Camera calibration method, device, equipment and computer-readable storage medium

Abstract

The present disclosure provides a camera calibration method, apparatus, device and computer-readable storage medium. The method includes: obtaining first image information of a target calibration field collected by cameras installed in a preset number of target vehicles, and the target vehicles have completed camera calibration operations; performing a three-dimensional reconstruction operation on the target calibration field based on the first image information to obtain multiple three-dimensional position information of each traffic element in the target calibration field, and target three-dimensional position information with an accuracy exceeding a preset threshold in the target calibration field; obtaining second image information collected by a camera on an uncalibrated vehicle in the target calibration field; and performing a calibration operation on the camera in the uncalibrated vehicle based on the multiple three-dimensional position information, the target three-dimensional position information and the second image information. There is no need to calibrate each vehicle individually, which improves the efficiency of vehicle calibration. In addition to the calibration operation of the target vehicle, there is no need for staff to manually perform calibration operations on the vehicle, which saves human resources on the basis of improving calibration efficiency.

Description

Technical Field

[0001] The present disclosure relates to the field of intelligent transportation, and in particular to a camera calibration method, apparatus, device, and computer-readable storage medium. Background Art

[0002] In the field of intelligent transportation, high-precision maps are a fundamental service relied upon by many businesses. For example, in the online ride-hailing industry, services such as user order placement, platform dispatching, driver pickup, route planning, and navigation all rely on high-precision electronic maps. To achieve high-precision maps, and considering cost and data collection efficiency, dashcams installed in online ride-hailing vehicles are generally used for data collection and 3D reconstruction. However, since dashcam cameras typically exhibit significant image distortion, obtaining stable and accurate 3D scenes requires calibration of both camera intrinsic parameters and camera extrinsic parameters relative to the ground. These intrinsic parameters primarily include the camera's focal length, principal axis, and image distortion, while extrinsic parameters include the camera's mounting position and height relative to the ground.

[0003] In existing technology, the production of high-precision maps requires the use of vehicles to collect road conditions. Before using vehicles to collect road image information, personnel must first manually calibrate the cameras installed on each vehicle. Due to the large number of vehicles used for image collection, manual calibration is inefficient and labor-intensive. Summary of the Invention

[0004] The present disclosure provides a camera calibration method, apparatus, device, and computer-readable storage medium, which are used to solve the technical problems of low calibration efficiency, difficulty in creating calibration scenes, and resource consumption in existing camera calibration methods.

[0005] A first aspect of the present disclosure is to provide a camera calibration method, comprising:

[0006] Acquire first image information of a target calibration field captured by cameras provided in a preset number of target vehicles;

[0007] Performing a three-dimensional reconstruction operation on the target calibration field according to the first image information to obtain a plurality of three-dimensional position information of each traffic element in the target calibration field, and three-dimensional position information of targets in the target calibration field whose accuracy exceeds a preset threshold;

[0008] Acquire second image information captured by a camera on an uncalibrated vehicle in the target calibration field;

[0009] A calibration operation is performed on the camera in the uncalibrated vehicle according to the multiple three-dimensional position information, the target three-dimensional position information and the second image information.

[0010] A second aspect of the present disclosure is to provide a camera calibration device, comprising:

[0011] A first acquisition module is used to acquire first image information of a target calibration field captured by cameras provided in a preset number of target vehicles;

[0012] a reconstruction module, configured to perform a three-dimensional reconstruction operation on the target calibration field based on the first image information, to obtain a plurality of three-dimensional position information of each traffic element in the target calibration field, and three-dimensional position information of targets in the target calibration field whose accuracy exceeds a preset threshold;

[0013] A second acquisition module is used to acquire second image information captured by a camera on an uncalibrated vehicle in the target calibration field;

[0014] The first calibration module is configured to perform a calibration operation on the camera in the uncalibrated vehicle according to the multiple three-dimensional position information, the target three-dimensional position information, and the second image information.

[0015] A third aspect of the present disclosure is to provide a camera calibration device, comprising: a memory, a processor;

[0016] Memory; a memory for storing instructions executable by the processor;

[0017] The processor is configured to call program instructions in the memory to execute the camera calibration method as described in the first aspect.

[0018] A fourth aspect of the present disclosure is to provide a computer-readable storage medium, wherein the computer-readable storage medium stores computer-executable instructions, and when the computer-executable instructions are executed by a processor, they are used to implement the camera calibration method as described in the first aspect.

