Data calibration method, electronic device and computer readable storage device

Abstract

The present application discloses a data calibration method. The method includes: acquiring an optical image and a radar image of a target area; wherein the radar image includes at least one object point; performing target detection on the optical image to determine at least one object area in the optical image; determining a matching target area and target point in at least one object area and at least one object point; determining a conversion relationship between the target area and the target point to determine a conversion relationship between pixel points of the optical image and coordinate points of the radar image, so as to calibrate the object area and the object point. The present application also discloses an electronic device and a computer-readable storage device. Through the above-mentioned method, the present application can quickly and accurately achieve matching correspondence of target objects in radar and video sensors.

Description

Technical Field

[0001] The present application relates to the field of data processing, and in particular to a data calibration method, an electronic device, and a computer-readable storage device. Background Art

[0002] With the advancement of science and technology, radar monitoring technology has played an increasingly important role in the transportation sector. Radar acquires measurement information of moving targets with a high detection rate. Target tracking algorithms, based on this radar input, quickly and accurately calculate the target's trajectory and estimated motion information, further enabling classification and identification. However, radar only identifies a point as a target. Beyond its motion, it cannot capture any other information related to that point, making it impossible to match it with actual objects. Video targets, on the other hand, can capture additional information about the target, such as its appearance, shape, and state. Accurately calibrating both radar and video sensors to accurately match targets has become a hot topic and a challenge in current research. Summary of the Invention

[0003] The main purpose of this application is to propose a data calibration method, an electronic device, and a computer-readable storage device that can quickly and accurately achieve matching correspondence between target objects in radar and video sensors.

[0004] To address the above technical issues, the first technical solution adopted in this application is to provide a data calibration method. The method includes: acquiring an optical image and a radar image of a target area; wherein the radar image includes at least one object point; performing target detection on the optical image to determine at least one object area in the optical image; determining matching target areas and target points in the at least one object area and at least one object point; and determining a conversion relationship between the target area and the target point to determine a conversion relationship between pixel points in the optical image and coordinate points in the radar image, thereby calibrating the object area and the object point.

[0005] To solve the above technical problem, the second technical solution adopted in this application is to provide an electronic device. The electronic device includes a memory and a processor, wherein the memory is used to store program data, and the program data can be executed by the processor to implement the method described in the first technical solution.

[0006] To solve the above technical problems, the third technical solution adopted by this application is to provide a computer-readable storage device that stores program data and can be executed by a processor to implement the method described in the first technical solution.

[0007] The beneficial effects of the present application are as follows: the present application processes the optical image and the radar image together, determines the matching relationship between the target in the optical image and the radar image, determines the matching relationship of the target's coordinate point information based on the matching relationship, and calculates the coordinate transformation relationship based on the target's coordinate point information, thereby achieving matching correspondence between the target in the optical image and the target in the radar image. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0009] Figure 1 This is a schematic diagram of an optical image of the present application;

[0010] Figure 2 This is a schematic diagram of a radar image of the present application;

[0011] Figure 3 This is a flow chart of the first embodiment of the data calibration method of this application;

[0012] Figure 4 This is a flow chart of the second embodiment of the data calibration method of this application;

[0013] Figure 5 This is another optical image schematic diagram of the present application;

[0014] Figure 6 It is a schematic diagram of the topological relationship of the optical image of this application;

[0015] Figure 7 This is a schematic diagram of the radar image and its topological relationship in this application;

[0016] Figure 8 This is a flow chart of the third embodiment of the data calibration method of this application;

[0017] Figure 9 This is a flow chart of the fourth embodiment of the data calibration method of the present application;

[0018] Figure 10 This is a flowchart of the fifth embodiment of the data calibration method of the present application;

[0019] Figure 11 This is a flow chart of the sixth embodiment of the data calibration method of the present application;

[0020] Figure 12 This is a flow chart of the seventh embodiment of the data calibration method of this application.

