High-precision map abnormal data processing method and device, equipment and storage medium

Abstract

The application relates to a high-precision map abnormal data processing method, device and equipment and a storage medium. The method comprises the following steps: acquiring multiple data extracted by an image about a certain road element, and importing the multiple data into the same array; when the array meets a preset screening condition, removing abnormal data in the array according to a preset filtering rule until the array no longer meets the preset screening condition, and taking the current array as a target array, wherein the abnormal data is the data with the largest deviation from the current array. By locking the data with the largest deviation from the current array and removing the abnormal data from the current array, the data in the target array can maintain accuracy, and the production accuracy of the high-precision map can also be well maintained when the data in the target array is used to construct the high-precision map subsequently.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of high-definition map, and particularly relates to a high-definition map abnormal data processing method and device, equipment and a storage medium. BACKGROUND

[0002] A high-definition map is a reconstruction of a road environment. The high-definition map contains a large number of road elements such as lane lines, traffic lights and traffic signs. The high-definition map forms an accurate three-dimensional representation of a road network. Unmanned driving, vehicle positioning, path planning and vehicle control all rely on the high-definition map.

[0003] In related technologies, in order to improve the production accuracy of the high-definition map as much as possible, N photos captured by a vehicle-mounted camera are identified and analyzed, and a plurality of data for constructing a certain road element are extracted from the N photos. However, due to reasons such as mismatching, identification accuracy and analysis algorithm, some of the extracted data may deviate greatly from the remaining data. If such abnormal data is introduced into the construction of the road element in the subsequent process, the production accuracy of the high-definition map will be inevitably affected. SUMMARY

[0004] To solve or partially solve the problems in related technologies, the present application provides a high-definition map abnormal data processing method, device, equipment and storage medium, which can improve the production accuracy of the high-definition map.

[0005] The first aspect of the present application provides a high-definition map abnormal data processing method, comprising:

[0006] obtaining a plurality of data extracted from an image about a certain road element, and importing the plurality of data into a same array;

[0007] when the array meets a preset screening condition, removing abnormal data in the array according to a preset filtering rule until the array no longer meets the preset screening condition, and taking the current array as a target array, wherein the abnormal data is the data with the largest deviation from the current array.

[0008] Preferably, before the step of when the array meets a preset screening condition, the method further comprises:

[0009] counting an array length of the array, and calculating an array mean and an array variance corresponding to the array;

[0010] The preset screening condition comprises:

[0011] the array length corresponding to the array is greater than a preset length value; and

[0012] the array variance corresponding to the array is greater than a preset variance threshold.

[0013] Preferably, removing abnormal data from the array according to preset filtering rules includes:

[0014] Calculate the difference between each data point in the array and the mean of the array corresponding to the array, and obtain several differences;

[0015] The absolute values ​​of several of the aforementioned differences are sorted, and the abnormal data that needs to be removed is filtered out based on the sorting results.

[0016] Preferably, the step of sorting the absolute values ​​of the plurality of differences and filtering out the abnormal data to be removed based on the sorting results includes:

[0017] The absolute values ​​of several differences are sorted in ascending order, and the data corresponding to the absolute value of the largest difference is locked as abnormal data, and the abnormal data is removed from the array.

[0018] Preferably, the acquisition of multiple data points extracted from the image regarding a certain road element includes:

[0019] The image is obtained by extracting multiple data points related to the traffic sign.

[0020] A second aspect of this application provides a high-precision map anomaly data processing apparatus, comprising:

[0021] The acquisition module is used to acquire multiple data points extracted from an image regarding a certain road element, and import the multiple data points into the same array;

[0022] The filtering module is used to remove abnormal data in the array according to the preset filtering rules when the array meets the preset filtering conditions, until the array no longer meets the preset filtering conditions, and then take the current array as the target array, wherein the abnormal data is the data with the largest deviation from the current array.

[0023] Preferably, the filtering module is further used to calculate the array length of the array and to calculate the array mean and array variance corresponding to the array;

[0024] The preset filtering conditions include:

[0025] The array corresponding to the array has a length greater than a preset length value; and

[0026] The variance of the array corresponding to the array is greater than a preset variance threshold.

[0027] Preferably, the filtering module removes abnormal data from the array according to preset filtering rules, including:

[0028] Calculate the difference between each data point in the array and the mean of the array corresponding to the array, and obtain several differences;

[0029] The absolute values ​​of several of the aforementioned differences are sorted, and the abnormal data that needs to be removed is filtered out based on the sorting results.

