Traffic signal information acquisition method and related equipment

Abstract

The present invention discloses a method for obtaining traffic signal information and related equipment. The method includes: obtaining a first score for all traffic signal light groups based on intersection image information and an intelligent recognition algorithm, wherein the first score is determined based on the degree of matching between the intelligent recognition algorithm and the traffic light feature information; determining a second score based on the distance between the image position of the traffic signal light group and a preset position, wherein the preset position is the position where the target traffic signal light group is most likely to appear based on feature training; determining the traffic signal light group with the largest product of the first score and the second score as the target traffic signal light group; and identifying the color information and shape information of the target traffic signal light group to obtain traffic signal information at the current intersection. The method proposed in this application uses deep learning traffic signal light group detection as an aid to complete traffic signal light matching, saves computing power, is quick to build, has low cost, high accuracy, and sustainable development and optimization.

Description

Technical Field

[0001] This specification relates to the field of autonomous driving, and more specifically, the present invention relates to a method for obtaining traffic signal information and related equipment. Background Art

[0002] With the advancement of autonomous driving, traffic light recognition at urban intersections has become a pressing technical challenge. Current methods for traffic light recognition often use images of the entire intersection captured by a vehicle. However, this method's accuracy decreases when there are multiple lights at an intersection, or when two intersections are very close together and multiple lights appear in a single image. Furthermore, intelligent recognition using normal images consumes a significant amount of computation and significantly reduces computational speed. Other traffic light information acquisition methods require precise GPS positioning and high-precision maps to provide accurate traffic light coordinates. Furthermore, the mapping from real-world coordinates to coordinates captured by the front-facing camera is crucial. This method has two drawbacks: first, the equipment relies heavily on GPS and high-precision maps for location information, which is costly; second, it requires extremely accurate location data from GPS and high-precision maps. Furthermore, the calibration error of the on-board front-facing camera must be extremely small, making data collection and camera calibration difficult in real-world scenarios. Summary of the Invention

[0003] The Summary of the Invention introduces a series of simplified concepts that will be further described in the Detailed Description of the Invention. The Summary of the Invention is not intended to limit the key features and essential features of the claimed technical solution, nor is it intended to determine the scope of protection of the claimed technical solution.

[0004] In order to provide a fast and accurate method for obtaining traffic signal information, in a first aspect, the present invention proposes a method for obtaining traffic signal information, the method comprising:

[0005] Obtaining first scores for all traffic light groups based on the intersection image information and the intelligent recognition algorithm, wherein the first scores are determined based on a degree of matching between the intelligent recognition algorithm and the traffic light feature information;

[0006] Determining a second score based on the distance between the image position of the traffic light group and a preset position, wherein the preset position is the position where the target traffic light group is most likely to appear based on feature training;

[0007] determining the traffic light group having the largest product of the first score and the second score as the target traffic light group;

[0008] The color information and shape information of the target traffic light group are identified to obtain the traffic signal information of the current intersection.

[0009] Optionally, the above-mentioned obtaining of the first scores of all traffic light groups based on the intersection image information and the intelligent recognition algorithm includes:

[0010] The confidence of all traffic light groups using the target detection model YOLOX based on the intersection image information is used as the first score.

[0011] Optionally, determining the second score based on the distance between the image position of the traffic light group and a preset position includes:

[0012] The second score is determined according to the distance between the image position of the traffic light group having the first score greater than the preset score and the preset position.

[0013] Optionally, the above also includes:

[0014] Obtaining type information of the above-mentioned signal light group, wherein the above-mentioned traffic light group includes a cantilever signal light group and a column signal light group;

[0015] The method of determining the second score based on the distance between the image position of the traffic light group and the preset position includes:

[0016] When the traffic light group is a cantilever traffic light group, determining the second score according to the distance between the image position of the traffic light group and the preset position and the first weight coefficient;

[0017] and / or,

[0018] When the traffic light group is a column type traffic light group, the second score is determined according to the distance between the image position of the traffic light group and the preset position and the second weight coefficient, wherein the first weight coefficient is greater than the second weight coefficient.

[0019] Optionally, the preset position includes a preset center and a preset radius;

[0020] The method of determining the second score based on the distance between the image position of the traffic light group and the preset position includes:

[0021] Determine distance information based on the traffic light group image position and the preset center;

[0022] The second score is determined according to a ratio of the distance information to the preset radius.

