Vehicle positioning method and device, electronic equipment and computer readable storage medium

Abstract

The application discloses a vehicle positioning method and device, electronic equipment and a computer readable storage medium. The method comprises the following steps: acquiring a road image collected by a roadside camera, performing vehicle target detection on the road image by using a first preset target detection model; generating a vehicle target image according to the vehicle target detection result and the road image, and determining the size transformation relationship between the vehicle target image and the road image; determining the vehicle target grounding point information in the vehicle target image according to the vehicle target image; and determining the absolute position of the vehicle target according to the vehicle target grounding point information and the size transformation relationship between the vehicle target image and the road image. The application processes the existing 2D target detection result, thereby obtaining a more accurate grounding point position of the vehicle target in the image, improving the vehicle positioning accuracy realized based on the 2D target detection, and increasing the processing logic without damaging and discarding the original 2D target detection data and processing flow, and the applicability is stronger.

Description

Technical Field

[0001] This application relates to the field of vehicle positioning technology, and in particular to a vehicle positioning method, device, electronic device, and computer-readable storage medium. Background Technology

[0002] Currently, the main technology for roadside object detection is 2D object detection technology based on deep learning. Although 3D object detection technology is also used, 2D object detection technology is relatively mature, data is easy to label, and 2D object detection usually has low computational cost, making it easy to deploy and run on roadside computing equipment.

[0003] Since the ultimate goal of roadside vehicle target detection is to output the vehicle target's location information in the world coordinate system, if 2D target detection is used, a point of the 2D bounding box in the detected image is usually transformed to the world coordinate system according to a pre-marked coordinate system transformation relationship. This coordinate system transformation relationship is the transformation relationship between the image coordinate system and the world ground coordinate system. However, because the roadside camera equipment is fixed, the bounding box obtained from vehicle target detection is often a bounding box containing the maximum edge of the vehicle target, making it impossible to directly determine which point is the grounding point of the vehicle target. Figure 1 As shown, a schematic diagram of a vehicle target detection result in the prior art is provided.

[0004] The existing solution is to directly transform the position of the bottom center point of the rectangle in the image to the world coordinate system as the final positioning result of the vehicle target. However, when the vehicle target is not in the middle area of ​​the image captured by the roadside camera, but on the side of the image such as the edge lane, the position obtained by transforming the bottom center point of the rectangle to the world coordinate system will have a large deviation from the actual vehicle target position, thus affecting the final vehicle positioning accuracy. Summary of the Invention

[0005] This application provides a vehicle positioning method, apparatus, electronic device, and computer-readable storage medium to improve the positioning accuracy of vehicle positioning based on image detection.

[0006] The embodiments of this application adopt the following technical solutions:

[0007] In a first aspect, embodiments of this application provide a vehicle positioning method, wherein the method includes:

[0008] Acquire road images captured by roadside cameras, and use a first preset target detection model to perform vehicle target detection on the road images to obtain vehicle target detection results;

[0009] A vehicle target image is generated based on the vehicle target detection results and the road image, and the size transformation relationship between the vehicle target image and the road image is determined.

[0010] Determine the vehicle target grounding point information in the vehicle target image based on the vehicle target image;

[0011] The absolute position of the vehicle target is determined based on the vehicle target ground point information in the vehicle target image and the size transformation relationship between the vehicle target image and the road image.

[0012] Optionally, the step of generating a vehicle target image based on the vehicle target detection result and the road image, and determining the size transformation relationship between the vehicle target image and the road image, includes:

[0013] The road image is cropped based on the vehicle target detection results to obtain the vehicle target image;

[0014] Based on the size of the vehicle target image and the size of the road image, determine the size transformation relationship between the vehicle target image and the road image.

[0015] Optionally, determining the vehicle target grounding point information in the vehicle target image based on the vehicle target image includes:

[0016] Determine the front or rear area information of the vehicle target image based on the vehicle target image;

[0017] The vehicle target grounding point information in the vehicle target image is determined based on the information of the front or rear area of ​​the vehicle target image.

