Vehicle positioning method and device, controller, medium, program product and vehicle

Abstract

The invention relates to a vehicle positioning method and device, a controller, a medium, a program product and a vehicle, the vehicle is equipped with positioning equipment, and the positioning method comprises the following steps: performing feature matching on a target image and a field image acquired by the vehicle in real time; a position of the vehicle is determined based on the coordinates of the positioning device. On the basis, the problem that the vehicle depends on GNSS positioning in a GNSS signal missing scene can be solved, and the blank of vehicle position judgment when no positioning signal exists is filled.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of vehicle positioning, in particular to a vehicle positioning method, device, controller, medium, program product and vehicle. BACKGROUND

[0002] Road mapping needs to collect road data at specified locations to update ground features. Existing solutions rely on vehicles

[0003] GNSS positioning determines whether the vehicle enters the collection area. However, in actual application, the GNSS signal is often missing or invalid in the collection area. Although the vehicle is equipped with a positioning device that can continuously obtain coordinates, it cannot determine whether the vehicle enters the collection area, and can only collect all data, causing redundancy and high cost problems. SUMMARY

[0004] The embodiments of the present application provide a vehicle positioning method, device, controller, medium, program product and vehicle to solve the problem of full data collection causing redundancy and high cost due to the missing or invalid GNSS signal in the collection area in the prior art.

[0005] The embodiments of the present application provide a vehicle positioning method, wherein the vehicle is equipped with a positioning device, and the positioning method comprises: performing feature matching on a target image and a live image collected by the vehicle in real time; and if the matching is successful, determining the position of the vehicle based on the coordinates of the positioning device.

[0006] In an embodiment of the present application, the feature matching on the target image and the live image collected by the vehicle in real time comprises: determining a target window based on first feature points of the target image, and determining a search window based on second feature points of the live image; calculating a Gaussian curvature correlation coefficient of the target window and the search window, and determining a feature point pair with the Gaussian curvature correlation coefficient greater than a threshold curvature, the feature point pair being the first feature points and the second feature points satisfying a corresponding relationship; correcting the live image based on coordinate information of the feature point pair; and matching the corrected live image and the target image.

[0007] In an embodiment of the present application, the matching of the corrected live image and the target image comprises: calculating a gray correlation coefficient of the corrected live image and the target image; determining a comprehensive correlation coefficient and a feature point matching rate of the corrected live image and the target image based on the gray correlation coefficient and the Gaussian curvature correlation coefficient; and when the comprehensive correlation coefficient and the feature point matching rate are both higher than a preset threshold, it is determined that the matching of the target image and the live image is successful.

[0008] In an embodiment of the present application, the calculating the Gaussian curvature correlation coefficients of the target window and the search window, and determining the feature point pairs with the Gaussian curvature correlation coefficients greater than a threshold curvature comprises: determining a first gray plane function of the target window and a second gray plane function of the search window respectively; determining a first Gaussian curvature set of all feature points in the target window based on the first gray plane function, and determining a second Gaussian curvature set of all feature points in the search window based on the second gray plane function; determining the Gaussian curvature correlation coefficients of the target window and the search window based on the first Gaussian curvature set and the second Gaussian curvature set; and determining the feature point pairs with the Gaussian curvature correlation coefficients greater than a threshold curvature.

[0009] In an embodiment of the present application, the correcting the live image based on the coordinate information of the feature point pairs comprises: determining a geometric distortion window based on the target window, wherein the geometric distortion window comprises at least three groups of the feature point pairs; constructing an error equation based on the feature point pairs to calculate geometric distortion parameters; and correcting the live image based on the geometric distortion parameters.

[0010] In an embodiment of the present application, after the correcting the live image based on the geometric distortion parameters, the method further comprises: performing resampling processing on the corrected live image.

[0011] In an embodiment of the present application, a pixel size of the geometric distortion window is greater than a pixel size of the target window.

[0012] In an embodiment of the present application, the target window and the search window have the same size, and the target window and the search window have a preset odd pixel size in width and height.

[0013] In an embodiment of the present application, before the performing feature matching on the target image and the live image collected by the vehicle in real time, the method further comprises: converting the target image into a gray image; calculating an information amount of the gray image; and determining an image with an information amount greater than a preset threshold as an effective target image.

[0014] In an embodiment of the present application, the vehicle positioning method further comprises: if the matching is successful, starting data collection of the vehicle, and uploading the collected target image to a server; and if the matching fails, not starting data collection of the vehicle.