[0023] Optionally, the above method further includes:

[0024] Control the vehicle camera to obtain vehicle driving image information;

[0025] The intersection image information is obtained by cropping the upper portion of the vehicle driving image information according to a first preset ratio and the lower portion of the vehicle driving image information according to a second preset ratio, wherein the first preset ratio is smaller than the second preset ratio.

[0026] Optionally, the above method further includes:

[0027] Obtain the lane line position information corresponding to the target vehicle;

[0028] Determine a position correction coefficient based on the lane line position information;

[0029] The method of determining the second score based on the distance between the image position of the traffic light group and the preset position includes:

[0030] A second score is determined according to the position correction coefficient and the distance between the image position of the traffic light group and the preset position.

[0031] In a second aspect, the present invention further provides a traffic signal information acquisition device, comprising:

[0032] a first obtaining unit, configured to obtain first scores of all traffic light groups based on the intersection image information and the intelligent recognition algorithm, wherein the first scores are determined based on a degree of matching between the intelligent recognition algorithm and the traffic light feature information;

[0033] a first determining unit, configured to determine a second score based on a distance between the image position of the traffic light group and a preset position, wherein the preset position is a position where the target traffic light group is most likely to appear based on feature training;

[0034] The second determination unit is used to determine the traffic light group with the largest product of the first score and the second score as the target traffic light group;

[0035] The second acquiring unit is configured to identify the color information and shape information of the target traffic light group to acquire the traffic signal information of the current intersection.

[0036] In a third aspect, an electronic device comprises: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor is configured to implement the steps of the traffic signal information acquisition method of any one of the first aspects described above when executing the computer program stored in the memory.

[0037] In a fourth aspect, the present invention further proposes a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the method for obtaining traffic signal information according to any one of the above items in the first aspect is implemented.

[0038] In summary, the traffic signal information acquisition method of the embodiment of the present application includes: obtaining a first score for all traffic signal light groups based on intersection image information and an intelligent recognition algorithm, wherein the first score is determined based on the degree of match between the intelligent recognition algorithm and traffic signal feature information; determining a second score based on the distance between the image position of the traffic signal light group and a preset position, wherein the preset position is the position where the target traffic signal light group is most likely to appear based on feature training; determining the traffic signal light group with the largest product of the first score and the second score as the target traffic signal light group; and identifying the color information and shape information of the target traffic signal light group to obtain traffic signal information for the current intersection. The traffic signal information acquisition method provided in the embodiment of the present application uses an intelligent recognition algorithm on the intersection image information to first obtain a first score, obtain the degree of match between the image and the signal light group, and then obtain a second score to obtain the degree of match between the signal light group in the image and the target signal light group. The traffic signal light group with the largest product of the first score and the second score is determined as the target traffic signal light group, and the color information and shape information of the target traffic signal light group are identified to obtain traffic signal information for the current intersection. The method proposed in this application does not require GPS and high-precision maps for traffic light matching. Instead, it uses deep learning traffic light group detection as an aid to complete traffic light matching, saving computing power, and is quick to build, low-cost, high-precision, and sustainable development and optimization.

[0039] The traffic signal information acquisition method of the present invention, and other advantages, objectives and features of the present invention will be partially reflected in the following description, and will also be understood by those skilled in the art through research and practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Various other advantages and benefits will become apparent to those skilled in the art upon reading the detailed description of the preferred embodiment below. The accompanying drawings are for illustration purposes only and are not to be considered as limiting the present description. The same reference symbols are used throughout the drawings to represent the same components. In the drawings:

[0041] Figure 1 A flow chart of a method for obtaining traffic signal information provided in an embodiment of the present application;

[0042] Figure 2 A schematic diagram of a scenario for obtaining traffic signal information at an intersection provided in an embodiment of the present application;

[0043] Figure 3 A schematic diagram of the structure of a traffic signal information acquisition device provided in an embodiment of the present application;

[0044] Figure 4 A schematic diagram of the structure of an electronic device for acquiring traffic signal information provided in an embodiment of the present application. DETAILED DESCRIPTION

[0045] The traffic signal information acquisition method provided in the embodiment of the present application first obtains a first score for the intersection image information using an intelligent recognition algorithm to determine the degree of match between the image and the signal light group. A second score is then obtained to determine the degree of match between the signal light group in the image and the target signal light group. The traffic signal light group with the largest product of the first and second scores is used as the target traffic signal light group. The color and shape information of the target traffic signal light group are then identified to obtain the traffic signal information of the current intersection. The method proposed in the present application does not require GPS and high-precision maps for traffic signal light matching. Instead, it uses deep learning traffic signal light group detection as an aid to complete traffic signal light matching, saving computing power, and is quick to build, low-cost, high-precision, and sustainable for development and optimization.