[0018] Optionally, determining the front or rear region information of the vehicle target image based on the vehicle target image includes:

[0019] The vehicle target image is semantically segmented using a preset semantic segmentation model to obtain information about the front or rear regions of the vehicle target image; and / or,

[0020] The vehicle target image is detected by using a second preset target detection model to detect the front and / or rear of the vehicle, thereby obtaining the front or rear region information in the vehicle target image.

[0021] Optionally, determining the vehicle target grounding point information in the vehicle target image based on the vehicle front region information or vehicle rear region information in the vehicle target image includes:

[0022] The vehicle front grounding point information in the image coordinate system corresponding to the vehicle target image is determined based on the vehicle front area information in the vehicle target image, or the vehicle rear grounding point information in the image coordinate system corresponding to the vehicle target image is determined based on the vehicle rear area information in the vehicle target image.

[0023] The vehicle target image is used as the vehicle's front ground point information or rear ground point information in the image coordinate system corresponding to the vehicle target image.

[0024] Optionally, determining the absolute position of the vehicle target based on the vehicle target ground point information in the vehicle target image and the size transformation relationship between the vehicle target image and the road image includes:

[0025] Based on the size transformation relationship between the vehicle target image and the road image, the vehicle target ground point information in the vehicle target image is mapped onto the road image to obtain the vehicle target ground point information in the road image;

[0026] Based on the transformation relationship between the image coordinate system corresponding to the road image and the world coordinate system, the ground point information of the vehicle target in the road image is transformed to the world coordinate system to obtain the absolute position of the vehicle target.

[0027] Optionally, determining the absolute position of the vehicle target based on the vehicle target ground point information in the vehicle target image and the size transformation relationship between the vehicle target image and the road image includes:

[0028] Based on the vehicle target ground point information in the vehicle target image, the size transformation relationship between the vehicle target image and the road image, and the transformation relationship between the image coordinate system corresponding to the road image and the world coordinate system, the vehicle target ground point information in the world coordinate system is determined.

[0029] Determine the current heading angle and vehicle target type of the vehicle target;

[0030] Based on the vehicle target ground point information in the world coordinate system, the current heading angle of the vehicle target, and the vehicle target type, the center position of the bottom surface of the vehicle target in the world coordinate system is determined as the absolute position of the vehicle target.

[0031] Secondly, embodiments of this application also provide a vehicle positioning device, wherein the device includes:

[0032] The target detection unit is used to acquire road images captured by roadside cameras and use a first preset target detection model to perform vehicle target detection on the road images to obtain vehicle target detection results.

[0033] The generation unit is configured to generate a vehicle target image based on the vehicle target detection result and the road image, and determine the size transformation relationship between the vehicle target image and the road image;

[0034] The first determining unit is configured to determine the vehicle target grounding point information in the vehicle target image based on the vehicle target image;

[0035] The second determining unit is used to determine the absolute position of the vehicle target based on the vehicle target ground point information in the vehicle target image and the size transformation relationship between the vehicle target image and the road image.

[0036] Thirdly, embodiments of this application also provide an electronic device, including:

[0037] Processor; and

[0038] A memory configured to store computer-executable instructions, which, when executed, cause the processor to perform any of the methods described above.

[0039] Fourthly, embodiments of this application also provide a computer-readable storage medium that stores one or more programs, which, when executed by an electronic device including multiple applications, cause the electronic device to perform any of the methods described above.

[0040] The above-mentioned at least one technical solution adopted in the embodiments of this application can achieve the following beneficial effects: The vehicle positioning method of the embodiments of this application first acquires road images collected by roadside cameras, and performs vehicle target detection on the road images using a first preset target detection model to obtain vehicle target detection results; then, a vehicle target image is generated based on the vehicle target detection results and the road image, and the size transformation relationship between the vehicle target image and the road image is determined; then, the vehicle target ground point information in the vehicle target image is determined based on the vehicle target image; finally, the absolute position of the vehicle target is determined based on the vehicle target ground point information in the vehicle target image and the size transformation relationship between the vehicle target image and the road image. The embodiments of this application process the existing 2D target detection results to obtain more accurate ground point positions of vehicle targets in the image, thereby improving the vehicle positioning accuracy based on 2D target detection. Furthermore, the processing logic added in the embodiments of this application does not destroy or discard the original 2D target detection data and processing flow, making it more applicable. Attached Figure Description