[0015] Correspondingly, an embodiment of the present application provides a controller comprising one or more processors and a memory, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the steps of the vehicle positioning method described above.

[0016] Accordingly, this application provides a storage medium, characterized in that it includes a computer program, which, when run on a controller, causes the controller to perform the steps of the vehicle positioning method described above.

[0017] Accordingly, this application provides a computer program product, including a computer program or instructions, which, when executed by a processor, implement the steps of the vehicle positioning method described above.

[0018] Furthermore, this application embodiment also provides a vehicle, which includes the controller described above.

[0019] The vehicle positioning method, device, controller, medium, program product, and vehicle provided in this application perform feature matching between the target image and the on-site image collected in real time by the vehicle. If the matching is successful, the vehicle positioning coordinates at the corresponding time are obtained through the vehicle positioning device to determine the vehicle position. This can solve the dependence of vehicles on GNSS positioning in the absence of GNSS signal, fill the gap in vehicle position determination when there is no positioning signal, and replace the full collection mode. It can reduce redundant data from the source, reduce the traffic consumption of vehicle mobile network and the overall collection cost, and ensure the relevance and economy of road mapping data collection. Attached Figure Description

[0020] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0021] Figure 1 This is a flowchart illustrating one embodiment of the vehicle positioning method of this application;

[0022] Figure 2 This is a schematic diagram of the grayscale conversion of the target image in this application;

[0023] Figure 3 This is a schematic diagram illustrating the determination of the valid target image in this application;

[0024] Figure 4 This application Figure 1 A flowchart illustrating an implementation method for step S100;

[0025] Figure 5 This is a schematic diagram of one embodiment of the target window and search window in this application;

[0026] Figure 6 This application Figure 4 A flowchart illustrating an embodiment of step S120;

[0027] Figure 7 is a flowchart of an embodiment of step S130 in the present application Figure 4

[0028] Figure 8 is a flowchart of an embodiment of step S140 in the present application Figure 4

[0029] Figure 9 is a schematic diagram of a vehicle positioning device module provided by an embodiment of the present application

[0030] Figure 10 is a schematic diagram of a controller in an embodiment of the present application DETAILED DESCRIPTION

[0031] The present application will be further described below in conjunction with the drawings and embodiments. It is particularly pointed out that the following embodiments are only for the purpose of illustrating the present application, but do not limit the scope of the present application. Similarly, the following embodiments are only part of the embodiments of the present application, but not all the embodiments of the present application, and all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0032] The terms "first", "second", "third" in the present application are only for the purpose of description, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features limited by "first", "second", "third" can explicitly or implicitly include at least one of the features. In the description of the present application, the meaning of "a plurality of" is at least two, such as two, three, etc., unless otherwise specifically limited. All directional indications (such as up, down, left, right, front, back, etc.) in the embodiments of the present application are only used to explain the relative positional relationship, movement condition, etc. between the components in a certain posture (as shown in the drawings), and if the certain posture changes, the directional indications also change accordingly. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device including a series of steps or units is not limited to the listed steps or units, but can optionally include steps or units not listed, or can optionally include other steps or units inherent to the process, method, product or device.

[0033] ​​Reference herein to "an implementation" means that a particular feature, structure, or characteristic described in connection with the implementation can be included in at least one implementation of the application. The appearances of the phrase in various places in the specification are not necessarily all referring to the same implementation nor are separate or alternative implementations mutually exclusive of other implementations. One of ordinary skill in the art will readily recognize from the disclosure herein, that the implementations described herein can be incorporated into other implementations.

[0034] Reference is made to Figure 1 , Figure 1 is a flowchart of an embodiment of the vehicle positioning method provided by the application, as Figure 1 The vehicle positioning method provided by the application comprises the following steps:

[0035] S100, feature matching is performed on the target image and the live image collected by the vehicle in real time.

[0036] It can be understood that, before the vehicle positioning method of the application is performed, the target image also needs to be preprocessed. The target image of the application can be selected from known road sign images, and the selection rule is that the image is clear and has distinct levels, and has no invalid area such as all white or all black. The target image of the road sign is preset, and the source of acquisition is a known image set, that is, before the vehicle performs the positioning task, an image that meets the conditions is selected from the collected road sign image library as the target image before the vehicle performs the positioning task, and is stored in the vehicle system or the associated server. That is, these known images are usually images collected historically and containing clear road sign features, which are determined as target images after screening (such as meeting the conditions of clarity, distinct levels, and information quantity higher than a threshold) and used for subsequent matching analysis with the live image collected by the vehicle in real time.