[0046] The terms "first", "second", "third", "fourth", etc. (if any) in the specification and claims of this application and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments described herein can be implemented in a sequence other than that illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or that are inherent to these processes, methods, products or devices. The technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the embodiments described are only part of the embodiments of the present application, not all of the embodiments.

[0047] See also Figure 1 , is a flow chart of a method for obtaining traffic signal information provided in an embodiment of the present application, which may specifically include:

[0048] S110. Obtaining first scores for all traffic light groups based on the intersection image information and the intelligent recognition algorithm, wherein the first scores are determined based on a degree of matching between the intelligent recognition algorithm and the traffic light feature information;

[0049] Exemplarily, the intersection image information may be obtained by a camera mounted on the front bumper of the vehicle or a camera mounted on the upper end of the windshield of the vehicle. After obtaining the intersection line information, the vehicle side uses an intelligent recognition algorithm to calculate the first score of all traffic light groups in the intersection image information. The first score is the degree of matching between the image elements in the intersection image and the traffic light characteristic information calculated according to the intelligent recognition algorithm. The higher the degree of matching, the higher the score, that is, the greater the possibility that the image element is a traffic light group.

[0050] S120, determining a second score based on the distance between the image position of the traffic light group and a preset position, wherein the preset position is a position where the target traffic light group is most likely to appear based on feature training;

[0051] For example, the second score is determined by obtaining the distance between the image position of the traffic light group and a preset position. The closer the distance between the image position of the traffic light group and the preset position, the higher the second score, indicating a greater likelihood that the traffic light group is the target traffic light group to be observed for the current direction of travel at the current intersection. It should be noted that the preset position is the location where the target traffic light group is most likely to appear, determined through feature training based on a large number of data samples.

[0052] S130: Determine the traffic light group with the largest product of the first score and the second score as the target traffic light group;

[0053] Exemplarily, the traffic light group with the largest product of the first score and the second score is determined as the target traffic light group. The first score represents the degree of match between the image and the real traffic light, and the second score represents the degree of match between the traffic lights at different positions and the most likely positions of the target traffic light. Therefore, the product of the two can well determine which traffic light image in the image is the target traffic light group.

[0054] S140: Identify the color information and shape information of the target traffic signal light group to obtain traffic signal information of the current intersection.

[0055] For example, after identifying the target traffic light group, the color confidence and shape information of the target traffic light group are determined. The color information includes: red, yellow and green, and the shape information includes round lights, arrow lights and countdown numbers, etc. The combined information of color and shape can accurately determine the traffic signal information at the current intersection, thereby controlling the safe driving of vehicles.

[0056] In summary, the traffic signal information acquisition method provided in the embodiment of the present application uses an intelligent recognition algorithm to first obtain a first score for the intersection image information, and obtain the degree of matching with the signal light group in the image. Then, a second score is obtained to obtain the degree of matching between the signal light group in the image and the target signal light group. The traffic signal light group with the largest product of the first score and the second score is used as the target traffic signal light group. The color information and shape information of the target traffic signal light group are identified to obtain the traffic signal information of the current intersection. The method proposed in this application does not require GPS and high-precision maps for traffic signal light matching. Instead, it uses deep learning traffic signal light group detection as an aid to complete traffic signal light matching, saving the equipment cost of GPS and high-precision maps and the cost of collecting and drawing high-precision maps. It is quick to build, low-cost, high-precision, and sustainable for development and optimization.

[0057] In some examples, the above-mentioned method of obtaining the first scores of all traffic light groups based on the intersection image information and the intelligent recognition algorithm includes:

[0058] The confidence of all traffic light groups using the target detection model YOLOX based on the intersection image information is used as the first score.

[0059] For example, using YOLOX can reduce computational complexity, eliminate IoU calculations, and generate fewer prediction boxes. This alleviates the imbalance between positive and negative samples. The prediction boxes for the anchor-free method are only one-third of those for the anchor-based method, and most of these prediction boxes are negative samples. Therefore, the anchor-free method can reduce the number of negative samples, further alleviating the imbalance between positive and negative samples. It also avoids anchor parameter adjustment: The scale of the anchor box in the anchor-based method is a hyperparameter, and different hyperparameter settings can affect model performance. The anchor-free method avoids this.