[0030] A third aspect of this application provides an electronic device, comprising:

[0031] Processor; and

[0032] The memory stores executable code, which, when executed by the processor, causes the processor to perform the high-precision map anomaly data processing method as described above.

[0033] A fourth aspect of this application provides a computer-readable storage medium having executable code stored thereon, which, when executed by a processor of an electronic device, causes the processor to perform the high-precision map anomaly data processing method as described above.

[0034] The technical solution provided in this application may include the following beneficial effects:

[0035] The technical solution of this application acquires multiple data points from an image related to a specific road element and imports these data points into the same array. When the array meets preset filtering conditions, abnormal data in the array is removed according to preset filtering rules until the array no longer meets the preset filtering conditions. The current array is then used as the target array, where abnormal data is the data with the largest deviation from the current array. By locking the data with the largest deviation from the current array and removing abnormal data from the current array, the accuracy of the data in the target array can be maintained. This ensures that the accuracy of the high-precision map can be well maintained when using the data in the target array to construct the map.

[0036] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and do not limit this application. Attached Figure Description

[0037] The above and other objects, features and advantages of this application will become more apparent from the more detailed description of exemplary embodiments thereof in conjunction with the accompanying drawings, wherein the same reference numerals generally represent the same components in the exemplary embodiments thereof.

[0038] Figure 1 A flowchart illustrating a high-precision map anomaly data processing method according to an embodiment of this application is shown;

[0039] Figure 2A flowchart illustrating a high-precision map anomaly data processing method according to another embodiment of this application is shown;

[0040] Figure 3 A flowchart of a high-precision map anomaly data processing method according to an embodiment of this application is shown;

[0041] Figure 4 A schematic diagram of the structure of a high-precision map anomaly data processing device according to an embodiment of this application is shown;

[0042] Figure 5 A schematic diagram of the structure of a high-precision map anomaly data processing device according to another embodiment of this application is shown;

[0043] Figure 6 This is a schematic diagram of the structure of an electronic device shown in an embodiment of this application. Detailed Implementation

[0044] Embodiments of this application will now be described in more detail with reference to the accompanying drawings. While embodiments of this application are shown in the drawings, it should be understood that this application may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided to make this application more thorough and complete, and to fully convey the scope of this application to those skilled in the art.

[0045] The terminology used in this application is for the purpose of describing particular embodiments only and is not intended to be limiting of the application. The singular forms “a,” “the,” and “the” used in this application and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used herein refers to and includes any or all possible combinations of one or more of the associated listed items.

[0046] It should be understood that although the terms "first," "second," "third," etc., may be used in this application to describe various information, this information should not be limited to these terms. These terms are only used to distinguish information of the same type from one another. For example, without departing from the scope of this application, first information may also be referred to as second information, and similarly, second information may also be referred to as first information. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this application, "multiple" means two or more, unless otherwise explicitly specified.

[0047] Currently, in related technologies, due to reasons such as mismatch, recognition accuracy, and parsing algorithms, some data extracted from images may deviate significantly from the rest of the data. If such abnormal data is subsequently introduced into the construction of road elements, it will inevitably affect the accuracy of high-precision map production.

[0048] Therefore, in order to solve the above-mentioned technical problems, this application provides a method, apparatus, device and storage medium for processing abnormal data of high-precision maps, which can improve the production accuracy of high-precision maps.

[0049] The technical principles of this application are described in detail below with reference to the accompanying drawings.

[0050] Figure 1 A flowchart illustrating a high-precision map anomaly data processing method according to an embodiment of this application is shown.

[0051] Please see Figure 1 A method for processing abnormal data in high-precision maps includes the following steps:

[0052] Step S111: Obtain multiple data points extracted from the image regarding road elements, and import the multiple data points into the same array.

[0053] The image is a photograph taken by an onboard camera, containing numerous road elements such as traffic signs (road signs), lane lines, and stop lines. Multiple data points from the image for a specific road element in the same dimension are imported into a single array. An array is a collection of multiple data points for a single road element. Arrays have a length characteristic, which is the number of data points they contain.

[0054] Step S112: When the array meets the preset filtering conditions, remove abnormal data in the array according to the preset filtering rules until the array no longer meets the preset filtering conditions. Then, take the current array as the target array, where the abnormal data is the data with the greatest deviation from the current array.