[0037] The live image is a road image collected by the vehicle in real time through a camera, and is used for matching analysis with the preset target image. In the application, the vehicle is equipped with a positioning device, and the positioning coordinates of the vehicle are obtained in real time during the driving of the vehicle.

[0038] Further, the target image is converted into a gray-scale image.

[0039] Reference is further made to Figure 2 , Figure 2 is a schematic diagram of the gray-scale conversion of the target image of the application. The selected target image is a color image, which needs to be further converted into a gray-scale image. The conversion method uses the gray-scale conversion formula:

[0040]

[0041] Further, the information quantity of the gray-scale image is calculated.

[0042] It can be understood that, for a target image with n gray values, the gray g i The probability of occurrence is p i , the frequency of the gray scale can be approximated, and the information amount H is represented as:

[0043] f i The frequency of the gray g i

[0044] The image with an information amount greater than the threshold value is determined as an effective target image.

[0045] Please further combine Figure 3 , Figure 3 is a schematic diagram of the determination of the effective target image of the present application. It can be understood that, if the image texture is not rich enough and the gray scale distribution is single, the information amount is lower than the threshold value, and it is set as an invalid area, and the target image needs to be selected again.

[0046] It can be understood that the information amount of the present application is essentially the image gray scale entropy based on the statistical characteristics of the gray scale distribution of the target image, which corresponds to the Shannon entropy in information theory, and is used to quantify the texture richness and gray scale diversity of the target image. It is a key indicator for screening effective target images that can be used for subsequent feature extraction and matching. In a specific application scenario of the present application, the information amount threshold value can be set to 30. If the information amount H is lower than the preset threshold value, it is determined as an invalid area, and the target image needs to be selected again. Only the effective target images with an information amount H higher than the threshold value are retained. It can be understood that, a high information amount image has a diverse gray scale distribution, such as a road sign image, which contains different gray scale regions such as sign background color, text, and patterns. The gray scale frequency pi is uniformly distributed, the H value is higher according to the formula, the image texture is rich, and enough effective feature points can be extracted by the Moravec operator to provide a reliable basis for subsequent window matching and Gaussian curvature calculation. A low information amount image refers to an image with a single gray scale distribution, such as a full white or full black image, or an image blurred to only 1-2 gray scales. The gray scale frequency pi is highly concentrated, and the calculated H value is very low, such as less than 30. The image has no effective texture, and cannot extract feature points that meet the matching requirements. If used for subsequent matching, it will lead to matching failure, and therefore needs to be set as an invalid area and the target image needs to be selected again.

[0047] Of course, in other embodiments, the information amount threshold value can also be set to other values based on specific application scenarios, which are not limited here.

[0048] ​In the above embodiments, by preprocessing the target image, the interference of color differences on subsequent feature matching can be avoided, the complexity of data processing can be reduced, and the subsequent feature extraction and matching process can be based on images with sufficient texture features from the data source, reducing the amount of invalid calculations, thereby improving the accuracy of subsequent geometric distortion correction and comprehensive correlation coefficient matching, and ultimately ensuring the reliability of vehicle positioning results.

[0049] Furthermore, please combine this further. Figure 4 , Figure 4 This application Figure 1 A flowchart illustrating an implementation method for step S100 is shown below. Figure 4 Step S100 further includes the following sub-steps:

[0050] S110, determine the target window based on the first feature points of the target image, and determine the search window based on the second feature points of the scene image.

[0051] In this embodiment, the Moravec operator can be used to extract a feature point from the target image and the scene image respectively, which can be used as the core reference point for subsequent matching.

[0052] The target window of the target image is determined based on the first feature point of the target image, and the search window of the scene image is determined based on the second feature point of the scene image.

[0053] Further integration Figure 5 , Figure 5 This is a schematic diagram of one embodiment of the target window and search window of this application. The target window of the image is determined with the current first feature point as its center. In this embodiment, the target window can be a preset odd-numbered pixel size. This application describes the application in detail using an example where the target window has 5 pixels horizontally and 5 pixels vertically. The current first feature point can be a random feature point in the target image; no specific limitation is made here.