[0019] The camera calibration method, apparatus, device, and computer-readable storage medium provided by the present disclosure perform calibration operations on a preset number of target vehicles in advance, and collect first image information of the target calibration field through the calibrated target vehicles. Based on the first image information, a three-dimensional reconstruction operation is performed on the target calibration field to obtain multiple three-dimensional position information of each traffic element in the target calibration field, as well as target three-dimensional position information in the target calibration field with an accuracy exceeding a preset threshold. When an uncalibrated vehicle enters the target calibration field, second image information collected by the uncalibrated vehicle in the target calibration field is obtained. Then, based on the multiple three-dimensional position information, the target three-dimensional position information, and the second image information, the camera in the uncalibrated vehicle can be calibrated. There is no need to calibrate each vehicle individually, which improves the efficiency of vehicle calibration. In addition to the calibration operation of the target vehicle, there is no need for staff to manually calibrate the vehicle, which saves human resources on the basis of improving calibration efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the embodiments of the present disclosure or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present disclosure. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0021] Figure 1 A schematic diagram of the network architecture on which this disclosure is based;

[0022] Figure 2 A flowchart of a camera calibration method according to the first embodiment of the present disclosure is provided;

[0023] Figure 3 A flowchart of a camera calibration method according to the second embodiment of the present disclosure is provided;

[0024] Figure 4 A flowchart of a camera calibration method according to a third embodiment of the present disclosure is provided;

[0025] Figure 5 A schematic diagram of a divided target area provided in an embodiment of the present disclosure;

[0026] Figure 6 A flowchart of a camera calibration method according to a fourth embodiment of the present disclosure is provided;

[0027] Figure 7 A schematic diagram of the structure of a camera calibration device provided in Embodiment 5 of the present disclosure;

[0028] Figure 8 A schematic diagram of the structure of a camera calibration device provided in Example 6 of the present disclosure;

[0029] Figure 9 A schematic structural diagram of a camera calibration device provided in Example 7 of the present disclosure;

[0030] Figure 10 A schematic diagram of the structure of a camera calibration device provided in Example 8 of the present disclosure;

[0031] Figure 11 This is a structural diagram of the camera calibration device provided in Example 9 of the present disclosure. DETAILED DESCRIPTION

[0032] To make the objectives, technical solutions, and advantages of the embodiments of the present disclosure more clear, the technical solutions in the embodiments of the present disclosure will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present disclosure. Obviously, the described embodiments are only part of the embodiments of the present disclosure, not all of the embodiments. All other embodiments obtained based on the embodiments in the present disclosure are within the scope of protection of the present disclosure.

[0033] In response to the technical problems mentioned above, such as low calibration efficiency, difficulty in creating calibration scenes, and resource consumption in existing camera calibration methods, the present disclosure provides a camera calibration method, apparatus, device, and computer-readable storage medium.

[0034] It should be noted that the camera calibration method, apparatus, device and computer-readable storage medium provided in the present disclosure can be used in scenarios for testing various application software.

[0035] In the prior art, to calibrate the intrinsic and extrinsic parameters of the camera in a dashcam, projective calibration is generally used. Specifically, a precisely machined calibration template is used, typically with corner points of known 3D coordinates placed on it. These known corner points can be used to calibrate the camera. Well-known methods include Zhang's calibration and Tsai's calibration. However, using these methods for camera calibration requires dedicated personnel to calibrate each device individually, which is time-consuming.

[0036] In the process of solving the above-mentioned technical problems, the inventors discovered through research that staff can pre-calibrate a preset number of target vehicles. These target vehicles then perform image acquisition, and a 3D reconstruction of the calibration field is performed based on the image information collected by the target vehicles. The purpose of 3D reconstruction is to restore the camera's 6-DOF pose, including the 3-DOF rotation R and the 3-DOF offset t. This allows subsequent calibration of uncalibrated vehicles that travel to the calibration field based on the 3D reconstructed scene. This eliminates the need for manual camera calibration for each vehicle, effectively saving human resources and improving calibration efficiency.

[0037] Figure 1 This is a schematic diagram of the network architecture based on which the present disclosure is based. Figure 1 As shown, the network architecture based on the present disclosure includes at least: vehicle 1 and server 2. The server 2 is provided with a camera calibration device, which is written in C / C++, Java, Shell, or Python. Vehicle 1 can be any vehicle capable of being equipped with a camera for image acquisition.