[0021] Figure 13 This is a schematic structural diagram of the first embodiment of the electronic device of the present application;

[0022] Figure 14 It is a structural diagram of the first embodiment of the computer-readable storage device of the present application. DETAILED DESCRIPTION

[0023] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0024] The terms "first," "second," and the like in this application are used to distinguish between different objects, not to describe a particular order. Furthermore, the terms "including," "having," and any variations thereof, are intended to cover non-exclusive inclusions. For example, a process, method, system, product, or apparatus comprising a series of steps or elements is not limited to the listed steps or elements, but may optionally include steps or elements not listed, or may optionally include other steps or elements inherent to the process, method, product, or apparatus.

[0025] References herein to "embodiments" mean that a particular feature, structure, or characteristic described in connection with the embodiments may be included in at least one embodiment of the present application. The appearance of this phrase in various places in the specification does not necessarily refer to the same embodiment, nor does it constitute an independent or alternative embodiment that is mutually exclusive of other embodiments. It is understood, both explicitly and implicitly, by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0026] In modern society, security protection is gaining increasing public attention, leading to the emergence of a growing number of security-related products. The application areas of security protection are also expanding, and related technologies are constantly advancing. Traditional security terminal equipment primarily relies on visible light cameras. Their advantage lies in obtaining target classification information, but they struggle to obtain information about the target's movement. For example, while they can easily identify the target type—human or machine—they struggle to estimate the target's speed and spatial position. Target recognition accuracy is lower at long distances, at night, and in rainy and foggy conditions compared to that of close-range targets. This means that the temporal and spatial accuracy of targets within the monitored area is inconsistent, and can even vary significantly. Millimeter-wave radar actively transmits electromagnetic waves and receives signals of the same frequency. It has a very high detection probability for moving objects or those with a large RCS (radar cross section), but a lower detection probability (not zero) for stationary objects. Millimeter-wave radar operates 24 hours a day and is less affected by weather.

[0027] Radar acquires measurement information (spatial position and velocity) of moving targets with a high probability of detection, but radar cannot achieve a high target recognition rate. Video / images can obtain highly accurate target recognition information, but are less likely to obtain information about the target's motion and spatial position. Radar and video sensors are two distinct devices, although both can detect targets. Radar's advantage lies in its ability to obtain accurate spatial and motion information, while video's advantage lies in its high-precision target recognition accuracy. However, video cannot always achieve high target recognition accuracy, for example, due to sparse tree cover, long distances, and poor lighting. Therefore, integrated radar and video surveillance technology, which combines radar and video, is gaining increasing attention. By effectively fusing radar and video data, leveraging their strengths and complementing their weaknesses, we can achieve even higher target recognition accuracy and obtain corresponding information about the target's motion and spatial position.

[0028] In surveillance areas such as roads, bridges, parks, and squares, target types are acquired through video, while spatial position and speed are determined through radar. Radar and video targets are then correctly matched and fused. Based on this correct fusion, data services can be provided for target behavior recognition (e.g., parallel driving, driving against traffic, tailgating, dangerous behavior, and traffic violations).

[0029] However, it should be noted that a radar target is a point target, which can be understood as a point position within a coordinate system. The image of a video target is a rectangular area, which corresponds to a quadrilateral area within the coordinate system. Accurately matching and calibrating the video target image with the radar target can be achieved through the data calibration method proposed in this application. The specific implementation steps can be implemented through the following examples.

[0030] The principle of the data calibration method of this application is as follows: Figure 1 As shown, Figure 1 The image captured by the video sensor contains multiple vehicles. The video sensor can accurately detect the number of vehicles and the corresponding image frame positions, and the video tracking algorithm can output the position of the target image frame in real time. Selecting a suitable point in the image frame as a replacement can establish a time coordinate sequence:

[0031] t i , BoxPt i , t i+1 , BoxPt i+1 , t i+2 , BoxPt i+2 、……

[0032] Among them, BoxPt i =(u i ,vi )=U i Indicates the position of a certain point in the video target frame, t i represents time, and i represents index.

