordinates in the three-dimensional space coordinate system;

[0071] A distance acquisition module, used to determine the distance information of each light source from the ground according to each second coordinate;

[0072] The vehicle light determination module 140 includes: a second vehicle light determination module;

[0073] The second vehicle light determination module is configured to determine that, if there is a light source whose distance information satisfies a preset distance range in the multiple light source information, and the distance information representing the light source satisfies a second preset condition, then determine that the light source is the vehicle light of the preceding vehicle.

[0074] The preset mapping relationship is expressed by the following formula:

[0075]

[0076] in, is the second coordinate, (p cx , p cy ) is the center point coordinate, cu, cv, fu, fv and D are the internal parameters of the camera that captured the image information, M cam2dev is the external parameter of the camera.

[0077] The preset aspect ratio range is 1.1 to 2.5.

[0078] Optionally, the preceding vehicle position determination device 100 further includes: a driving module;

[0079] The driving module is used to control the driving according to the position of the preceding vehicle.

[0080] Regarding the device in the above embodiment, the specific manner in which each module performs operations has been described in detail in the embodiment of the method, and will not be elaborated here.

[0081] The present disclosure also provides a computer-readable storage medium having computer program instructions stored thereon, and when the program instructions are executed by a processor, the steps of the method for determining the position of a preceding vehicle provided by the present disclosure are implemented.

[0082] Figure 3 2 is a block diagram of a vehicle for determining the position of a preceding vehicle according to an exemplary embodiment. For example, vehicle 200 may be a hybrid vehicle, a non-hybrid vehicle, an electric vehicle, a fuel cell vehicle, or other types of vehicles. Vehicle 200 may be an autonomous vehicle, a semi-autonomous vehicle, or a non-autonomous vehicle.

[0083] Reference Figure 3, the vehicle 200 may include various subsystems, such as an infotainment system 210, a perception system 220, a decision control system 230, a drive system 240, and a computing platform 250. The vehicle 200 may also include more or fewer subsystems, and each subsystem may include multiple components. In addition, each subsystem and each component of the vehicle 200 may be interconnected by wire or wireless means.

[0084] In some embodiments, the infotainment system 210 may include a communication system, an entertainment system, and a navigation system, among others.

[0085] The perception system 220 may include several sensors for sensing information about the environment around the vehicle 200. For example, the perception system 220 may include a global positioning system (the global positioning system may be a GPS system, or a Beidou system or other positioning systems), an inertial measurement unit (IMU), a laser radar, a millimeter wave radar, an ultrasonic radar, and a camera.

[0086] The decision control system 230 may include a computing system, a vehicle controller, a steering system, a throttle, and a braking system.

[0087] The drive system 240 may include components that provide power movement for the vehicle 200. In one embodiment, the drive system 240 may include an engine, an energy source, a transmission system, and wheels. The engine may be one or a combination of an internal combustion engine, an electric motor, and an air compression engine. The engine can convert energy provided by the energy source into mechanical energy.

[0088] Some or all functions of the vehicle 200 are controlled by a computing platform 250. The computing platform 250 may include at least one processor 251 and a memory 252, and the processor 251 may execute instructions 253 stored in the memory 252.

[0089] The processor 251 may be any conventional processor, such as a commercially available CPU. The processor may also include a graphics processor (Graphic Process Unit, GPU), a field programmable gate array (Field Programmable Gate Array, FPGA), a system on chip (System on Chip, SOC), an application specific integrated circuit (Application Specific Integrated Circuit, ASIC) or a combination thereof.

[0090] The memory 252 may be implemented by any type of volatile or nonvolatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk or optical disk.

[0091] In addition to instructions 253 , memory 252 may also store data, such as road maps, route information, and data such as the location, direction, and speed of the vehicle. The data stored in memory 252 may be used by computing platform 250 .

[0092] In the disclosed embodiment, the processor 251 may execute the instruction 253 to complete all or part of the steps of the preceding vehicle position determination method described above.

[0093] In summary, the method, device, vehicle and medium for determining the position of the leading vehicle provided by the present invention first photographs the leading vehicle with its lights on to obtain image information, wherein the image information includes information of various light sources such as interference light sources and headlights, and then uses a pre-trained recognition model to perform preliminary recognition of the image information to obtain a light source recognition result corresponding to each light source information. Since the interference light sources present in the image information may interfere with the recognition of the headlights, size information and distance information are obtained based on the light source recognition result corresponding to each light source information. After the light source recognition result is re-detected with the distance information and / or size information, the headlights of the leading vehicle are determined, thereby determining the position of the leading vehicle, ensuring the accuracy of the headlight position recognition, and thus ensuring the accuracy of the positioning of the leading vehicle.