[0041] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments of this application and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings:

[0042] Figure 1 This is a schematic diagram of a vehicle target detection result in the prior art;

[0043] Figure 2 This is a flowchart illustrating a vehicle positioning method according to an embodiment of this application;

[0044] Figure 3 This is a schematic diagram of the semantic segmentation result of a vehicle rear region in an embodiment of this application;

[0045] Figure 4 This is a schematic diagram of target detection results in the rear area of ​​a vehicle according to an embodiment of this application;

[0046] Figure 5 This is a schematic diagram of the structure of a vehicle positioning device according to an embodiment of this application;

[0047] Figure 6 This is a schematic diagram of the structure of an electronic device according to an embodiment of this application. Detailed Implementation

[0048] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions of this application will be clearly and completely described below in conjunction with specific embodiments and corresponding drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0049] The technical solutions provided by the various embodiments of this application are described in detail below with reference to the accompanying drawings.

[0050] This application provides a vehicle positioning method, such as... Figure 2 The diagram shows a flowchart of a vehicle positioning method according to an embodiment of this application. The method includes at least the following steps S210 to S240:

[0051] Step S210: Acquire road images captured by roadside cameras, and use the first preset target detection model to perform vehicle target detection on the road images to obtain vehicle target detection results.

[0052] The vehicle localization method of this application embodiment can be executed by a roadside device. When performing vehicle localization, it is necessary to first acquire the road image currently captured by the roadside camera, and then use an existing 2D object detection model to detect vehicle targets in the road image, thereby obtaining the vehicle target detection result, which mainly includes the position of the bounding box corresponding to the vehicle target in the image. The 2D object detection model here can be implemented using models such as YOLO or SSD. How to specifically perform vehicle target detection can be flexibly determined by those skilled in the art in conjunction with existing technology, and is not specifically limited here.

[0053] Step S220: Generate a vehicle target image based on the vehicle target detection result and the road image, and determine the size transformation relationship between the vehicle target image and the road image.

[0054] After obtaining the vehicle target detection results, the vehicle target image can be segmented from the road image based on the position of the vehicle target bounding box in the road image. Since the vehicle target bounding box is often a bounding box that contains the maximum edge of the vehicle target, the vehicle target image segmented in this way can be regarded as an image that exactly covers the complete vehicle target.

[0055] Since the vehicle target image is segmented from the road image, the size transformation relationship between the two can also be determined based on the size of the road image and the vehicle target image.

[0056] Step S230: Determine the vehicle target grounding point information in the vehicle target image based on the vehicle target image.

[0057] The vehicle target grounding point is the contact point between the vehicle target and the ground. In this embodiment, the contact point between the center point of the front or rear of the vehicle and the ground can be used as the vehicle target grounding point, and the center point of the front or rear of the vehicle can be further detected or segmented from the vehicle target image.

[0058] Step S240: Determine the absolute position of the vehicle target based on the vehicle target ground point information in the vehicle target image and the size transformation relationship between the vehicle target image and the road image.

[0059] Since the vehicle target ground point information is in the image coordinate system corresponding to the vehicle target image, it is necessary to combine the size transformation relationship between the target image and the road image determined in the previous steps to map the vehicle target ground point information back into the road image, and then determine the absolute position of the vehicle target based on the transformation relationship between the image coordinate system corresponding to the road image and the world coordinate system.

[0060] This application embodiment processes existing 2D target detection results to obtain a more accurate grounding point position of the vehicle target in the image, thereby improving the vehicle positioning accuracy based on 2D target detection. Furthermore, the processing logic added in this application embodiment does not destroy or discard the original 2D target detection data and processing flow, making it more applicable.

[0061] In some embodiments of this application, generating a vehicle target image based on the vehicle target detection result and the road image, and determining the size transformation relationship between the vehicle target image and the road image includes: cropping the road image based on the vehicle target detection result to obtain the vehicle target image; and determining the size transformation relationship between the vehicle target image and the road image based on the size of the vehicle target image and the size of the road image.