[0033] like Figure 2 As shown, Figure 2 is the image acquired by the radar sensor. Similarly, a similar time coordinate sequence can be obtained for the target points in the radar image:

[0034] t i , CartPt i , t i+1 , CartPt i+1 , t i+2 , CartPt i+2 、……

[0035] Among them, CartPt i =(x i ,y i )=X i Indicates the position of a certain point of the radar target.

[0036] By combining the two, we can obtain the corresponding relationship between the position of a certain point in the video target frame and a certain point on the radar target at the same time.

[0037] t i , BoxPt i , CartPt i , t i+1 , BoxPt i+1 , CartPt i+1 , t i+2 , BoxPt i+2 , CartPt i+2 、…

[0038] The mapping relationship between a certain point in the video target frame and the radar target point can be obtained based on the coordinate data set. Based on this mapping relationship, the calibration correspondence between the video target and the radar target can be achieved.

[0039] like Figure 3 As shown, Figure 3 This is a flow chart of the first embodiment of the data calibration method of this application. The method includes the following steps:

[0040] S11: Acquire optical images and radar images of the target area.

[0041] Acquire an optical image and a radar image of the target area. The target area is the area monitored by the radar sensor and the video sensor. The optical image is the image captured by the video sensor, and the radar image is the image captured by the radar sensor. The radar image includes at least one object point. The optical image and radar image must be taken at the same time to ensure accurate target matching.

[0042] S12: Perform target detection on the optical image to determine at least one object area in the optical image.

[0043] An optical image is acquired, wherein the optical image includes at least one object region. The object region may be, for example, an image region of a vehicle. This object region is used for subsequent calculation of the transformation relationship. Therefore, at least one object region is required in the optical image. The more object regions there are, the more accurate the obtained transformation relationship.

[0044] S13: Determine a matching target region and target point among the at least one object region and the at least one object point.

[0045] After acquiring the object area and the object point in the radar, the object area and the object point are matched, and the successfully matched object area and object point are used as the target area and target point for subsequent conversion relationship calculation. The detailed matching process can be implemented in the following second embodiment.

[0046] S14: Determine a conversion relationship between the target area and the target point, so as to determine a conversion relationship between the pixel points of the optical image and the coordinate points of the radar image, so as to calibrate the object area and the object point.

[0047] After acquiring multiple pairs of target areas and target points, the coordinate mapping relationship is calculated based on the coordinate information of the target areas and target points. The coordinate conversion coefficient between a point in the target area and a target point is obtained. This coefficient is also the conversion coefficient between the optical image pixel point and the radar image coordinate point, which can represent the conversion relationship between the two. This allows the calibration of the object area in the optical image and the corresponding object point in the radar image.

[0048] like Figure 4 As shown, Figure 4 This is a flow chart of the second embodiment of the data calibration method of this application. This method is a further extension of step S13. The method includes the following steps:

[0049] S21: Establish a first topological map according to at least one object area.

[0050] Obtain the object area in the optical image, and establish a topological map based on the object area to represent the topological relationship between the object areas. Figure 5 As shown, Figure 5 An optical image is obtained, in which there are 6 object areas. Figure 6 Based on Figure 5 The obtained topology diagram includes multiple directions, such as left, right, up, down, upper left, lower left, upper right, lower right, etc.

[0051] S22: Establish a second topological map according to at least one object point.

[0052] Similarly, the topological relationship diagram of the target points can be obtained based on the radar image, such as Figure 7 shown.

[0053] S23: Matching the first topological map and the second topological map to determine matching target areas and target points.

[0054] According to the topological relationship reflected by the topological relationship graph of the object area and the topological relationship graph of the object point, the object area and the object point can be matched based on the node position information of the same topological relationship. For example, Figure 6 、 Figure 7 In the topological relationship, the matching relationships of A-1, B-2, C-3, D-4, F-5, and E-6 can be obtained. Select a point in the target area to match the target point to obtain the corresponding coordinate pair.