[0038] Figure 2 This is a flow chart of the camera calibration method provided in the first embodiment of the present disclosure, as shown in FIG. Figure 1As shown, the method includes:

[0039] Step 101: Acquire first image information of a target calibration field captured by cameras installed in a preset number of target vehicles, wherein the target vehicles have completed camera calibration operations.

[0040] The execution subject of this embodiment is a camera calibration device, which can be coupled to a server. The server can be connected to multiple target vehicles for communication, thereby enabling information exchange with multiple target vehicles.

[0041] In this embodiment, to avoid manual calibration of each vehicle for information collection, a preset number of target vehicles can be set in advance, wherein the target vehicles are vehicles that have undergone internal and external parameter calibration. The number of target vehicles can be set according to actual needs and is not limited by this disclosure. For example, the number of target vehicles can be 5.

[0042] The target vehicle is provided with a camera for collecting image information. It should be noted that the camera can be an independent device or a camera coupled to a driving recorder or other device for data collection, and the present disclosure does not limit this.

[0043] After the camera calibration device establishes a communication connection with the target vehicle, when the target vehicle drives to a preset target calibration site, an image acquisition operation can be performed at the target calibration site using a preset camera. The camera calibration device can obtain first image information captured by a camera set in the target vehicle.

[0044] Step 102: Perform a three-dimensional reconstruction operation on the target calibration field based on the first image information to obtain multiple three-dimensional position information of each traffic element in the target calibration field, as well as three-dimensional position information of targets in the target calibration field whose accuracy exceeds a preset threshold.

[0045] In this embodiment, after acquiring the first image information captured by the target vehicle, a three-dimensional reconstruction operation can be performed on the target calibration field based on the first image information. Since the target vehicle has already completed the calibration operation of the intrinsic and extrinsic parameters, the three-dimensional scene corresponding to the target calibration field reconstructed based on the first image information can accurately represent the actual conditions of the target calibration field.

[0046] In practical applications, the target calibration field may contain a variety of traffic elements, such as traffic lights, roadblocks, trees, lane markings, etc. Therefore, to improve the accuracy of subsequent camera calibration, after obtaining the 3D scene corresponding to the target calibration field, it is necessary to determine multiple 3D position information of each traffic element in the 3D scene and filter these multiple 3D position information to determine the target 3D position information in the target calibration field whose accuracy exceeds a preset threshold.

[0047] Step 103: Acquire second image information captured by a camera on an uncalibrated vehicle in the target calibration field.

[0048] In this embodiment, after completing the reconstruction of the three-dimensional scene and obtaining the three-dimensional position information, other vehicles that have not been camera-annotated may be prompted to drive to the target calibration site to perform information collection operations.

[0049] Accordingly, the camera calibration device can obtain second image information captured by a camera on an uncalibrated vehicle within the target calibration field. The camera can be a standalone device or a camera coupled to a driving recorder or other device for data acquisition, and this disclosure does not limit this.

[0050] Step 104: calibrate the camera in the uncalibrated vehicle according to the multiple three-dimensional position information, the target three-dimensional position information, and the second image information.

[0051] In this embodiment, once the camera calibration device acquires the second image information captured by the camera on an uncalibrated vehicle, it can calibrate the camera on the uncalibrated vehicle based on the multiple 3D position information, the target's 3D position information, and the second image information. This allows operators to calibrate only a preset number of vehicles; the remaining vehicles can be automatically calibrated, eliminating unnecessary manual work and improving calibration efficiency.

[0052] The camera calibration method provided in this embodiment pre-calibrates a preset number of target vehicles and collects first image information of a target calibration field using the calibrated target vehicles. Based on the first image information, a three-dimensional reconstruction operation is performed on the target calibration field to obtain multiple three-dimensional position information of each traffic element in the target calibration field, as well as target three-dimensional position information in the target calibration field with an accuracy exceeding a preset threshold. When an uncalibrated vehicle enters the target calibration field, second image information collected by the uncalibrated vehicle in the target calibration field is obtained. Furthermore, based on the multiple three-dimensional position information, the target three-dimensional position information, and the second image information, the camera in the uncalibrated vehicle can be calibrated. This eliminates the need to calibrate each vehicle individually, improving vehicle calibration efficiency. Furthermore, aside from the calibration of the target vehicle, manual vehicle calibration is not required, thereby improving calibration efficiency and saving human resources.