[0062] The vehicle target detection result includes the specific locations of the four corner points of the vehicle target bounding box. Therefore, the image region corresponding to the vehicle target bounding box can be cropped from the road image based on the specific locations of the four corner points of the vehicle target bounding box, thus obtaining the vehicle target image.

[0063] Since the specific positions of the four corner points of the vehicle target bounding box in the road image have been determined, the size of the vehicle target image, including its length and width, can be determined based on these positions. The size of the road image can also be determined based on the parameter settings of the roadside camera. Therefore, by combining the length and width of the vehicle target image and the road image, the size transformation relationship between the vehicle target image and the road image can be calculated. This size transformation relationship allows the position of any pixel in the vehicle target image to be mapped to the original road image.

[0064] In some embodiments of this application, determining the vehicle target grounding point information in the vehicle target image based on the vehicle target image includes: determining the front area information or rear area information in the vehicle target image based on the vehicle target image; and determining the vehicle target grounding point information in the vehicle target image based on the front area information or rear area information in the vehicle target image.

[0065] The vehicle target grounding point information in the vehicle target image determined in this application embodiment can be mainly determined based on the front or rear regions detected from the vehicle target image. Since the angle at which the vehicle target appears in the field of view of the roadside camera is different, the roadside camera may only capture the front or only the rear of the vehicle. Therefore, whether to determine the vehicle target grounding point information based on the front region information or the rear region information depends mainly on the actual detection results of the front / rear regions.

[0066] In some embodiments of this application, determining the front or rear region information of the vehicle target image based on the vehicle target image includes: performing semantic segmentation on the vehicle target image using a preset semantic segmentation model to obtain the front or rear region information of the vehicle target image; and / or, performing front and / or rear detection on the vehicle target image using a second preset target detection model to obtain the front or rear region information of the vehicle target image.

[0067] In this application embodiment, the determination of the vehicle front / rear region can be achieved in two ways. One is to use a pre-trained semantic segmentation model, such as a semantic segmentation model based on U-Net or FCN (Fully Convolutional Networks), to perform semantic segmentation processing on the vehicle target image, thereby segmenting the vehicle front or rear region in the vehicle target image. Figure 3 The diagram illustrates a semantic segmentation result of a vehicle rear region according to an embodiment of this application. The gray area in the diagram represents the segmented vehicle rear region. Of course, those skilled in the art can flexibly determine the specific method of semantic segmentation based on existing technology, and no specific limitations are made here.

[0068] Another approach is to use a pre-trained object detection model, such as the aforementioned 2D object detection model based on YOLO networks, to perform secondary detection on the front / rear targets in the vehicle target image, thereby obtaining the target detection results for the front / rear targets. Specifically, this can include the bounding boxes of the front / rear targets in the vehicle target image. Figure 4 The diagram illustrates a target detection result in the rear area of ​​a vehicle according to an embodiment of this application. The rectangular area in the diagram represents the detected rear area of ​​the vehicle. Of course, those skilled in the art can flexibly determine the specific method of target detection based on existing technology, and no specific limitations are made here.

[0069] Of course, in order to further improve the accuracy of determining the front / rear area of ​​the vehicle, the above two methods can be used simultaneously to fuse the segmentation results and detection results, thereby determining the final front or rear area information.

[0070] In some embodiments of this application, determining the vehicle target ground point information in the vehicle target image based on the vehicle front region information or the vehicle rear region information in the vehicle target image includes: determining the vehicle front ground point information in the image coordinate system corresponding to the vehicle target image based on the vehicle front region information in the vehicle target image, or determining the vehicle rear ground point information in the image coordinate system corresponding to the vehicle target image based on the vehicle rear region information in the vehicle target image; and using the vehicle front ground point information or the vehicle rear ground point information in the image coordinate system corresponding to the vehicle target image as the vehicle target ground point information.

[0071] If the front area information is determined from the vehicle target image, the front ground point information in the corresponding image coordinate system of the vehicle target image can be determined based on the front area information in the vehicle target image. The front ground point can be defined as the contact point between the front area and the ground. Specifically, the center point of the bottom edge of the detected or segmented front area can be used as the front ground point.