[0054] Furthermore, the search window is determined based on the second feature points of the scene image, and in this application, the target window and the search window are the same size, both being a preset odd-dimensional pixel size, i.e., 5x5, 25 grids.

[0055] In the above implementation, the target window and the search window are the same size, which ensures that the feature calculation of the target window and the search window is based on the same number of pixel samples, avoiding matching errors caused by differences in sample size, and providing a unified benchmark for subsequent correlation coefficient calculation.

[0056] S120, calculate the Gaussian curvature correlation coefficient between the target window and the search window, and determine the feature point pairs whose first Gaussian curvature correlation coefficient is greater than the threshold curvature.

[0057] Please further combine Figure 6 , Figure 6 is an embodiment of step S120 in the present application Figure 4 , as shown in the flow chart of FIG. 2. Figure 6 Step S120 further comprises the following sub-steps:

[0058] S121, respectively determining a first gray surface function of the target window and a second gray surface function of the search window.

[0059] In the present application, the image surface is described by a gray surface function. The gray values in each target window and search window are not distributed in disorder, but present certain spatial variation characteristics, such as the gradient trend of the gray with the pixel coordinates (x, y). In the present application, the gray variation law is mathematically modeled, and the gray function of the target window in the small block shadow is (expressed by a quadratic surface function):

[0060] f(x, y) = a0 + a1x + a2y + a3x 2 +a4y 2 +a5xy

[0061] wherein a0, a1, a2, a4 and a5 are coefficients to be calculated.

[0062] Further, the error equation is:

[0063]

[0064] wherein v m is the correction number, H m is the gray value, a'0, a'1, a'2, a'3, a'4, a'5 are unknown parameter estimation values, which are used to form the normal equation to solve the estimation values of the unknown parameters, and the first gray surface function of the target window is obtained as:

[0065] g(x, y) = a'0 + a'1x + a'2y + a'3x 2 +a'4y 2 +a'5xy

[0066] Similarly, based on the same method, the second gray surface function of the search window can be obtained as:

[0067] h(x, y) = b'0 + b'1x + b'2y + b'3x 2 +b'4y 2 +b'5xy

[0068] S122, determining a first Gaussian curvature set of all feature points in the target window based on the first gray surface function, and determining a second Gaussian curvature set of all feature points in the search window based on the second gray surface function.

[0069] Further, the first Gaussian curvature set c and the second Gaussian curvature set r of all feature points in the target window and the search window are calculated based on the first gray surface function and the second gray surface function respectively.

[0070] The Gaussian curvature is used to describe the concave-convex information of a surface, and the expression is as follows:

[0071]

[0072] The first Gaussian curvature set c of the target window is as follows:

[0073] c = {c1, c2, …, c n}

[0074] The second Gaussian curvature set r of the search window is as follows:

[0075] r = {r1, r2, …, r n}

[0076] S123, determining the Gaussian curvature correlation coefficient of the target window and the search window based on the first Gaussian curvature set and the second Gaussian curvature set.

[0077] Further, the Gaussian curvature correlation coefficient of the target window and the search window is determined as follows:

[0078] p1 = p(c, r)

[0079] S124, determining the feature point pair whose Gaussian curvature correlation coefficient is greater than a threshold curvature.

[0080] The feature point pair is the first feature point and the second feature point that satisfy the corresponding relationship, that is, the feature point pair that satisfies the preset condition and is composed of the feature point of the target image and the feature point of the field image. In the above application embodiment, the feature point pair with high similarity is retained by the correlation coefficient screening, which provides an effective reference point for subsequent geometric distortion correction and accurate matching, and the invalid feature point pair with low similarity is eliminated.

[0081] S130, correcting the field image based on the coordinate information of the feature point pair.

[0082] Please further combine Figure 7 , Figure 7 is the flowchart of an embodiment of step S130 in the present application Figure 4 , as Figure 7 Step S130 further includes the following sub-steps:

[0083] S131, determining a geometric distortion window based on the target window, the geometric distortion window including at least three groups of feature point pairs.

[0084] Optionally, in the embodiments of the present application, because the target window image and the search window image exist geometric distortion, it is necessary to calculate the geometric distortion parameters to correct the live image. In the present application, the pixel size of the geometric distortion window is defined to be larger than the pixel size of the target window. For example, if the size of the target window is 7x7, the size of the geometric distortion window can be, for example, 7x7 or 9x9.