[0053] Figure 3 This is a flow chart of the camera calibration method provided in the second embodiment of the present disclosure. Based on the first embodiment, Figure 3 As shown, step 102 specifically includes:

[0054] Step 201: Perform a three-dimensional reconstruction operation on the target calibration field according to the first image information to obtain a three-dimensional scene corresponding to the target calibration field.

[0055] Step 202: Determine multiple three-dimensional position information of each traffic element in the three-dimensional scene.

[0056] Step 203: Filter the plurality of three-dimensional position information to obtain target three-dimensional position information having an accuracy exceeding a preset threshold in the target calibration field.

[0057] In this embodiment, after acquiring the first image information captured by the target vehicle, a 3D reconstruction operation can be performed on the target calibration field based on the first image information. The purpose of the 3D reconstruction is to restore the camera's 6-DOF pose, including the 3-DOF rotation R and the 3-DOF offset t. Since the target vehicle has completed the calibration of the intrinsic and extrinsic parameters, the 3D scene corresponding to the target calibration field reconstructed based on the first image information can accurately represent the actual condition of the target calibration field.

[0058] After obtaining the 3D scene corresponding to the target calibration site, it is necessary to determine multiple 3D positions of each traffic element in the 3D scene, as the target calibration site often contains multiple traffic elements. To improve the accuracy of subsequent calibration, this multiple 3D positions can be filtered to obtain the target 3D position information whose accuracy exceeds a preset threshold.

[0059] Furthermore, based on the first embodiment, step 203 specifically includes:

[0060] Three-dimensional position information with a reprojection error less than a preset threshold and / or uniform spatial distribution among the multiple three-dimensional position information is obtained as the target three-dimensional position information.

[0061] In this embodiment, in order to implement the screening operation on multiple 3D position information, 3D position information with a reprojection error less than a preset threshold and / or uniform spatial distribution among the multiple 3D position information can be obtained as the target 3D position information.

[0062] Specifically, the reprojection error corresponding to each piece of 3D position information can be determined separately, and the 3D position information with a reprojection error less than a preset threshold is used as the target 3D position information. The preset threshold can be 3-4 pixels. Alternatively, the preset threshold can be set by the operator based on actual needs, and this disclosure does not impose any restrictions on this.

[0063] Alternatively, the number of cameras capable of observing each piece of 3D position information can be determined separately. The more times a piece of 3D position information is observed, the higher the confidence level of the recovered position and the more uniform its spatial distribution. Therefore, 3D position information can be filtered based on the number of cameras, with 3D position information whose observations exceed a preset threshold being selected as the target 3D position information.

[0064] The camera calibration method provided in this embodiment performs a three-dimensional reconstruction operation on the target calibration field based on the first image information, and determines multiple three-dimensional position information of each traffic element in the three-dimensional scene and target three-dimensional position information with an accuracy exceeding a preset threshold. This enables uncalibrated vehicles to be automatically calibrated based on the above information, eliminating the need for manual calibration by staff, thereby improving calibration efficiency and saving human resources.

[0065] Figure 4 This is a flowchart of a camera calibration method provided in the third embodiment of the present disclosure. Based on any of the above embodiments, Figure 4 As shown, before step 101, the following steps are also included:

[0066] Step 301: Obtain the operational data of multiple online ride-hailing vehicles from a preset database.

[0067] Step 302: Filter the online-hailing vehicles according to the operating data to obtain a preset number of online-hailing vehicles.

[0068] Step 303: Obtain third image information collected by the preset number of online-hailing vehicles at the target calibration site.

[0069] Step 304: Calibrate the preset number of online-hailing vehicles based on the third image information to obtain the preset number of target vehicles.

[0070] The network architecture underlying the present disclosure may further include a database, and the camera calibration device may be connected to the database and the vehicle, respectively. The database stores a large amount of historical operation data of online ride-hailing services.

[0071] In this embodiment, before collecting the first image information through the target vehicle, it is first necessary to determine the target vehicle. Specifically, the operating data of multiple online ride-hailing vehicles can be obtained from the database. The online ride-hailing vehicles are screened based on the operating data to obtain a preset number of online ride-hailing vehicles. The screening rules can specifically be to obtain the credit rating information, online time information, camera installation posture information, and the quality of the returned image corresponding to each online ride-hailing vehicle in the operating data, and select online ride-hailing vehicles with higher credit rating information, longer online time information, appropriate camera installation posture information, and higher quality of the returned image.