[0072] The x-coordinate of the center point of the bottom edge of the vehicle's front area can be determined by averaging the maximum and minimum x-coordinates of the corresponding area. The x-coordinate can be determined by the maximum y-coordinate of the corresponding area. Since image coordinate systems are typically established with the top-left corner of the image as the origin, the horizontal axis pointing to the right, and the vertical axis pointing downwards, the pixel with the maximum y-coordinate of the front area is also the pixel closest to the ground. Similarly, if the rear area information is determined from the vehicle target image, the rear ground contact point information can be determined using the same process.

[0073] In some embodiments of this application, determining the absolute position of the vehicle target based on the vehicle target ground point information in the vehicle target image and the size transformation relationship between the vehicle target image and the road image includes: mapping the vehicle target ground point information in the vehicle target image to the road image according to the size transformation relationship between the vehicle target image and the road image to obtain the vehicle target ground point information in the road image; and transforming the vehicle target ground point information in the road image to the world coordinate system according to the transformation relationship between the image coordinate system corresponding to the road image and the world coordinate system to obtain the absolute position of the vehicle target.

[0074] After determining the vehicle target ground point information, it is necessary to transform the vehicle target ground point information from the image coordinate system to the world coordinate system to obtain the corresponding absolute position. However, since the vehicle target image and the road image correspond to different image coordinate systems, the transformation relationship between the pre-calibrated image coordinate system and the world coordinate system is the transformation relationship between the image coordinate system and the world coordinate system corresponding to the original road image captured by the roadside camera.

[0075] Based on this, the embodiments of this application can first map the vehicle target ground point information in the vehicle target image back to the original road image according to the size transformation relationship between the vehicle target image and the road image, thereby determining the position of the target ground point in the road image. Finally, according to the transformation relationship between the image coordinate system corresponding to the road image and the world coordinate system, the vehicle target ground point information in the road image is transformed to the world coordinate system, thereby obtaining the absolute position of the vehicle target ground point.

[0076] In some embodiments of this application, determining the absolute position of the vehicle target based on the vehicle target ground point information in the vehicle target image and the size transformation relationship between the vehicle target image and the road image includes: determining the vehicle target ground point information in the world coordinate system based on the vehicle target ground point information in the vehicle target image, the size transformation relationship between the vehicle target image and the road image, and the transformation relationship between the image coordinate system corresponding to the road image and the world coordinate system; determining the current heading angle and vehicle target type of the vehicle target; and determining the center position of the bottom surface of the vehicle target in the world coordinate system based on the vehicle target ground point information in the world coordinate system, the current heading angle and vehicle target type of the vehicle target, as the absolute position of the vehicle target.

[0077] As described in the foregoing embodiments, the absolute position of the vehicle target ground point can be determined based on the vehicle target ground point information in the vehicle target image, the size transformation relationship between the vehicle target image and the road image, and the transformation relationship between the image coordinate system corresponding to the road image and the world coordinate system. To further improve vehicle positioning accuracy, embodiments in this field can also combine information such as the current heading angle of the vehicle target and the vehicle target type to compensate the position of the front / rear ground point to the center position of the bottom surface of the vehicle target.

[0078] Specifically, based on the orientation of the roadside camera and the current heading angle of the vehicle target, it can be determined whether the current vehicle target's bounding box belongs to the front, back, or side of the vehicle target. For example, if the vehicle target's movement direction and the camera's orientation are close to 180 degrees, it means that the target is detected as the front in the roadside camera's field of view. If the vehicle target's movement direction and the roadside camera's orientation are close, then the vehicle target is detected as the back in the roadside camera's field of view. Otherwise, it can be regarded as the side of the vehicle target.

[0079] If the detected target is the front or back of a vehicle, the front / rear contact point can be shifted a fixed distance, such as half the length of the vehicle body, towards the camera's direction, depending on the type of vehicle (e.g., bus or car). If the detected target is the side, the front / rear contact point can be shifted a fixed distance, such as half the width of the vehicle body, towards the camera's direction, depending on the type of vehicle. In other words, the detected front / rear contact point is compensated for by different amounts towards the actual center of the vehicle target, depending on the orientation and target type.