[0055] like Figure 8 As shown, Figure 8 This is a flow chart of the third embodiment of the data calibration method of this application. This method is a further extension of step S14. The method includes the following steps:

[0056] S31: Calculating coordinate conversion coefficients based on the coordinates of at least four coordinate pairs.

[0057] Get multiple coordinate pairs to calculate the coordinate conversion coefficient. The relevant calculation steps are as follows:

[0058] Assume that the coordinate information of the lower edge midpoint of the target frame of the target area in the coordinate system of the optical image is (u i ,v i ), and the coordinate information of the target point in the radar image coordinate system is (x i ,y i ), i represents the index.

[0059] Then the following homogeneous equation can be constructed:

[0060]

[0061] Through further exchange, we can get:

[0062] x i =a 11 ·ui +a 12 ·v i +a 13 (2)

[0063] y i =a 21 ·u i +a 22 ·v i +a 23 (3)

[0064] 1=a 31 ·u i +a 32 ·v i +a 33 (4)

[0065] By treating equations (2), (3) and (4) equivalently, we can obtain:

[0066]

[0067]

[0068] Then convert the unknown coordinate values ​​into proportional values ​​and solve them:

[0069]

[0070]

[0071] Formulas (7) and (8) are the calculation formulas for converting the midpoint of the lower edge of the target area into the target point. Further construction of the coefficient matrix can be obtained:

[0072]

[0073] Where W = [b 11 , b 12 , b 13 , b 21 , b 22 , b 23 , b 21 , b 32 ] T .

[0074] From this, we can see that there are 8 unknown quantities that need to be solved for the conversion coefficient. Based on the information of the four pairs of coordinate points obtained, these 8 unknown quantities can be correspondingly solved for the conversion coefficient. The coordinate information of the four pairs of coordinate points obtained is used to construct the following measurement equation:

[0075]

[0076] Simplify formula (10) as follows:

[0077] AW=b (11)

[0078] The final coordinate transformation coefficient is:

[0079] W=(A T A) -1 A T b (12)

[0080] The conversion between the midpoint of the lower edge of the target area and the target point can be achieved based on the coordinate conversion coefficient.

[0081] like Figure 9 As shown, Figure 9 This is a flow chart of a fourth embodiment of the data calibration method of this application. This method is a further extension of step S14 and includes the following steps:

[0082] S41: Randomly assign coordinate pairs into calibration set and test set.

[0083] Obtain multiple coordinate pairs and randomly assign and combine them to form a calibration set and a test set. The calibration set and the test set have the same number of coordinate pairs. Divide the coordinate pairs into multiple sets with the same number of coordinate pairs. Select one of them as the calibration set and the remaining sets as the test sets. Each calibration set can correspond to at least one test set. Each set must contain at least four coordinate pairs.

[0084] S42: Calculate the coordinate conversion coefficient based on the calibration set, and calculate the average calibration error of the coordinate conversion coefficient on the test set based on the test set.

[0085] The coordinate conversion coefficients can be calculated based on the coordinate information of the coordinate pairs in the calibration set. The corresponding calculation steps can be referred to the above embodiment and will not be repeated here. After obtaining the coordinate conversion coefficients, the coordinate conversion coefficients are brought into the test set to calculate the average calibration error. The calculation process is as follows:

[0086]

[0087]

[0088]

[0089]

[0090] Among them, e i represents the calibration error of the target, represents the average calibration error on this test set.

[0091] S43: Setting an average calibration error change threshold.

[0092] Before determining whether the average calibration error has changed, a threshold is set to determine whether the average calibration error has changed significantly. The description here does not limit the implementation time of the setting step, and it can be before the determination step S44.

[0093] S44: Determine whether the average calibration error changes significantly.