[0060] In summary, the traffic signal information acquisition method provided in the embodiment of the present application uses the YOLOX model to quickly obtain the first scores of all traffic light groups in the intersection image information.

[0061] In some examples, determining the second score based on the distance between the image position of the traffic light group and a preset position includes:

[0062] The second score is determined according to the distance between the image position of the traffic light group having the first score greater than the preset score and the preset position.

[0063] For example, after obtaining the first score, the traffic light groups in the image are first screened. A preset score can be set, such as 0.8, and all traffic light groups with a score lower than or equal to 0.8 are eliminated. The eliminated traffic light groups have a low degree of match with the real traffic light groups, which may be the brakes of the vehicle in front, etc. The eliminated traffic light groups do not participate in the calculation process of the second score, thereby saving a lot of computing power and accelerating the recognition process of traffic light group information.

[0064] In summary, the traffic signal information acquisition method provided in the embodiment of the present application first filters the image of the traffic light group through the first score and the preset score. The eliminated traffic light groups do not participate in the calculation process of the second score, thereby saving a lot of computing power and accelerating the recognition process of the traffic light group information.

[0065] In some examples, the above also includes:

[0066] Obtaining type information of the above-mentioned signal light group, wherein the above-mentioned traffic light group includes a cantilever signal light group and a column signal light group;

[0067] The method of determining the second score based on the distance between the image position of the traffic light group and the preset position includes:

[0068] When the traffic light group is a cantilever traffic light group, determining the second score according to the distance between the image position of the traffic light group and the preset position and the first weight coefficient;

[0069] and / or,

[0070] When the traffic light group is a column type traffic light group, the second score is determined according to the distance between the image position of the traffic light group and the preset position and the second weight coefficient, wherein the first weight coefficient is greater than the second weight coefficient.

[0071] For example, traffic signal light groups can be divided into cantilever signal light groups and column signal light groups. Some intersections have both cantilever signal light groups and column signal light groups, such as Figure 2 As shown, in this case, the cantilever signal light group is the signal light group that provides information to vehicles, and the column signal light group is the signal light group that provides signals to pedestrians. Some intersections only have column signal light groups (such as T-shaped intersections). In this embodiment of the application, different weight coefficients are matched to different types of traffic lights to calculate the second score. The cantilever signal light group can be set with a first weight coefficient of 1.1, and the column signal light group can be set with a second weight coefficient of 0.8. By setting different weight coefficients, the current intersection information reflected by different types of traffic light groups can be well characterized.

[0072] In some examples, the preset position includes a preset center and a preset radius;

[0073] The method of determining the second score based on the distance between the image position of the traffic light group and the preset position includes:

[0074] Determine distance information based on the traffic light group image position and the preset center;

[0075] The second score is determined according to a ratio of the distance information to the preset radius.

[0076] Exemplarily, the preset position includes a preset center and a preset radius. For example, in image information of a certain intersection, the coordinates of the preset center are P_center(center_x, center_y), the coordinates of the center points of different light group images are P(x, y), and the distance information between the position of the traffic light group line and the preset center is: The first weight coefficient is weight_overhang = 1.1, and the second weight coefficient is weight_column = 0.8. The area within the preset radius of the preset center corresponds to the area where the target traffic light is most likely to appear.

[0077] Therefore, the product of the first score and the second score of the cantilever signal light group score_cantilever is:

[0078] score_overhang = conf_bbox * weight_overhang * (distance / r * 5%) (1)

[0079] The product of the first score and the second score of the column signal light group, score_column, is:

[0080] score_column=conf_bbox*weight_column*(distance / r*5%) (2)

[0081] In the above two formulas, conf_bbox is the first score.

[0082] In summary, the traffic signal information acquisition method provided in the embodiment of the present application can well characterize the matching degree between different types of signal light groups and the target light group through formula (1) and formula (2), and the signal information of the intersection obtained by this method is more accurate.

[0083] In some examples, the method further includes:

[0084] Control the vehicle camera to obtain vehicle driving image information;

[0085] The intersection image information is obtained by cropping the upper portion of the vehicle driving image information according to a first preset ratio and the lower portion of the vehicle driving image information according to a second preset ratio, wherein the first preset ratio is smaller than the second preset ratio.