[0055] Due to factors such as mismatches, recognition accuracy, and parsing algorithms, each data point in the array may not accurately reflect the road element. Specifically, each data point in the array has its own corresponding value, but each data point deviates somewhat from the actual data in the array. To ensure the accuracy of constructing road elements using the data in the array, abnormal data in the array that meets preset filtering criteria is removed according to preset filtering rules. That is, data with significant deviations from the array (these are called abnormal data) are filtered out from the array, ensuring that the filtered array data does not have a significant discrepancy with the actual data.

[0056] The technical solution of this embodiment locks the data with the largest deviation from the current array and removes abnormal data from the current array, which enables the data in the target array to maintain accuracy. This ensures that when using the data in the target array to build a high-precision map, the accuracy of the high-precision map can also be well maintained.

[0057] Figure 2 This illustration shows a flowchart of a high-precision map anomaly data processing method according to another embodiment of this application. Figure 2 relatively Figure 1 In particular, it details the preset filtering conditions and preset filtering rules.

[0058] Please see Figure 2 A method for processing abnormal data in high-precision maps includes the following steps:

[0059] Step S211: Obtain multiple data points extracted from the image regarding a certain road element, and import the multiple data points into the same array.

[0060] The image is a photograph taken by an onboard camera, containing numerous road elements such as traffic signs (road signs), lane markings, and stop lines. Multiple data points about a specific road element from the image are imported into the same array. An array is a collection of multiple data points for a single road element. Arrays have a length, which is the number of data points they contain.

[0061] Step S212: Calculate the array length and the array mean and variance of the corresponding array. When the array length is greater than the preset length value and the array variance is greater than the preset variance threshold, remove abnormal data in the array according to the preset filtering rules until the array no longer meets the condition that the array length is greater than the preset length value or the array variance is less than the preset variance threshold. Then, take the current array as the target array, where abnormal data is the data with the largest deviation from the current array.

[0062] It should be noted that the array length is the number of data points contained in the array. Since each data point in the array has its own value, the array mean and variance can be calculated.

[0063] The formula for calculating the mean of an array is as follows:

[0064]

[0065] Where mean represents the array mean, X i This represents the i-th data in the array, where N is the array length.

[0066] The formula for calculating the variance of an array is as follows:

[0067]

[0068] Where std represents the array variance, X iThis represents the i-th data in the array, mean represents the array average, and N is the array length.

[0069] The mean and variance of the corresponding array can be calculated using formulas (1) and (2).

[0070] To determine whether an array meets the preset filtering conditions, the array is considered to meet the preset filtering conditions when the array length N is greater than the preset length value len and the array variance std is greater than the preset variance threshold STD. In this case, the data in the array needs to be filtered.

[0071] The algorithm removes outlier data from the array according to preset filtering rules. These rules include: calculating the difference between each data point in the array and the array mean, obtaining several differences; sorting the absolute values ​​of these differences; and filtering out outlier data based on the sorting results. For example, the absolute values ​​of the differences can be sorted in ascending order, and the data corresponding to the largest absolute value of the difference can be identified as outlier data and removed from the array.

[0072] Through the above process, abnormal data in the array can be quickly identified and filtered (i.e., the identified abnormal data is removed from the array), and the array is updated. Based on the updated array, the array length, mean, and variance are recalculated and statistically analyzed. This continues until the array no longer meets the conditions of having an array length greater than a preset length value or an array variance less than a preset variance threshold. At this point, the current array is used as the target array. The resulting target array is the array with the abnormal data removed.

[0073] The technical solution of this embodiment locks the data with the largest deviation from the current array and removes abnormal data from the current array, which enables the data in the target array to maintain accuracy. This ensures that when using the data in the target array to build a high-precision map, the accuracy of the high-precision map can also be well maintained.

[0074] Figure 3 A flowchart of a high-precision map anomaly data processing method according to an embodiment of this application is shown. Figure 3 The application scenarios of handling road elements—traffic signs—are specifically discussed.

[0075] Please see Figure 3 A method for handling anomalies in high-precision maps, comprising the following steps:

[0076] Step S311: Transfer multiple data points X regarding traffic signs i Import into the array data and proceed to step S312.

[0077] Step S312: Calculate the array length N corresponding to the array data, and calculate the array mean and array variance std corresponding to the array data, then proceed to step S313.