[0094] When a coordinate conversion coefficient and the corresponding average calibration error of the test set are obtained, the coordinate pairs are randomly redistributed and divided into several coordinate pair sets with the same number of coordinate pairs as the previous set. A coordinate pair set different from the previous calibration set is selected as a new calibration set, and the remaining coordinate pair sets are selected as new test sets. The coordinate coefficients and average calibration errors of the new calibration set and the test set are calculated again. It is determined whether the difference in the obtained average calibration error is greater than the set average calibration error threshold. If so, it indicates that there is a problem with the data set, resulting in the calibration result being related to the division of the data set, and data inconsistency occurs. Therefore, it is necessary to re-collect data and calculate the coordinate conversion coefficient again, that is, re-execute step S11. If not, it indicates that the coordinate conversion coefficient is accurate, and step S45 is executed. The process of re-random allocation is performed at least once. The more times the allocation is performed, the higher the accuracy of the obtained coordinate conversion coefficient can be guaranteed.

[0095] S45: Using the coordinate conversion coefficient as the conversion coefficient corresponding to the matching of the target area and the target point.

[0096] When the change in the average calibration error of the reacquired data set is less than the preset average calibration error threshold, the coordinate conversion coefficients obtained for the first time are used as the calibration relationship for matching the targets in the optical image and the radar image, realizing the conversion between the target area and the target point.

[0097] like Figure 10 As shown, Figure 10 This is a flow chart of a fifth embodiment of the data calibration method of the present application, which includes the following steps:

[0098] S51: Acquire an optical image and a radar image of the target area.

[0099] An optical image and a radar image of a target area are acquired at the same time, and the radar image includes at least one object point.

[0100] S52: Perform target detection on the optical image to determine at least one object area in the optical image.

[0101] The acquired optical image is detected to identify at least one object region within it, matching it to an object point in the radar image. In the transportation field, the objects detected can be vehicles. Each detected target vehicle forms an object region encompassing the target vehicle.

[0102] S53: Determine whether there is mutual occlusion between at least one object region in the optical image.

[0103] Mutual occlusion can be defined as overlapping object areas, or when the distance between the lower edge of one object area and the upper edge of another object area in the same lane is less than a preset value. The specific definition of occlusion can be freely set based on actual conditions.

[0104] If there is an occlusion, step S51 is executed to reacquire the image data.

[0105] If there is no occlusion, step S54 is executed.

[0106] S54: Determine a matching target region and target point among the at least one object region and the at least one object point.

[0107] After obtaining the object area and the object points in the radar, the object area and the object points are matched, and the successfully matched object area and object points are used as the target area and target points used in subsequent conversion relationship calculations.

[0108] S55: Determine a conversion relationship between the target area and the target point, so as to determine a conversion relationship between the pixel points of the optical image and the coordinate points of the radar image, so as to calibrate the object area and the object point.

[0109] After acquiring multiple pairs of target areas and target points, the coordinate mapping relationship is calculated based on the coordinate information of the target areas and target points. The coordinate conversion coefficient between a point in the target area and a target point is obtained. This coefficient is also the conversion coefficient between the optical image pixel point and the radar image coordinate point, which can represent the conversion relationship between the two. This allows the calibration of the object area in the optical image and the corresponding object point in the radar image.

[0110] like Figure 11 As shown, Figure 11 This is a flow chart of a sixth embodiment of the data calibration method of the present application, which includes the following steps:

[0111] S61: Acquire an optical image and a radar image of the target area.

[0112] An optical image and a radar image of a target area are acquired at the same time, and the radar image includes at least one object point.

[0113] S62: Perform target detection on the optical image to determine at least one object area in the optical image.

[0114] The acquired optical image is detected to identify at least one object region within it, matching it to an object point in the radar image. In the transportation field, the objects detected can be vehicles. Each detected target vehicle forms an object region encompassing the target vehicle.

[0115] S63: Determine whether the number of at least one object region and at least one object point is consistent.

[0116] If they are inconsistent, step S61 is executed to reacquire the image data.

[0117] If they are consistent, step S64 is executed.

[0118] S64: Determine a matching target region and target point among the at least one object region and the at least one object point.

[0119] An equal number of object regions and object points are used to be matched with each other to obtain a target region and a target point.

[0120] S65: Determine a conversion relationship between the target area and the target point, so as to determine a conversion relationship between the pixel points of the optical image and the coordinate points of the radar image.