[0086] For example, there is a lot of useless information in the vehicle driving image information obtained by the vehicle's camera. In order to improve the speed of traffic light recognition, the vehicle driving image information needs to be cropped. Because the traffic light is located at a higher position, there is more useless information in the lower part of the image. Therefore, the second preset ratio for cropping the lower part is higher than the first preset ratio for cropping the upper part. For example, the first preset ratio can be 10%, and the second preset ratio can be 20%.

[0087] In summary, the traffic signal information acquisition method provided in the embodiment of the present application uses the first preset ratio and the second preset ratio to crop the vehicle driving image information, which can filter out most of the invalid information and improve the calculation and recognition speed.

[0088] In some examples, the method further includes:

[0089] Obtain the lane line position information corresponding to the target vehicle;

[0090] Determine a position correction coefficient based on the lane line position information;

[0091] The method of determining the second score based on the distance between the image position of the traffic light group and the preset position includes:

[0092] A second score is determined according to the position correction coefficient and the distance between the image position of the traffic light group and the preset position.

[0093] For example, when the vehicle is in different lane line positions, due to the different angles between the vehicle and the target traffic light group, the actual position of the target traffic light in the acquired image will be different because the vehicle is in different lane lines. The vehicle obtains the vehicle's precise position, that is, the lane line position information, through camera image information and / or high-precision map information, determines the position correction coefficient based on different lane line position information, and makes corrections to the position where the target traffic light is most likely to appear in the image based on the position correction coefficient and the preset position. The second score is calculated using the corrected position and the image position of each traffic light group in the image. The obtained second score can be matched with the current position of the vehicle, and the obtained traffic information recognition result is more accurate.

[0094] In summary, the traffic signal information acquisition method provided in the embodiment of the present application determines the position correction coefficient for the vehicle in different lane positions, and uses the corrected position and the image position of each traffic light group in the image to calculate the second score. The obtained second score can match the current position of the vehicle, and the obtained traffic information recognition result is more accurate.

[0095] See also Figure 3An embodiment of the traffic signal information acquisition device in the embodiment of the present application may include:

[0096] The first acquisition unit 21 is used to obtain the first score of all traffic light groups based on the intersection image information and the intelligent recognition algorithm, wherein the first score is determined based on the matching degree between the intelligent recognition algorithm and the traffic light feature information.

[0097] a first determining unit 22 for determining a second score based on a distance between the image position of the traffic light group and a preset position, wherein the preset position is a position where the target traffic light group is most likely to appear based on feature training;

[0098] The second determination unit 23 is used to determine the traffic light group with the largest product of the first score and the second score as the target traffic light group;

[0099] The second acquiring unit 24 is configured to identify the color information and shape information of the target traffic light group to acquire the traffic signal information of the current intersection.

[0100] like Figure 4 As shown, an embodiment of the present application further provides an electronic device 300, comprising a memory 310, a processor 320, and a computer program 311 stored on the memory 310 and executable on the processor. When the processor 320 executes the computer program 311, the steps of any of the above-mentioned methods for acquiring traffic signal information are implemented.

[0101] Since the electronic device introduced in this embodiment is a device used to implement a traffic signal information acquisition device in the embodiment of the present application, based on the method introduced in the embodiment of the present application, technical personnel in this field can understand the specific implementation of the electronic device of this embodiment and its various variations. Therefore, how the electronic device implements the method in the embodiment of the present application is no longer introduced in detail here. As long as the equipment used by technical personnel in this field to implement the method in the embodiment of the present application falls within the scope of protection of this application.

[0102] In the specific implementation process, the computer program 311 can be implemented when executed by the processor Figure 1 Any implementation manner in the corresponding embodiments.

[0103] It should be noted that, in the above embodiments, the description of each embodiment has its own focus. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant description of other embodiments.

[0104] Those skilled in the art will appreciate that the embodiments of the present application can be provided as methods, systems, or computer program products. Therefore, the present application can adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0105] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, 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, an embedded computer, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0106] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0107] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0108] The present application also provides a computer program product, which includes computer software instructions. When the computer software instructions are executed on a processing device, the processing device is caused to execute the following Figure 1 The process of obtaining traffic signal information in the corresponding embodiment.