[0072] A preset number of online-hailing vehicles are directed to travel to a target calibration site to collect information and obtain third image information. Calibration operations are then performed on the preset number of online-hailing vehicles based on the third image information to obtain the preset number of target vehicles. The calibration of the online-hailing vehicles can be performed automatically by the camera calibration device or by a human operator.

[0073] Furthermore, based on any of the above embodiments, before step 303, the following steps may be further included:

[0074] Dividing a preset target area according to the operation data to obtain a plurality of sub-areas;

[0075] A sub-region whose thermal value exceeds a preset threshold is selected from the multiple sub-regions as the target calibration field.

[0076] In this embodiment, before obtaining the third image information collected by the preset number of online-hailing vehicles in the target calibration field, it is also necessary to determine the target calibration field.

[0077] Specifically, the target area can be divided according to the operational data to obtain multiple sub-areas. Specifically, any area division algorithm can be used to divide the target area, and this disclosure does not limit this.

[0078] Determine the thermal value corresponding to each sub-region, and select the sub-region whose thermal value exceeds a preset threshold as the target calibration field. Optionally, the sub-region with the highest thermal value can be used as the target calibration field.

[0079] Figure 5 A schematic diagram of the target area after division provided by the embodiment of the present disclosure, such as Figure 5 As shown, the target area can be divided into five sub-areas according to the preset division rules and the operation data. Among the five sub-areas, the sub-area with a thermal value exceeding a preset threshold is selected as the target calibration field.

[0080] The camera calibration method provided in this embodiment obtains the target vehicle by screening online-hailing vehicles based on operational data before collecting the first image information of the target vehicle, thereby improving the quality of the first image information and providing a basis for subsequent camera labeling.

[0081] Figure 6 This is a flowchart of a camera calibration method provided in the fourth embodiment of the present disclosure. Based on any of the above embodiments, Figure 6 As shown, step 104 specifically includes:

[0082] Step 401: Determine the correspondence between the pixel points in the second image information and the plurality of three-dimensional position information.

[0083] Step 402: Perform an internal parameter calibration operation on the camera in the uncalibrated vehicle according to the corresponding relationship.

[0084] Step 403: Perform an extrinsic calibration operation on the camera in the uncalibrated vehicle according to the target three-dimensional position information.

[0085] In this embodiment, after obtaining multiple three-dimensional position information and target three-dimensional position information corresponding to the target calibration field respectively, the camera in the uncalibrated vehicle can be calibrated based on the multiple three-dimensional position information, the target three-dimensional position information and the second image information.

[0086] Specifically, a correspondence between pixel points in the second image information and multiple three-dimensional position information can be determined. Based on this correspondence, an intrinsic calibration operation is performed on the camera in the uncalibrated vehicle. The intrinsic calibration operation can be performed on the camera in the uncalibrated vehicle using a traditional calibration method, or using the Zhang calibration method. This is not limited in the present disclosure. Furthermore, an extrinsic calibration operation can be performed on the camera in the uncalibrated vehicle based on the target three-dimensional position information.

[0087] Specifically, the camera in the uncalibrated vehicle can be calibrated with external parameters using the PNP method according to the target three-dimensional position information.

[0088] The camera calibration method provided in this embodiment can calibrate cameras in uncalibrated vehicles based on multiple 3D position information, target 3D position information, and second image information. This allows operators to calibrate only a preset number of vehicles; the remaining vehicles can be automatically calibrated, eliminating unnecessary manual work and improving calibration efficiency.

[0089] Figure 7 This is a structural diagram of a camera calibration device provided in the fifth embodiment of the present disclosure, as shown in FIG. Figure 7 As shown, the device includes: a first acquisition module 51, a reconstruction module 52, a second acquisition module 53 and a first calibration module 54. The first acquisition module 51 is used to acquire first image information of a target calibration field captured by cameras installed in a preset number of target vehicles, wherein the target vehicles have completed the camera calibration operation. The reconstruction module 52 is used to perform a three-dimensional reconstruction operation on the target calibration field based on the first image information, and obtain multiple three-dimensional position information of each traffic element in the target calibration field, as well as target three-dimensional position information in the target calibration field whose accuracy exceeds a preset threshold. The second acquisition module 53 is used to acquire second image information captured by a camera on an uncalibrated vehicle in the target calibration field. The first calibration module 54 is used to perform a calibration operation on the camera in the uncalibrated vehicle based on the multiple three-dimensional position information, the target three-dimensional position information and the second image information.