[0121] After obtaining multiple pairs of target areas and target points, the coordinate mapping relationship is calculated based on the coordinate information of the target areas and target points to obtain the coordinate conversion coefficient between a point in the target area and a target point. This coefficient is also the conversion coefficient between the optical image pixel point and the radar image coordinate point, which can represent the conversion relationship between the two.

[0122] like Figure 12 As shown, Figure 12 This is a flow chart of the seventh embodiment of the data calibration method of the present application, which includes the following steps:

[0123] S71: Initialize parameters such as thresholds.

[0124] Set parameters such as calibration error threshold, average calibration error threshold, time threshold, etc.

[0125] S72: Collect optical images and radar images.

[0126] S73: Confirm whether the time of the optical image and the radar image is synchronized.

[0127] If they are synchronized, then steps S74 and S75 are executed. If they are not synchronized, then step S72 is executed to re-collect image data.

[0128] S74: Extracting object points in the radar image.

[0129] S75: Extracting the object region in the optical image.

[0130] S76: Determine whether the object area is blocked.

[0131] If there is no occlusion, then step S77 is executed. If there is occlusion, then step S72 is executed to re-collect image data.

[0132] S77: Determine whether the number of object points is consistent with the number of object areas.

[0133] If they are consistent, then step S78 is executed. If they are inconsistent, then step S72 is executed to re-acquire image data.

[0134] S78: Save data.

[0135] The data of the object area in the optical image of the frame and the corresponding object point in the radar image are saved.

[0136] S79: Topological relationship construction.

[0137] S80: Obtain a matching pair of target area and target point.

[0138] S81: Data accumulation.

[0139] Accumulate multiple frames of data for subsequent calculations.

[0140] S82: Establish a calibration set and a test set.

[0141] S83: Solve the coordinate transformation coefficient and calibration error.

[0142] S84: Determine whether the calibration error is greater than a threshold.

[0143] If not, then go to step S85. If greater, then go to step S72 to reacquire image data.

[0144] S85: Data consistency judgment.

[0145] Data consistency is determined by the change in the average calibration error. For details, see the fourth embodiment above. If the average calibration error is not greater than the set threshold, it indicates that no significant change has occurred, and step S86 is executed. If it is greater than the threshold, step S72 is executed to recapture the image data.

[0146] S86: Save the coordinate conversion coefficient.

[0147] S87: Position consistency judgment.

[0148] At regular intervals, data is reacquired, and the calibration and test sets are assigned to calculate the coordinate conversion coefficients and average calibration error. A determination is made as to whether the average calibration error has changed significantly. If the average calibration error is less than a set threshold, it indicates no significant change, and step S88 is executed. If it is greater than the threshold, indicating a change in the test environment, step S72 is executed to reacquire image data.

[0149] S88: The process ends.

[0150] The same or similar steps in this embodiment refer to the description of the above embodiment and are not repeated here.

[0151] like Figure 13 As shown, Figure 13 This is a schematic structural diagram of a first embodiment of an electronic device of the present application. The electronic device includes: a processor 110 and a memory 120.

[0152] The processor 110 controls the operation of the electronic device and may also be referred to as a CPU (Central Processing Unit). The processor 110 may be an integrated circuit chip with the ability to process signal sequences. The processor 110 may also be a general-purpose processor, a digital signal sequence processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic device, a discrete gate or transistor logic device, or a discrete hardware component. The general-purpose processor may be a microprocessor or any conventional processor.

[0153] The memory 120 stores instructions and data required for the processor 110 to operate.

[0154] The processor 110 is configured to execute instructions to implement the method described in any one of the first to third embodiments of the course management method of the present application and possible combined methods.

[0155] like Figure 14 As shown, Figure 14 This is a schematic diagram of the structure of the first embodiment of the computer-readable storage device of the present application. The computer-readable storage device includes a memory 210, which stores program data. When the program data is executed, the method provided by any embodiment and possible combination of the course management method of the present application is implemented.