[0078] Step S313: Determine whether the array length N is greater than the preset length value len and whether the array variance std is greater than the preset variance threshold SRTD. If yes, proceed to step S314; otherwise, proceed to step S317.

[0079] Step S314: Calculate the difference between each data point in the array data and the array mean, and obtain several difference values ​​D. i Proceed to step S315.

[0080] Step S315: Sort the absolute values ​​of several differences Di in ascending order, and select the data X corresponding to the absolute value of the largest difference. i Locked as abnormal data X error and the abnormal data X error Remove from array data and proceed to step S316.

[0081] Step S316: Update the array data, then return to step S313.

[0082] Step S317: Use the current array data as the target array data.

[0083] By taking the above steps, the data with the largest deviation from the current array can be quickly identified based on the array variance, and the abnormal data can be removed from the current array. This ensures that the data in the target array can maintain its accuracy, so that when the data in the target array is used to build a high-precision map, the accuracy of the high-precision map can also be well maintained.

[0084] Corresponding to the aforementioned functional method embodiments, this application also provides a high-precision map anomaly data processing device and corresponding embodiments.

[0085] Figure 4 A schematic diagram of the structure of a high-precision map anomaly data processing device according to an embodiment of this application is shown.

[0086] Please see Figure 4 A high-precision map anomaly data processing device 40 includes: an acquisition module 410 and a filtering module 420.

[0087] The acquisition module 410 is used to acquire multiple data points extracted from the image about a certain road element and import the multiple data points into the same array.

[0088] The image is a photograph taken by an onboard camera, containing numerous road elements such as traffic signs (road signs), lane lines, and stop lines. Multiple data points from the image for a specific road element in the same dimension are imported into a single array. An array is a collection of multiple data points for a single road element. Arrays have a length characteristic, which is the number of data points they contain.

[0089] The filtering module 420 is used to remove abnormal data in the array according to the preset filtering rules when the array meets the preset filtering conditions, until the array no longer meets the preset filtering conditions, and the current array is used as the target array, where the abnormal data is the data with the greatest deviation from the current array.

[0090] Due to factors such as mismatches, recognition accuracy, and parsing algorithms, each data point in the array may not accurately reflect the road element. Specifically, each data point in the array has its own corresponding value, but each data point deviates somewhat from the actual data in the array. To ensure the accuracy of constructing road elements using the data in the array, abnormal data in the array that meets preset filtering criteria is removed according to preset filtering rules. That is, data with significant deviations from the array (these are called abnormal data) are filtered out from the array, ensuring that the filtered array data does not have a significant discrepancy with the actual data.

[0091] In this embodiment, the acquisition module 410 acquires multiple data points extracted from an image regarding a specific road element and imports these data points into the same array. The filtering module 420, when the array meets preset filtering conditions, removes abnormal data from the array according to preset filtering rules until the array no longer meets the preset filtering conditions. The current array is then used as the target array, where abnormal data is the data with the largest deviation from the current array. By locking the data with the largest deviation from the current array and removing abnormal data from the current array, the accuracy of the data in the target array can be maintained. This ensures that when using the data in the target array to construct a high-precision map, the accuracy of the high-precision map can also be well maintained.

[0092] Figure 5 A schematic diagram of a high-precision map anomaly data processing device according to another embodiment of this application is shown.

[0093] Please see Figure 5 A high-precision map anomaly data processing device 50 includes: an acquisition module 510 and a filtering module 520.

[0094] For detailed information on the functions of module 510 and filter module 520, please refer to [link / reference]. Figure 4 The relevant descriptions in the document will not be repeated here.

[0095] The filtering module 520 is also used to count the array length and calculate the array mean and variance.

[0096] Accordingly, preset filtering conditions include: the array length corresponding to the array is greater than a preset length value; and the array variance corresponding to the array is greater than a preset variance threshold.

[0097] Accordingly, the filtering module 520 removes abnormal data from the array according to preset filtering rules, including: calculating the difference between each data in the array and the array mean, obtaining several differences; sorting the absolute values ​​of the several differences, and filtering out the abnormal data to be removed based on the sorting results.

[0098] Regarding the apparatus in the above embodiments, the specific manner in which each module and unit performs its operation has been described in detail in the method embodiments corresponding to the apparatus, and will not be elaborated further here.

[0099] Please see Figure 6 The electronic device 600 includes a processor 610 and a memory 620.

[0100] The processor 610 can be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor.