[0080] This application provides a vehicle positioning device 500, such as... Figure 5 The diagram shows a structural schematic of a vehicle positioning device according to an embodiment of this application. The device 500 includes: a target detection unit 510, a generation unit 520, a first determination unit 530, and a second determination unit 540, wherein:

[0081] The target detection unit 510 is used to acquire road images captured by roadside cameras and use a first preset target detection model to perform vehicle target detection on the road images to obtain vehicle target detection results.

[0082] The generation unit 520 is used to generate a vehicle target image based on the vehicle target detection result and the road image, and to determine the size transformation relationship between the vehicle target image and the road image;

[0083] The first determining unit 530 is used to determine the vehicle target grounding point information in the vehicle target image based on the vehicle target image;

[0084] The second determining unit 540 is used to determine the absolute position of the vehicle target based on the vehicle target ground point information in the vehicle target image and the size transformation relationship between the vehicle target image and the road image.

[0085] In some embodiments of this application, the generation unit 520 is specifically used to: crop the road image according to the vehicle target detection result to obtain the vehicle target image; and determine the size transformation relationship between the vehicle target image and the road image according to the size of the vehicle target image and the size of the road image.

[0086] In some embodiments of this application, the first determining unit 530 is specifically used to: determine the vehicle front region information or vehicle rear region information in the vehicle target image based on the vehicle target image; and determine the vehicle target grounding point information in the vehicle target image based on the vehicle front region information or vehicle rear region information in the vehicle target image.

[0087] In some embodiments of this application, the first determining unit 530 is specifically used to: perform semantic segmentation on the vehicle target image using a preset semantic segmentation model to obtain the front or rear region information of the vehicle target image; and / or, perform front and / or rear detection on the vehicle target image using a second preset target detection model to obtain the front or rear region information of the vehicle target image.

[0088] In some embodiments of this application, the first determining unit 530 is specifically used to: determine the front grounding point information of the vehicle target image in the image coordinate system corresponding to the vehicle target image based on the front area information in the vehicle target image, or determine the rear grounding point information of the vehicle target image in the image coordinate system corresponding to the vehicle target image based on the rear area information in the vehicle target image; and use the front grounding point information or the rear grounding point information in the image coordinate system corresponding to the vehicle target image as the vehicle target grounding point information.

[0089] In some embodiments of this application, the second determining unit 540 is specifically used to: map the vehicle target ground point information in the vehicle target image to the road image according to the size transformation relationship between the vehicle target image and the road image, to obtain the vehicle target ground point information in the road image; and transform the vehicle target ground point information in the road image to the world coordinate system according to the transformation relationship between the image coordinate system corresponding to the road image and the world coordinate system, to obtain the absolute position of the vehicle target.

[0090] In some embodiments of this application, the second determining unit is specifically used to: determine the vehicle target ground point information in the world coordinate system based on the vehicle target ground point information in the vehicle target image, the size transformation relationship between the vehicle target image and the road image, and the transformation relationship between the image coordinate system corresponding to the road image and the world coordinate system; determine the current heading angle and vehicle target type of the vehicle target; and determine the bottom center position of the vehicle target in the world coordinate system as the absolute position of the vehicle target based on the vehicle target ground point information in the world coordinate system, the current heading angle and vehicle target type of the vehicle target.

[0091] It is understood that the above-mentioned vehicle positioning device can realize each step of the vehicle positioning method provided in the foregoing embodiments. The relevant explanations of the vehicle positioning method are applicable to the vehicle positioning device and will not be repeated here.

[0092] Figure 6 This is a schematic diagram of the structure of an electronic device according to an embodiment of this application. Please refer to it. Figure 6At the hardware level, the electronic device includes a processor, and optionally also includes an internal bus, a network interface, and memory. The memory may include main memory, such as high-speed random-access memory (RAM), or non-volatile memory, such as at least one disk drive. Of course, the electronic device may also include other hardware required for other business operations.