[0109] A computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function according to the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website, computer, server or data center to another website, computer, server or data center via a wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) method. The computer-readable storage medium can be any available medium that a computer can store or a data storage device such as a server or data center that includes one or more available media integrated. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid state drive (SSD)).

[0110] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0111] In the several embodiments provided in this application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are merely schematic. For example, the division of units is only a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interface, device or unit, which can be electrical, mechanical or other forms.

[0112] Units described as separate components may or may not be physically separate, and components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0113] In addition, the functional units in the various embodiments of the present application may be integrated into a single processing unit, or each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.

[0114] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the various embodiments of the present application. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0115] The above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the above embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the above embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for obtaining traffic signal information, characterized in that: include: Obtaining first scores for all traffic light groups based on the intersection image information and the intelligent recognition algorithm, wherein the first scores are determined based on a degree of matching between the intelligent recognition algorithm and the traffic light feature information; Acquiring type information of the traffic signal light group, wherein the traffic signal light group includes a cantilever signal light group and a column signal light group; Determining a second score based on a distance between the image position of the traffic light group and a preset position, wherein the preset position is a position where the target traffic light group is most likely to appear based on feature training; the preset position includes a preset center and a preset radius; When the traffic light group is a cantilever traffic light group, determining a second score according to a distance between the image position of the traffic light group and a preset position and a first weight coefficient; When the traffic light group is a column-type traffic light group, determining a second score according to a distance between the image position of the traffic light group and a preset position and a second weight coefficient, wherein the first weight coefficient is greater than the second weight coefficient; determining distance information based on the image position of the traffic light group and the preset center; determining the second score according to a ratio of the distance information to the preset radius; determining a traffic light group having a maximum product of the first score and the second score as a target traffic light group; The color information and shape information of the target traffic light group are identified to obtain the traffic signal information of the current intersection.

2. The method according to claim 1, wherein The method of obtaining the first scores of all traffic light groups based on the intersection image information and the intelligent recognition algorithm includes: The confidence of all traffic light groups using the target detection model YOLOX based on the intersection image information is used as the first score.

3. The method according to claim 1, wherein The determining a second score according to the distance between the image position of the traffic light group and a preset position includes: The second score is determined according to the distance between the image position of the traffic light group having the first score greater than the preset score and the preset position.

4. The method according to claim 1, wherein Also includes: Control the vehicle camera to obtain vehicle driving image information; The intersection image information is obtained by cropping an upper portion of the vehicle driving image information according to a first preset ratio and a lower portion of the vehicle driving image information according to a second preset ratio, wherein the first preset ratio is smaller than the second preset ratio.

5. The method according to claim 1, wherein Also includes: Obtain the lane line position information corresponding to the target vehicle; Determining a position correction coefficient based on the lane line position information; The determining a second score according to the distance between the image position of the traffic light group and a preset position includes: A second score is determined according to the position correction coefficient and the distance between the image position of the traffic light group and the preset position.

6. A traffic signal information acquisition device, characterized in that: include: a first acquiring unit, configured to acquire first scores of all traffic light groups based on the intersection image information and the intelligent recognition algorithm, wherein the first scores are determined based on a degree of matching between the intelligent recognition algorithm and the traffic light feature information; Acquiring type information of the traffic signal light group, wherein the traffic signal light group includes a cantilever signal light group and a column signal light group; a first determining unit, configured to determine a second score based on a distance between the image position of the traffic light group and a preset position, wherein the preset position is a position where the target traffic light group is most likely to appear based on feature training; the preset position includes a preset center and a preset radius; When the traffic light group is a cantilever traffic light group, determining a second score according to a distance between the image position of the traffic light group and a preset position and a first weight coefficient; When the traffic light group is a column-type traffic light group, determining a second score according to a distance between the image position of the traffic light group and a preset position and a second weight coefficient, wherein the first weight coefficient is greater than the second weight coefficient; determining distance information based on the image position of the traffic light group and the preset center; determining the second score according to a ratio of the distance information to the preset radius; a second determining unit, configured to determine a traffic light group having a maximum product of the first score and the second score as a target traffic light group; The second acquiring unit is configured to identify the color information and shape information of the target traffic light group to acquire the traffic signal information of the current intersection.

7. An electronic device comprising: A memory and a processor, wherein the processor is configured to implement the steps of the traffic signal information acquisition method according to any one of claims 1 to 5 when executing a computer program stored in the memory.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the traffic signal information acquisition method according to any one of claims 1 to 5 is implemented.

Images