[0101] Memory 620 may include various types of storage units, such as system memory, read-only memory (ROM), and permanent storage devices. ROM may store static data or instructions required by the processor 610 or other modules of the computer. Permanent storage devices may be read-write storage devices. Permanent storage devices may be non-volatile storage devices that retain stored instructions and data even when the computer is powered off. In some embodiments, permanent storage devices use mass storage devices (e.g., magnetic or optical disks, flash memory) as permanent storage devices. In other embodiments, permanent storage devices may be removable storage devices (e.g., floppy disks, optical drives). System memory may be a read-write storage device or a volatile read-write storage device, such as dynamic random access memory. System memory may store some or all of the instructions and data required by the processor during operation. Furthermore, memory 620 may include any combination of computer-readable storage media, including various types of semiconductor memory chips (e.g., DRAM, SRAM, SDRAM, flash memory, programmable read-only memory), and disks and / or optical disks may also be used. The memory 620 stores executable code, which, when processed by the processor 610, can cause the processor 610 to execute part or all of the methods described above.

[0102] Furthermore, the method according to this application can also be implemented as a computer program or computer program product, which includes computer program code instructions for performing some or all of the steps in the method described above.

[0103] Alternatively, this application may be implemented as a computer-readable storage medium (or a non-transitory machine-readable storage medium or a machine-readable storage medium) storing executable code (or computer program or computer instruction code) thereon, which, when executed by a processor of an electronic device (or server, etc.), causes the processor to perform part or all of the steps of the methods described above according to this application.

[0104] The various embodiments of this application have been described above. These descriptions are exemplary and not exhaustive, nor are they limited to the disclosed embodiments. Many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the described embodiments. The terminology used herein is chosen to best explain the principles, practical application, or improvement of the technology in the market, or to enable others skilled in the art to understand the embodiments disclosed herein.

Claims

1. A method for processing abnormal data in high-precision maps, characterized in that, include: Obtain multiple data points of the same dimension from an image for a given road element, and import these multiple data points into the same array; When the array meets the preset filtering conditions, calculate the difference between each data in the array and the array mean corresponding to the array to obtain several differences; sort the absolute values ​​of the several differences, filter out the abnormal data that needs to be removed based on the sorting results, until the array no longer meets the preset filtering conditions, and then take the current array as the target array; The abnormal data is the data that deviates the most from the current array. The preset filtering conditions include the array length corresponding to the array being greater than a preset length value and the array variance corresponding to the array being greater than a preset variance threshold.

2. The high-precision map anomaly data processing method according to claim 1, characterized in that, Before the array meets the preset filtering conditions, the process further includes: The array length is calculated, and the array mean and variance are calculated.

3. The high-precision map anomaly data processing method according to claim 1, characterized in that, The process of sorting the absolute values ​​of several differences and filtering out outlier data to be removed based on the sorting results includes: The absolute values ​​of several differences are sorted in ascending order, and the data corresponding to the absolute value of the largest difference is locked as abnormal data, and the abnormal data is removed from the array.

4. The high-precision map anomaly data processing method according to any one of claims 1 to 3, characterized in that, The acquisition of multiple data points from an image regarding a specific road element includes: The image is obtained by extracting multiple data points related to the traffic sign.

5. A high-precision map anomaly data processing device, characterized in that, include: The acquisition module is used to acquire multiple data points of the same dimension of an image with respect to a certain road element, and import the multiple data points into the same array; The filtering module is used to calculate the difference between each data in the array and the mean of the array corresponding to the array when the array meets the preset filtering conditions, and obtain several differences; sort the absolute values ​​of the several differences, filter out the abnormal data that needs to be removed based on the sorting results, until the array no longer meets the preset filtering conditions, and then take the current array as the target array. The abnormal data is the data that deviates the most from the current array. The preset filtering conditions include the array length corresponding to the array being greater than a preset length value and the array variance corresponding to the array being greater than a preset variance threshold.

6. The high-precision map anomaly data processing device according to claim 5, characterized in that, The filtering module is also used to calculate the array length of the array and to calculate the array mean and array variance corresponding to the array.

7. An electronic device, characterized in that, include: processor; as well as A memory storing executable code, which, when executed by the processor, causes the processor to perform the high-precision map anomaly data processing method as described in any one of claims 1 to 4.

8. A computer-readable storage medium having executable code stored thereon, which, when executed by a processor of an electronic device, causes the processor to perform the high-precision map anomaly data processing method as described in any one of claims 1 to 4.

Images