[0093] The processor, network interface, and memory can be interconnected via an internal bus, which can be an ISA (Industry Standard Architecture) bus, a PCI (Peripheral Component Interconnect) bus, or an EISA (Extended Industry Standard Architecture) bus, etc. This bus can be divided into address bus, data bus, control bus, etc. For ease of representation, Figure 6 The symbol is represented by a single double-headed arrow, but this does not mean that there is only one bus or one type of bus.

[0094] Memory is used to store programs. Specifically, programs may include program code, which includes computer operation instructions. Memory may include main memory and non-volatile memory, and provides instructions and data to the processor.

[0095] The processor reads the corresponding computer program from non-volatile memory into main memory and then executes it, forming a vehicle positioning device at the logical level. The processor executes the program stored in memory and specifically performs the following operations:

[0096] Acquire road images captured by roadside cameras, and use a first preset target detection model to perform vehicle target detection on the road images to obtain vehicle target detection results;

[0097] A vehicle target image is generated based on the vehicle target detection results and the road image, and the size transformation relationship between the vehicle target image and the road image is determined.

[0098] Determine the vehicle target grounding point information in the vehicle target image based on the vehicle target image;

[0099] The absolute position of the vehicle target is determined based on the vehicle target ground point information in the vehicle target image and the size transformation relationship between the vehicle target image and the road image.

[0100] The above is as stated in this application. Figure 1The method executed by the vehicle positioning device disclosed in the illustrated embodiment can be applied to a processor or implemented by a processor. The processor may be an integrated circuit chip with signal processing capabilities. During implementation, each step of the above method can be completed by integrated logic circuits in the processor's hardware or by instructions in software form. The processor can be a general-purpose processor, including a Central Processing Unit (CPU), a Network Processor (NP), etc.; it can also be a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field-Programmable Gate Array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. It can implement or execute the methods, steps, and logic block diagrams disclosed in the embodiments of this application. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the method disclosed in the embodiments of this application can be directly embodied in the execution of a hardware decoding processor, or executed by a combination of hardware and software modules in the decoding processor. The software module can reside in a mature storage medium in the field, such as random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, or registers. This storage medium is located in memory, and the processor reads information from the memory and, in conjunction with its hardware, completes the steps of the above method.

[0101] The electronic device can also perform Figure 1 The method for executing the vehicle positioning device, and realizing the vehicle positioning device in Figure 1 The functions of the embodiments shown are not described in detail here.

[0102] This application also proposes a computer-readable storage medium that stores one or more programs, the programs including instructions that, when executed by an electronic device including multiple applications, enable the electronic device to perform... Figure 1 The method executed by the vehicle positioning device in the illustrated embodiment is specifically used to perform the following:

[0103] Acquire road images captured by roadside cameras, and use a first preset target detection model to perform vehicle target detection on the road images to obtain vehicle target detection results;

[0104] A vehicle target image is generated based on the vehicle target detection results and the road image, and the size transformation relationship between the vehicle target image and the road image is determined.

[0105] Determine the vehicle target grounding point information in the vehicle target image based on the vehicle target image;

[0106] The absolute position of the vehicle target is determined based on the vehicle target ground point information in the vehicle target image and the size transformation relationship between the vehicle target image and the road image.

[0107] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0108] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0109] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0110] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0111] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0112] Memory may include non-persistent storage in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0113] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can store information using any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0114] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0115] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0116] The above description is merely an embodiment of this application and is not intended to limit this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principle of this application should be included within the scope of the claims of this application.

Claims

1. A vehicle positioning method, wherein, The method includes: Acquire road images captured by roadside cameras, and use a first preset target detection model to perform vehicle target detection on the road images to obtain vehicle target detection results; A vehicle target image is generated based on the vehicle target detection results and the road image, and the size transformation relationship between the vehicle target image and the road image is determined. Determine the vehicle target grounding point information in the vehicle target image based on the vehicle target image; The absolute position of the vehicle target is determined based on the vehicle target ground point information in the vehicle target image and the size transformation relationship between the vehicle target image and the road image; Determining the absolute position of the vehicle target based on the vehicle target ground point information in the vehicle target image and the size transformation relationship between the vehicle target image and the road image includes: Based on the vehicle target ground point information in the vehicle target image, the size transformation relationship between the vehicle target image and the road image, and the transformation relationship between the image coordinate system corresponding to the road image and the world coordinate system, the vehicle target ground point information in the world coordinate system is determined. Determine the current heading angle and vehicle target type of the vehicle target; Based on the vehicle target ground point information in the world coordinate system, the current heading angle of the vehicle target, and the vehicle target type, the center position of the bottom surface of the vehicle target in the world coordinate system is determined as the absolute position of the vehicle target.