[0085] Further, the geometric distortion window includes at least three sets of feature point pairs, wherein a pixel coordinate of a point on the target image is (x, y), and a coordinate of a corresponding homonymous image point on the search image is (x', y')

[0086] x' = a 00 +a 10 x+a 01 y

[0087] y' = b 00 +b 10 x+b 01 y

[0088] S132, based on the error equation constructed based on the feature point pairs, the geometric distortion parameters are calculated.

[0089] Based on the pixel coordinates of the above feature point pairs, an error equation is constructed, and the geometric distortion parameters are calculated. According to the polynomial coefficients obtained by solving, the coordinates of the feature points on the search image are calculated, and pixel resampling is performed. The resampling method uses bilinear interpolation.

[0090] S133, based on the geometric distortion parameters, the live image is corrected.

[0091] Specifically, based on the calculated geometric distortion polynomial coefficients (geometric distortion parameters), the live image is corrected for geometric distortion. For each first feature point in the target window, its pixel coordinates (x, y) are obtained, (x, y) is substituted into the geometric distortion polynomial model, and the theoretical corrected coordinates (x', y') of the feature point on the search image after eliminating distortion are calculated. By traversing all the first feature points in the target window, the batch calculation of the corresponding corrected coordinates of the feature points on the search image is completed, the accurate mapping of the target feature points and the live image spatial position is realized, and the image geometric misplacement caused by shooting angle deviation, lens optical distortion, etc. is eliminated.

[0092] In the above embodiments, the error equation is constructed based on the feature point pairs to calculate the distortion parameters, so that the geometric distortion correction has a clear parameter calculation basis, and the spatial deformation of the live image caused by shooting angle, lens distortion, etc. can be effectively eliminated, the spatial alignment of the corrected image and the target image is realized, the matching difficulty caused by image geometric difference is solved, and the spatial consistency of subsequent matching is improved.

[0093] Further, the application further includes resampling processing on the corrected on-site image after correcting the on-site image based on the geometric distortion parameters.

[0094] Since the theoretical coordinates (x', y') obtained after correction can be non-integers, i.e., located at the gaps of the pixel grid of the on-site image, pixel resampling is needed to obtain effective gray values through the bilinear interpolation method, so that all the corrected coordinates are traversed to complete the resampling, generate the corrected on-site image eliminating geometric distortion and having continuous pixels, and ensure that the image details are effectively preserved in the correction process. For example, after geometric distortion correction, the feature point correction coordinates of the on-site image can be non-integers (such as (3.2, 5.7)), and the image pixel coordinates need to be integers, so the gray values of the non-integer coordinates need to be estimated through the interpolation algorithm. In the embodiment of the application, the bilinear interpolation is adopted, the target non-integer coordinates are taken as the center, the four nearest integer coordinate pixels around the target non-integer coordinates are selected, and the gray value of the target point is calculated according to the distance weight, so that the corrected image not only retains the original texture features, but also eliminates the pixel misplacement caused by geometric distortion.

[0095] In the above embodiment, the resampling processing is added after the image correction, which can solve the problem of missing pixel information caused by non-integer coordinates after geometric distortion correction, and make the pixel arrangement of the corrected on-site image regular and the gray information continuous through resampling, so as to avoid the image texture distortion caused by pixel misplacement and ensure that the corrected image can accurately reflect the original scene features, provide clear and complete image data for accurate comparison of the gray, curvature and other features in the subsequent matching process, and further guarantee the matching accuracy.

[0096] S140, matching the corrected on-site image and the target image.

[0097] Please further combine Figure 8 , Figure 8 is an embodiment of step S140 in the application Figure 4 The flowchart of an embodiment of step S140 in the application is shown in FIG. 6. Figure 8 Step S140 further includes the following sub-steps:

[0098] S141, calculating the gray correlation coefficient of the corrected on-site image and the target image.

[0099] Further, all the feature points are traversed, a search window is slid on the on-site image, and the gray correlation coefficient p2 of the on-site image after geometric distortion correction and the target image is calculated.

[0100] S142, determining the comprehensive correlation coefficient of the corrected on-site image and the target image and the feature point matching rate based on the gray correlation coefficient and the Gaussian curvature correlation coefficient.

[0101] Further, the comprehensive correlation coefficient and the feature point matching rate of the corrected live image and the target image are determined based on the gray correlation coefficient and the Gaussian curvature correlation coefficient, and are respectively:

[0102] p = 0.5p1 + 0.5p2

[0103]

[0104] wherein m is the number of successfully matched feature points, and n is the number of all feature points.