[0094] In another exemplary embodiment, a computer program product is also provided. The computer program product includes a computer program executable by a programmable device, and the computer program has a code portion for executing the above-mentioned preceding vehicle position determination method when executed by the programmable device.

[0095] Those skilled in the art will readily appreciate other embodiments of the present disclosure after considering the specification and practicing the present disclosure. This application is intended to cover any variations, uses or adaptations of the present disclosure, which follow the general principles of the present disclosure and include common knowledge or customary technical means in the art that are not disclosed in the present disclosure. The specification and examples are to be considered as exemplary only, and the true scope and spirit of the present disclosure are indicated by the following claims.

[0096] It should be understood that the present disclosure is not limited to the exact structures that have been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present disclosure is limited only by the appended claims.

Claims

1. A method for determining the position of a preceding vehicle, characterized in that: The method comprises: Obtaining image information including multiple light source information; Recognize the image information through a pre-trained recognition model to obtain a light source recognition result corresponding to each of the multiple light source information; According to the identification results of the multiple light sources, the size information of each of the multiple light sources and the distance information of each of the light sources from the ground are obtained; If there is a light source among the multiple light sources whose size information satisfies the first preset condition, and / or whose distance information satisfies the second preset condition, it is determined that the light source is a headlight of the preceding vehicle; The position of the preceding vehicle is determined according to the headlights of the preceding vehicle.

2. The method according to claim 1, characterized in that The light source recognition result corresponding to each light source information includes first coordinates in the image information of two vertices on a diagonal line of a rectangular detection box of each light source selected in the image information, and obtaining size information of each light source in the multiple light sources according to the multiple light source recognition results includes: Obtaining the length of the detection frame and the width of the detection frame according to the first coordinates of the two vertices; According to the ratio between the length and the width, an aspect ratio is obtained.

3. The method according to claim 2, characterized in that If there is a light source whose size information satisfies the first preset condition among the multiple light sources, determining that the light source is the headlight of the preceding vehicle includes: If there is a light source whose aspect ratio is within a preset aspect ratio range among the multiple light sources, and the size information representing the light source meets a first preset condition, it is determined that the light source is a headlight of the preceding vehicle.

4. The method according to claim 2 or 3, characterized in that: The step of obtaining distance information of each of the multiple light sources from the ground according to the multiple light source identification results includes: According to the first coordinates of the two vertices corresponding to each of the multiple light sources, obtaining the center point coordinates of the center point of the detection frame of each light source, wherein the center point is the intersection point of two diagonal lines of the detection frame; According to a preset mapping relationship, mapping each center point coordinate in the image information to a second coordinate in the three-dimensional space coordinate system corresponding to each center point coordinate, wherein the preset mapping relationship represents a correspondence between the coordinates in the image information and the coordinates in the three-dimensional space coordinate system; Determine the distance information of each light source from the ground according to each second coordinate; If the distance information satisfies the light source of the second preset condition, determining that the light source is the headlight of the preceding vehicle includes: If there is a light source whose distance information satisfies a preset distance range in the multiple types of light source information, and the distance information representing the light source satisfies a second preset condition, it is determined that the light source is a headlight of the preceding vehicle.

5. The method according to claim 4, characterized in that The preset mapping relationship is expressed by the following formula: in, is the second coordinate, (p cx , p cy ) is the center point coordinate, cu, cv, fu, fv and D are the internal parameters of the camera that captured the image information, M cam2dev is the external parameter of the camera.

6. The method according to claim 3, characterized in that The preset aspect ratio range is 1.1 to 2.

5.

7. The method according to any one of claims 1 to 3, characterized in that: The method further comprises: Driving is controlled according to the position of the preceding vehicle.

8. A device for determining the position of a preceding vehicle, characterized in that: include: A shooting module, used to obtain image information including information of multiple light sources; A recognition module, used to recognize the image information through a pre-trained recognition model to obtain a light source recognition result corresponding to each of the multiple light source information; An acquisition module, used for obtaining size information of each of the multiple light sources and distance information of each of the multiple light sources from the ground according to the multiple light source identification results; A vehicle light determination module, configured to determine that, if there is a light source among the multiple light sources whose size information satisfies a first preset condition and / or whose distance information satisfies a second preset condition, the light source is a vehicle light of a preceding vehicle; The position determination module is used to determine the position of the preceding vehicle according to the headlights of the preceding vehicle.

9. A vehicle, characterized in that: include: processor; a memory for storing processor-executable instructions; The processor is configured to implement the steps of any one of the methods of claims 1 to 7 when executed.

10. A computer-readable storage medium having computer program instructions stored thereon, characterized in that: When the program instructions are executed by a processor, the steps of the method described in any one of claims 1 to 7 are implemented.