2. The method of claim 1, wherein, The step of generating a vehicle target image based on the vehicle target detection result and the road image, and determining the size transformation relationship between the vehicle target image and the road image, includes: The road image is cropped based on the vehicle target detection results to obtain the vehicle target image; Based on the size of the vehicle target image and the size of the road image, determine the size transformation relationship between the vehicle target image and the road image.

3. The method as described in claim 1, wherein, The step of determining the vehicle target ground point information in the vehicle target image based on the vehicle target image includes: Determine the front or rear area information of the vehicle target image based on the vehicle target image; The vehicle target grounding point information in the vehicle target image is determined based on the information of the front or rear area of ​​the vehicle target image.

4. The method as described in claim 3, wherein, The step of determining the vehicle front region information or vehicle rear region information in the vehicle target image based on the vehicle target image includes: The vehicle target image is semantically segmented using a preset semantic segmentation model to obtain information about the front or rear regions of the vehicle target image; and / or, The vehicle target image is detected by using a second preset target detection model to detect the front and / or rear of the vehicle, thereby obtaining the front or rear region information in the vehicle target image.

5. The method as described in claim 3, wherein, Determining the vehicle target ground point information in the vehicle target image based on the vehicle front area information or vehicle rear area information in the vehicle target image includes: The vehicle front grounding point information in the image coordinate system corresponding to the vehicle target image is determined based on the vehicle front area information in the vehicle target image, or the vehicle rear grounding point information in the image coordinate system corresponding to the vehicle target image is determined based on the vehicle rear area information in the vehicle target image. The vehicle target image is used as the vehicle's front ground point information or rear ground point information in the image coordinate system corresponding to the vehicle target image.

6. The method of claim 1, wherein, Determining the absolute position of the vehicle target based on the vehicle target ground point information in the vehicle target image and the size transformation relationship between the vehicle target image and the road image includes: Based on the size transformation relationship between the vehicle target image and the road image, the vehicle target ground point information in the vehicle target image is mapped onto the road image to obtain the vehicle target ground point information in the road image; Based on the transformation relationship between the image coordinate system corresponding to the road image and the world coordinate system, the vehicle target ground point information in the road image is transformed to the world coordinate system to obtain the absolute position of the vehicle target.

7. A vehicle positioning device, wherein, The device includes: The target detection unit is used to acquire road images captured by roadside cameras and use a first preset target detection model to perform vehicle target detection on the road images to obtain vehicle target detection results. The generation unit is configured to generate a vehicle target image based on the vehicle target detection result and the road image, and determine the size transformation relationship between the vehicle target image and the road image; The first determining unit is configured to determine the vehicle target grounding point information in the vehicle target image based on the vehicle target image; The second determining unit is used to determine the absolute position of the vehicle target based on the vehicle target ground point information in the vehicle target image and the size transformation relationship between the vehicle target image and the road image; The second determining unit is specifically used for: Based on the vehicle target ground point information in the vehicle target image, the size transformation relationship between the vehicle target image and the road image, and the transformation relationship between the image coordinate system corresponding to the road image and the world coordinate system, the vehicle target ground point information in the world coordinate system is determined. Determine the current heading angle and vehicle target type of the vehicle target; Based on the vehicle target ground point information in the world coordinate system, the current heading angle of the vehicle target, and the vehicle target type, the center position of the bottom surface of the vehicle target in the world coordinate system is determined as the absolute position of the vehicle target.

8. An electronic device, comprising: processor; as well as A memory configured to store computer-executable instructions, which, when executed, cause the processor to perform the method of any one of claims 1 to 6.

9. A computer-readable storage medium storing one or more programs, which, when executed by an electronic device including a plurality of applications, cause the electronic device to perform the method of any one of claims 1 to 6.

Images