[0105] S143, when the comprehensive correlation coefficient and the feature point matching rate are both higher than a preset threshold, it is determined that the target image and the live image are successfully matched.

[0106] In the present application, in the image matching logic of the vehicle positioning method, the comprehensive correlation coefficient and the feature point matching rate are complementary to each other in local feature quality verification and overall matching effectiveness statistics, and both of them work together to determine the final matching result.

[0107] Optionally, when the comprehensive correlation coefficient p is greater than a specified threshold, it is considered that the live image and the target image are currently corresponding to the matched feature points, i.e. a single pair of successfully matched feature points. The final feature point matching rate is the number of single pair of successfully matched feature points / the total number of high-confidence feature points, i.e. the screening result of the comprehensive correlation coefficient directly determines the numerator of the matching rate calculation. In the scheme, both of them need to be higher than the respective preset threshold at the same time, so as to finally determine that the image matching is successful. That is, when the feature matching rate σ is also greater than a specified threshold, it is considered that the live image and the target image are successfully matched, otherwise they are not matched.

[0108] In the present application, the comprehensive correlation coefficient can ensure the quality of a single matching point and avoid interference of low similarity points, and the feature matching rate can ensure the effectiveness of overall matching and avoid local accidental compliance. Both of them depend on each other and work together to verify, and together constitute the core determination standard of image matching, so as to avoid vehicle positioning deviation caused by single index misjudgment.

[0109] S200, if the matching is successful, the position of the vehicle is determined based on the coordinates of the positioning device.

[0110] Further, if the target image and the live image collected by the vehicle in real time are successfully matched, the positioning coordinates of the vehicle at the corresponding time are obtained based on the positioning device to determine the driving position of the vehicle.

[0111] In addition, if the target image and the live image collected by the vehicle in real time are successfully matched, the data collection task of the vehicle is started, and the collected target image is uploaded to the server. Otherwise, if the matching fails, the data collection is not started, so as to reduce redundant data and traffic consumption.

[0112] The above embodiment can solve the dependence of the vehicle on GNSS positioning in a GNSS signal missing scene, fill the blank of vehicle position determination without positioning signal, and replace the full-quantity acquisition mode to reduce redundant data from the source, reduce the consumption of vehicle mobile network traffic and the overall acquisition cost, and ensure the pertinence and economy of road mapping data acquisition.

[0113] According to the method described in the above embodiment, the embodiment of the application further provides a vehicle positioning device 200 for executing the steps in the above vehicle positioning method. Please refer to Figure 9 , Figure 9 is a module schematic diagram of the vehicle positioning device 200 provided by the embodiment of the application. The vehicle is equipped with a positioning device, and the vehicle positioning device 200 comprises:

[0114] The feature matching module 210 performs feature matching on the target image and the live image collected by the vehicle in real time.

[0115] The determination module 220 determines the position of the vehicle based on the coordinates of the positioning device if the matching is successful.

[0116] It should be noted that the specific details of each module unit in the above vehicle positioning device 200 have been described in detail in the above embodiment of the vehicle positioning method, and will not be repeated here.

[0117] In the embodiment of the application, the term "module" or "unit" refers to a computer program or a part of a computer program with a predetermined function, and works with other related parts to achieve a predetermined target, and can be implemented entirely or partially by using software, hardware (such as processing circuitry or memory) or a combination thereof. Similarly, one processor (or multiple processors or memory) can be used to implement one or more modules or units. In addition, each module or unit can be a part of an overall module or unit that includes the functions of the module or unit.

[0118] In some embodiments, the vehicle positioning device in the embodiment of the application can be implemented in a hardware manner, such as an electronic device or a component in an electronic device, for example, an integrated circuit or a chip; the vehicle positioning device can also be implemented in a software manner, such as an application installed in an electronic device.

[0119] The embodiment of the application further provides a controller, as shown in Figure 10 which shows the structure schematic diagram of the controller related to the embodiment of the application, specifically:

[0120] The controller can include a processor 301 having one or more processing cores, a memory 302 having one or more storage media, a power supply 303, and an input unit 304, etc. Those skilled in the art can understand that Figure 10 The controller structure shown in the figure is not a limitation on the controller, and can include more or fewer components than shown, or combine certain components, or different component arrangements. Among them:

[0121] The processor 301 is the control center of the controller, and connects various parts of the entire controller through various interfaces and lines, and performs various functions of the controller and processes data by running or executing computer programs and / or modules stored in the memory 302, and calling data stored in the memory 302. Optionally, the processor 301 can include one or more processing cores; preferably, the processor 301 can integrate an application processor and a modem processor, wherein the application processor mainly processes the operating system, user interface, and application program, etc., and the modem processor mainly processes wireless communication. It can be understood that the above-mentioned modem processor can also not be integrated into the processor 301.