[0090] The camera calibration device provided in this embodiment pre-calibrates a preset number of target vehicles and collects first image information of a target calibration field from the calibrated target vehicles. Based on the first image information, a three-dimensional reconstruction operation is performed on the target calibration field to obtain multiple three-dimensional position information of each traffic element in the target calibration field, as well as target three-dimensional position information in the target calibration field with an accuracy exceeding a preset threshold. When an uncalibrated vehicle enters the target calibration field, second image information collected by the uncalibrated vehicle in the target calibration field is obtained. Furthermore, based on the multiple three-dimensional position information, the target three-dimensional position information, and the second image information, the camera in the uncalibrated vehicle can be calibrated. This eliminates the need to calibrate each vehicle individually, improving vehicle calibration efficiency. Furthermore, aside from the calibration of the target vehicle, manual vehicle calibration is no longer required, saving human resources while improving calibration efficiency.

[0091] Figure 8This is a schematic diagram of the structure of a camera calibration device provided in Example 6 of the present disclosure. Based on Example 5, the reconstruction module includes: a reconstruction unit 61, a position information determination unit 62, and a screening unit 63. The reconstruction unit 61 is used to perform a three-dimensional reconstruction operation on the target calibration field based on the first image information to obtain a three-dimensional scene corresponding to the target calibration field. The position information determination unit 62 is used to determine multiple three-dimensional position information of each traffic element in the three-dimensional scene. The screening unit 63 is used to perform a screening operation on the multiple three-dimensional position information to obtain target three-dimensional position information in the target calibration field whose accuracy exceeds a preset threshold.

[0092] Furthermore, based on the fifth embodiment, the screening unit is used to:

[0093] Three-dimensional position information with a reprojection error less than a preset threshold and / or uniform spatial distribution among the multiple three-dimensional position information is obtained as the target three-dimensional position information.

[0094] Figure 9 The schematic diagram of the structure of the camera calibration device provided in the seventh embodiment of the present disclosure, based on any of the above embodiments, further includes: an operation data acquisition module 71, a screening module 72, a third acquisition module 73, and a second calibration module 74. Among them, the operation data acquisition module 71 is used to obtain the operation data of multiple online-hailing vehicles in a preset database; the screening module 72 is used to screen the online-hailing vehicles according to the operation data to obtain a preset number of online-hailing vehicles; the third acquisition module 73 is used to obtain third image information collected by the preset number of online-hailing vehicles at the target calibration site; the second calibration module 74 is used to perform a calibration operation on the preset number of online-hailing vehicles according to the third image information to obtain the preset number of target vehicles.

[0095] Furthermore, based on any of the above embodiments, the device further includes:

[0096] A division module, configured to divide a preset target area according to the operation data to obtain a plurality of sub-areas;

[0097] A selection module is used to select a sub-region whose thermal value exceeds a preset threshold from the multiple sub-regions as the target calibration field.

[0098] Figure 10A schematic diagram of the structure of a camera calibration device provided in Example 8 of the present disclosure, based on any of the above embodiments, wherein the first calibration module includes: a determination unit 81, an intrinsic parameter calibration unit 82, and an extrinsic parameter calibration unit 83. The determination unit 81 is configured to determine the correspondence between the pixel points in the second image information and the multiple three-dimensional position information. The intrinsic parameter calibration unit 82 is configured to perform an intrinsic parameter calibration operation on the camera in the uncalibrated vehicle based on the correspondence. The extrinsic parameter calibration unit 83 is configured to perform an extrinsic parameter calibration operation on the camera in the uncalibrated vehicle based on the target three-dimensional position information.

[0099] Furthermore, based on any of the above embodiments, the extrinsic parameter calibration unit is used to:

[0100] According to the target three-dimensional position information, an extrinsic calibration operation is performed on the camera in the uncalibrated vehicle using a PNP method.