[0156] The memory 210 may include a read-only memory (ROM), a random access memory (RAM), a flash memory, a hard disk, an optical disk, and the like.

[0157] In summary, through the above embodiments, the present application processes the optical image and the radar image together, determines the matching relationship between the optical image and the target in the radar image, determines the matching relationship of the target's coordinate point information based on the matching relationship, and calculates the coordinate transformation relationship based on the target's coordinate point information, thereby achieving matching correspondence between the target in the optical image and the target in the radar image.

[0158] In the several embodiments provided in this application, it should be understood that the disclosed methods and devices can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the modules or units is merely a logical functional division. In actual implementation, other division methods may be used, such as combining or integrating multiple units or components into another system, or ignoring or not implementing certain features.

[0159] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of this embodiment.

[0160] In addition, each functional unit in each embodiment of the present application may be integrated into a processing unit, each unit may exist physically separately, or two or more units may be integrated into a single unit. The above-mentioned integrated units may be implemented in the form of hardware or software functional units.

[0161] If the integrated units in the above other embodiments are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) or a processor to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0162] The above description is merely an embodiment of the present application and does not limit the patent scope of the present application. Any equivalent structure or equivalent process transformation made using the contents of the present application specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present application.

Claims

1. A data calibration method, characterized in that: The method comprises: Acquire an optical image and a radar image of a target area; wherein the radar image includes at least one object point; performing target detection on the optical image to determine at least one object region in the optical image; determining a matching target region and target point among the at least one object region and the at least one object point; determining a conversion relationship between the target area and the target point, so as to determine a conversion relationship between pixel points of the optical image and coordinate points of the radar image, so as to calibrate the target area and the target point; The determining of the matching target area and target point among the at least one object area and the at least one object point includes: Creating a first topological map based on the at least one object region in the optical image; and establishing a second topological map according to the at least one object point in the radar image, wherein the optical image and the radar image are images taken at the same time; Matching the first topological map and the second topological map to determine matching target areas and target points; The determining of the matching target area and target point among the at least one object area and the at least one object point further includes: determining whether the number of the at least one object area is consistent with the number of the at least one object point; If not, reacquire the optical image and the radar image of the target area.

2. The data calibration method according to claim 1, characterized in that: The matching the first topological map and the second topological map to determine the matching target area and target point includes: The coordinates of a point in the target area are matched with the coordinates of the matched target point to obtain a coordinate pair.

3. The data calibration method according to claim 2, characterized in that: Determining the conversion relationship between the target area and the target point includes: A coordinate conversion coefficient is calculated based on the coordinates of at least four of the coordinate pairs.

4. The data calibration method according to claim 3, characterized in that: The determining of the conversion relationship between the target area and the target point further includes: Randomly allocating the coordinate pairs into a calibration set and a test set, wherein one calibration set corresponds to at least one test set, and the calibration set and the test set contain an equal number of coordinate pairs; Calculating a coordinate transformation coefficient based on the calibration set, and calculating an average calibration error of the coordinate transformation coefficient on the test set based on the test set; Determining whether the average calibration error changes significantly; If so, reacquire the data and calculate the coordinate conversion coefficient.

5. The data calibration method according to claim 4, characterized in that: The determining whether the average calibration error is less than before the significant change further includes: Setting an average calibration error change threshold; Determining whether the average calibration error changes significantly includes: If the change in the average calibration error is greater than the average calibration error change threshold, the average calibration error has changed significantly.

6. The data calibration method according to claim 1, characterized in that: The performing target detection on the optical image to determine at least one object area in the optical image includes: determining whether there is mutual occlusion between the at least one object region in the optical image; If so, reacquire the optical image and the radar image of the target area.

7. An electronic device, characterized in that: The system comprises a memory and a processor, wherein the memory is used to store program data, and the program data can be executed by the processor to implement the method according to any one of claims 1 to 6.

8. A computer-readable storage device, characterized in that: Program data is stored and can be executed by a processor to implement the method according to any one of claims 1 to 6.

Images