[0122] The memory 302 can be used to store computer programs and modules, and the processor 301 executes various function applications and seat controls by running the computer programs and modules stored in the memory 302. The memory 302 can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, computer programs required by at least one function (such as an audible and light prompt function, a seat control function, etc.), etc.; the data storage area can store data created according to the use of the controller, etc. In addition, the memory 302 can include a high-speed random access memory, and can also include a non-volatile memory, such as at least one magnetic disk storage device, a flash memory device, or other volatile solid-state memory device. Accordingly, the memory 302 can also include a memory controller to provide the processor 301 with access to the memory 302.

[0123] The controller also includes a power supply 303 for powering various components, and preferably the power supply 303 can be logically connected to the processor 301 through a power management system, so as to realize the functions of managing charging, discharging, and power consumption management, etc. through the power management system. The power supply 303 can also include one or more direct current or alternating current power supplies, a recharging system, a power failure detection circuit, a power converter or inverter, a power state indicator, etc. Any component.

[0124] The controller can also include an input unit 304, which can be used to receive input digital or character information, and generate keyboard, mouse, joystick, optical or trackball signal inputs related to user settings and function controls.

[0125] Although not shown, the controller can further include a display unit, etc., which will not be described here. Specifically in the present embodiment, the processor 301 in the controller will load the executable file corresponding to the process of one or more computer programs into the memory 302 according to the following instructions, and run the computer program stored in the memory 302 by the processor 301, so as to realize various functions, such as: performing feature matching on the target image and the live image collected in real time by the vehicle; if the matching is successful, obtaining the positioning coordinates of the vehicle at the corresponding time, and determining the position of the vehicle.

[0126] The controller provided in the present application can.

[0127] The specific implementation manners and corresponding beneficial effects of the above operations can be referred to the detailed description of the vehicle positioning method above, which will not be described here.

[0128] Those skilled in the art can understand that all or part of the steps in the various methods of the above embodiments can be completed by a computer program, or by a computer program controlling related hardware, which can be stored in a storage medium and loaded and executed by a processor.

[0129] Therefore, the embodiments of the present application provide a storage medium, which stores a computer program capable of being loaded by a processor to execute the steps in any vehicle positioning method provided by the embodiments of the present application. For example, the computer program can execute the following steps:

[0130] performing feature matching on the target image and the live image collected in real time by the vehicle;

[0131] if the matching is successful, obtaining the positioning coordinates of the vehicle at the corresponding time, and determining the position of the vehicle.

[0132] The storage medium provided in the embodiments of the present application can.

[0133] The specific implementation manners and corresponding beneficial effects of the above operations can be referred to the previous embodiments, which will not be described here.

[0134] The storage medium can include a read only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, etc.

[0135] Since the computer program stored in the storage medium can execute the steps in any vehicle positioning method provided by the embodiments of the present application, the beneficial effects that can be achieved by any vehicle positioning method provided by the embodiments of the present application can be achieved, which will not be described here.

[0136] According to an aspect of the present application, there is provided a computer program product or computer program comprising computer instructions stored in a storage medium. A processor of a computer device reads the computer instructions from the storage medium, and the processor executes the computer instructions to cause the computer device to perform the above-mentioned method for positioning a vehicle.

[0137] The embodiments of the present application also provide a vehicle comprising the above-mentioned controller.

[0138] The specific structure of the vehicle is not limited in the present application. The specific implementation of the above-mentioned operations of the vehicle seat control system and the corresponding benefits are also applicable to the vehicle. For details, please refer to the detailed description of the seat control system above, which will not be repeated here.

[0139] The above describes in detail the vehicle positioning method, device, controller, medium, program product and vehicle provided by the embodiments of the present application. The principles and implementation manners of the present application are described by applying specific examples in this paper. The above-mentioned embodiments are only used to help understand the method of the present application and its core idea. Meanwhile, for those skilled in the art, according to the idea of the present application, the specific implementation manner and application range will be changed. In summary, the content of the specification should not be understood as a limitation of the present application.