[0101] Figure 11 This is a structural diagram of a camera calibration device provided in Example 9 of the present disclosure, as shown in FIG. Figure 11 As shown, the camera calibration device includes: a memory 91, a processor 92;

[0102] Memory 91; Memory 91 for storing instructions executable by the processor 92;

[0103] The processor 92 is configured to call program instructions in the memory 91 to execute the method described in any of the above embodiments.

[0104] Memory 91 is used to store programs. Specifically, the programs may include program code, which includes computer operating instructions. Memory 91 may include high-speed RAM memory or non-volatile memory, such as at least one disk storage device.

[0105] The processor 92 may be a central processing unit (CPU), an application specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of the present disclosure.

[0106] Optionally, in a specific implementation, if the memory 91 and the processor 92 are implemented independently, the memory 91 and the processor 92 can be connected to each other via a bus and communicate with each other. The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus. The bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 11 Only one thick line is used in the diagram, but this does not mean that there is only one bus or one type of bus.

[0107] Optionally, in a specific implementation, if the memory 91 and the processor 92 are integrated on a chip, the memory 91 and the processor 92 can communicate with each other through an internal interface.

[0108] Yet another embodiment of the present disclosure provides a computer-readable storage medium, wherein the computer-readable storage medium stores computer-executable instructions. When the computer-executable instructions are executed by a processor, they are used to implement the method described in any of the above embodiments.

[0109] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the device described above can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.

[0110] Those skilled in the art will appreciate that all or part of the steps in the above-described method embodiments can be implemented using hardware associated with program instructions. The aforementioned program can be stored in a computer-readable storage medium. When executed, the program performs the steps of the above-described method embodiments. The aforementioned storage medium includes various media capable of storing program code, such as ROM, RAM, magnetic disks, or optical disks.

[0111] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present disclosure, rather than to limit them. Although the present disclosure has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present disclosure.

Claims

1. A camera calibration method, characterized in that: include: Acquiring first image information of a target calibration field captured by cameras installed in a preset number of target vehicles, wherein when the target vehicles drive to the target calibration field, the cameras perform image capture operations in the target calibration field, and the cameras in the target vehicles have completed calibration operations of intrinsic parameters and extrinsic parameters; Performing a three-dimensional reconstruction operation on the target calibration field according to the first image information to obtain a plurality of three-dimensional position information of each traffic element in the target calibration field, and three-dimensional position information of targets in the target calibration field whose accuracy exceeds a preset threshold; Acquire second image information captured by a camera on an uncalibrated vehicle in the target calibration field; Performing a calibration operation on the camera in the uncalibrated vehicle according to the multiple three-dimensional position information, the target three-dimensional position information, and the second image information, including: performing an external parameter calibration operation on the camera in the uncalibrated vehicle according to the target three-dimensional position information; Before obtaining the first image information of the target calibration field collected by a preset number of target vehicles, the method further includes: Obtain the operational data of multiple online ride-hailing services from a preset database; Filtering the online-hailing vehicles according to the operation data to obtain a preset number of online-hailing vehicles, wherein the filtering criteria include camera installation posture information and quality of the returned images; Obtaining third image information collected by the preset number of online-hailing vehicles at the target calibration site; The preset number of online-hailing vehicles are calibrated according to the third image information to obtain the preset number of target vehicles.

2. The method according to claim 1, characterized in that The performing a three-dimensional reconstruction operation on the target calibration field according to the first image information to obtain a plurality of three-dimensional position information of each traffic element in the target calibration field, and three-dimensional position information of targets in the target calibration field whose accuracy exceeds a preset threshold, includes: Performing a three-dimensional reconstruction operation on the target calibration field according to the first image information to obtain a three-dimensional scene corresponding to the target calibration field; Determining multiple three-dimensional position information of each traffic element in the three-dimensional scene; A screening operation is performed on the plurality of three-dimensional position information to obtain target three-dimensional position information having an accuracy exceeding a preset threshold in the target calibration field.

3. The method according to claim 2, characterized in that The filtering operation on the plurality of three-dimensional position information to obtain the target three-dimensional position information with an accuracy exceeding a preset threshold in the target calibration field includes: Three-dimensional position information with a reprojection error less than a preset threshold and / or uniform spatial distribution among the multiple three-dimensional position information is obtained as the target three-dimensional position information.

4. The method according to claim 1, wherein Before obtaining the third image information of the preset number of online-hailing vehicles collected at the target calibration site, the method further includes: Dividing a preset target area according to the operation data to obtain a plurality of sub-areas; A sub-region whose thermal value exceeds a preset threshold is selected from the multiple sub-regions as the target calibration field.

5. The method according to any one of claims 1 to 3, characterized in that The step of calibrating the camera in the uncalibrated vehicle according to the plurality of three-dimensional position information, the target three-dimensional position information, and the second image information further includes: Determining a correspondence between pixel points in the second image information and the plurality of three-dimensional position information; According to the corresponding relationship, an intrinsic parameter calibration operation is performed on the camera in the uncalibrated vehicle.

6. The method according to claim 5, characterized in that The performing an external parameter calibration operation on the camera in the uncalibrated vehicle according to the target three-dimensional position information includes: According to the target three-dimensional position information, an extrinsic calibration operation is performed on the camera in the uncalibrated vehicle using a PNP method.

7. A camera calibration device, characterized in that: include: a first acquisition module, configured to acquire first image information of a target calibration field acquired by cameras installed in a preset number of target vehicles, wherein when the target vehicles drive to the target calibration field, the cameras perform image acquisition operations in the target calibration field, and the cameras in the target vehicles have completed calibration operations of intrinsic and extrinsic parameters; a reconstruction module, configured to perform a three-dimensional reconstruction operation on the target calibration field based on the first image information, to obtain a plurality of three-dimensional position information of each traffic element in the target calibration field, and three-dimensional position information of targets in the target calibration field whose accuracy exceeds a preset threshold; A second acquisition module is used to acquire second image information captured by a camera on an uncalibrated vehicle in the target calibration field; a first calibration module, configured to perform a calibration operation on the camera in the uncalibrated vehicle according to the plurality of three-dimensional position information, the target three-dimensional position information, and the second image information; The first calibration module includes: an extrinsic parameter calibration unit, configured to perform an extrinsic parameter calibration operation on the camera in the uncalibrated vehicle according to the target three-dimensional position information; The device further comprises: Operation data acquisition module, used to obtain the operation data of multiple online ride-hailing vehicles from a preset database; a screening module, configured to screen the online-hailing vehicles according to the operation data to obtain a preset number of online-hailing vehicles, wherein the screening criteria include camera installation posture information and the quality of the returned image; A third acquisition module is used to acquire third image information collected by the preset number of online-hailing vehicles at the target calibration site; The second calibration module is used to calibrate the preset number of online-hailing vehicles based on the third image information to obtain the preset number of target vehicles.

8. The device according to claim 7, characterized in that The reconstruction module includes: a reconstruction unit, configured to perform a three-dimensional reconstruction operation on the target calibration field according to the first image information to obtain a three-dimensional scene corresponding to the target calibration field; a position information determining unit, configured to determine a plurality of three-dimensional position information of each traffic element in the three-dimensional scene; The screening unit is used to perform a screening operation on the multiple three-dimensional position information to obtain the target three-dimensional position information whose accuracy exceeds a preset threshold in the target calibration field.

9. The device according to claim 8, characterized in that The screening unit is used to: Three-dimensional position information with a reprojection error less than a preset threshold and / or uniform spatial distribution among the multiple three-dimensional position information is obtained as the target three-dimensional position information.

10. The device according to claim 7, characterized in that The device further comprises: A division module, configured to divide a preset target area according to the operation data to obtain a plurality of sub-areas; A selection module is used to select a sub-region whose thermal value exceeds a preset threshold from the multiple sub-regions as the target calibration field.

11. The device according to any one of claims 7 to 9, characterized in that: The first calibration module also includes: a determining unit, configured to determine a correspondence between pixel points in the second image information and the plurality of three-dimensional position information; An internal parameter calibration unit is used to perform an internal parameter calibration operation on the camera in the uncalibrated vehicle according to the corresponding relationship.

12. The device according to claim 11, characterized in that The external parameter calibration unit is used for: According to the target three-dimensional position information, an extrinsic calibration operation is performed on the camera in the uncalibrated vehicle using a PNP method.

13. A camera calibration device, characterized in that: include: Memory, processor; Memory; a memory for storing instructions executable by the processor; The processor is configured to call program instructions in the memory to execute the camera calibration method according to any one of claims 1 to 6.

14. A computer-readable storage medium, characterized in that The computer-readable storage medium stores computer-executable instructions, which, when executed by a processor, are used to implement the camera calibration method according to any one of claims 1 to 6.

Images