Claims

1. A vehicle positioning method, characterized in that, The vehicle is equipped with a positioning device, and the vehicle positioning method includes: Feature matching is performed between the target image and the real-time on-site images acquired by the vehicle; If a match is successful, the vehicle's location is determined based on the coordinates of the positioning device.

2. The positioning method according to claim 1, characterized in that, The feature matching between the target image and the real-time captured scene image of the vehicle includes: A target window is determined based on the first feature points of the target image, and a search window is determined based on the second feature points of the scene image; Calculate the Gaussian curvature correlation coefficient between the target window and the search window, and determine the feature point pairs whose Gaussian curvature correlation coefficient is greater than a threshold curvature. The feature point pairs are the first feature point and the second feature point that satisfy the correspondence relationship. The scene image is corrected based on the coordinate information of the feature point pairs; The corrected on-site image and the target image are matched.

3. The positioning method according to claim 2, characterized in that, The step of matching the corrected on-site image and the target image includes: Calculate the grayscale correlation coefficient between the corrected on-site image and the target image; The comprehensive correlation coefficient and feature point matching rate between the corrected on-site image and the target image are determined based on the gray-level correlation coefficient and the Gaussian curvature correlation coefficient. When both the comprehensive correlation coefficient and the feature point matching rate are higher than a preset threshold, it is determined that the target image and the scene image are successfully matched.

4. The vehicle positioning method according to claim 2, characterized in that, The step of calculating the Gaussian curvature correlation coefficient between the target window and the search window, and determining feature point pairs whose Gaussian curvature correlation coefficient is greater than a threshold curvature, includes: Determine the first grayscale surface function of the target window and the second grayscale surface function of the search window respectively; The first Gaussian curvature set of all feature points in the target window is determined based on the first gray-scale surface function, and the second Gaussian curvature set of all feature points in the search window is determined based on the second gray-scale surface function. The Gaussian curvature correlation coefficient between the target window and the search window is determined based on the first Gaussian curvature set and the second Gaussian curvature set; Identify feature point pairs whose Gaussian curvature correlation coefficient is greater than the threshold curvature.

5. The vehicle positioning method according to claim 2, characterized in that, The correction of the scene image based on the coordinate information of the feature point pairs includes: A geometric distortion window is determined based on the target window, and the geometric distortion window includes at least three sets of the feature point pairs; Based on the feature point pairs, an error equation is constructed to calculate the geometric distortion parameters; The on-site image is corrected based on the geometric distortion parameters.

6. The vehicle positioning method according to claim 5, characterized in that, After correcting the scene image based on the geometric distortion parameters, the process includes: The corrected on-site image is then resampled.

7. The vehicle positioning method according to claim 5, characterized in that, The pixel size of the geometrically distorted window is larger than the pixel size of the target window.

8. The vehicle positioning method according to claim 2, characterized in that, The target window and the search window are the same size, and the width and height of the target window and the search window are both preset odd-numbered pixel dimensions.

9. The vehicle positioning method according to claim 1, characterized in that, Before performing feature matching on the target image and the real-time acquired on-site images of the vehicle, the following steps are included: Convert the target image into a grayscale image; Calculate the information content of the grayscale image; Images with an information content greater than a preset threshold are identified as valid target images.

10. The vehicle positioning method according to claim 1, characterized in that, The vehicle positioning method also includes: If a match is successful, data acquisition of the vehicle is initiated, and the acquired on-site images are uploaded to the server; If a match fails, data collection for the vehicle will not be initiated.

11. A vehicle positioning device, characterized in that, The vehicle is equipped with a positioning device, the positioning device comprising: The feature matching module performs feature matching between the target image and the real-time captured images of the vehicle. If a match is found, the module determines the vehicle's location based on the coordinates of the positioning device.

12. A controller, characterized in that, It includes one or more processors and a memory, the memory storing a computer program that, when executed by the processor, causes the processor to perform the steps of any of the vehicle positioning methods of claims 1-10.

13. A storage medium, characterized in that, Includes a computer program, which, when run on a controller, causes the controller to perform the steps of any of the vehicle positioning methods of claims 1-10.

14. A computer program product, characterized in that, It includes a computer program or instructions that, when executed by a processor, implement the steps of the vehicle positioning method according to any one of claims 1-10.

15. A vehicle, characterized in that, The vehicle includes the controller as described in claim 12.

Images