Map data processing method, device and equipment

Abstract

This specification discloses a method, apparatus, and device for processing map data. The method includes: acquiring map data to be processed; determining a set of map elements from the map data to be processed, comprising a first map element and a second map element that are adjacent in location; and performing map element denoising processing on the set of map elements based on the confidence values ​​of the first map element and the second map element to obtain denoised map data. Thus, by performing map element denoising processing on the set of map elements based on the confidence values ​​of the first map element and the second map element, the accuracy of the map data processing results is improved.

Description

Technical Field

[0001] This application relates to the field of map data processing technology, and in particular to a method, apparatus and equipment for processing map data. Background Technology

[0002] In order to make full use of the map data collected by the map data acquisition equipment, reduce the data acquisition cost, and improve the accuracy of the mapping results, the map data needs to be processed before mapping is completed, so as to obtain as accurate and complete road information as possible.

[0003] Currently, after acquiring map data, the number of times each map element is reported is counted based on the acquired map data. The reliability of each map element is then determined based on the number of reports. Finally, for any given map element, it is determined whether its reliability meets preset reliability requirements, and denoising processing is performed on the map element based on the determination result. In this way, by denoising the map data based on the reliability of each map element, a map is generated based on the processing results.

[0004] However, in related technologies, simply judging whether a map element is a noise element based on its reliability may lead to non-noise elements with low reliability being identified as noise elements, resulting in low accuracy of map data processing results. Summary of the Invention

[0005] This specification provides a method, apparatus, and device for processing map data to improve the accuracy of map data processing results.

[0006] To solve the above-mentioned technical problems, the embodiments in this specification are implemented as follows:

[0007] Firstly, embodiments of this specification provide a method for processing map data, including:

[0008] Obtain the map data to be processed;

[0009] From the map data to be processed, a set of map features is determined; the set of map features includes a first map feature and a second map feature that are adjacent in position.

[0010] Based on the confidence values ​​of the first map element and the second map element, map element denoising processing is performed on the map element set to obtain denoised map data.

[0011] Secondly, embodiments of this specification provide a map data processing apparatus, including:

[0012] The acquisition module is used to acquire map data to be processed;

[0013] The determining module is used to determine a set of map features from the map data to be processed; the set of map features includes a first map feature and a second map feature that are adjacent in position;

[0014] The denoising module is used to perform map element denoising processing on the map element set based on the confidence values ​​of the first map element and the second map element to obtain denoised map data.

[0015] Thirdly, embodiments of this specification provide a map data processing device, including:

[0016] At least one processor; and,

[0017] A memory communicatively connected to the at least one processor; wherein,

[0018] The memory stores instructions executable by the at least one processor, which, when executed by the at least one processor, enable the at least one processor to:

[0019] Obtain the map data to be processed;

[0020] From the map data to be processed, a set of map features is determined; the set of map features includes a first map feature and a second map feature that are adjacent in position.

[0021] Based on the confidence values ​​of the first map element and the second map element, map element denoising processing is performed on the map element set to obtain denoised map data.

[0022] At least one embodiment provided in this specification can achieve the following beneficial effects:

[0023] After acquiring the map data to be processed, a set of map features containing adjacent first and second map features is determined from the map data. Based on the confidence values ​​of the first and second map features, denoising processing is performed on the set of map features to obtain denoised map data. Therefore, by performing denoising processing on the set of map features based on the confidence values ​​of the first and second map features, the accuracy of the map data processing results is improved. Attached Figure Description

[0024] To more clearly illustrate the technical solutions in the embodiments or prior art of this specification, the drawings used in the description of the embodiments or prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0025] Figure 1 A flowchart illustrating a map data processing method provided in an embodiment of this specification;

[0026] Figure 2 This is a schematic diagram of the data collection location of a ground arrow provided in an embodiment of this specification;

[0027] Figure 3 A schematic diagram of a gantry frame provided for an embodiment of this specification;

[0028] Figure 4 A schematic diagram of sub-element grouping of a traffic sign provided as an embodiment of this specification;

[0029] Figure 5 A schematic diagram of traffic sign noise reduction provided in the embodiments of this specification;

[0030] Figure 6 A schematic diagram of ground arrow noise reduction provided in the embodiments of this specification;

[0031] Figure 7 A schematic diagram of low-confidence ground arrow extraction provided for embodiments of this specification;

[0032] Figure 8 This is a schematic diagram of the structure of a map data processing device provided in the embodiments of this specification;

[0033] Figure 9 This is a schematic diagram of the structure of a map data processing device provided in an embodiment of this specification. Detailed Implementation

[0034] To make the objectives, technical solutions, and advantages of one or more embodiments of this specification clearer, the technical solutions of one or more embodiments of this specification will be clearly and completely described below in conjunction with specific embodiments and corresponding drawings. Obviously, the described embodiments are only a part of the embodiments of this specification, and not all of them. Based on the embodiments in this specification, all other embodiments obtained by those skilled in the art without creative effort are within the protection scope of one or more embodiments of this specification.

[0035] The technical solutions provided in the various embodiments of this specification are described in detail below with reference to the accompanying drawings.

[0036] Figure 1 This is a flowchart illustrating a map data processing method provided in an embodiment of this specification. From a programming perspective, the entity executing this process can be a computer, a server, or a corresponding task execution system. Figure 1 As shown, the process may include the following steps:

[0037] Step 101: Obtain the map data to be processed.

[0038] In the embodiments of this specification, the map data acquisition device (e.g., a map data acquisition vehicle, more specifically, a map data acquisition vehicle includes professional acquisition vehicles and ordinary social vehicles) is equipped with sensors for collecting map data. When the map data acquisition vehicle passes a certain map element, it can collect the information of the map element through the sensors installed on it. The map element can be a ground arrow, ground marking, traffic sign pole, traffic sign, traffic light, etc.

[0039] The acquisition of map features is often affected by factors such as the sensor itself, the acquisition time, the acquisition location of the map data acquisition device, and the weather when the map feature data is acquired. As a result, the map data acquired by the map data acquisition device contains a lot of noise data. Therefore, before constructing a map based on the map data acquired by the map data acquisition device, the map data needs to be denoised.

[0040] In practical applications, after acquiring raw map data collected by professional map data collection vehicles and ordinary social vehicles, the raw map data can be processed to obtain the map data to be processed. The map data to be processed includes type information, reporting frequency, and reliability value of each map element. For ground arrows, the map data to be processed also includes the longitudinal length, lateral width, lateral offset (perpendicular to the direction of road travel) relative to the standard reference line (which could be the centerline of the lane where the ground arrow is located), and longitudinal relative position (in the direction of road travel). The reporting frequency refers to the number of times a map element is collected; the reliability value of a map element can be calculated using the following formula:

[0041]

[0042] Here, x can be any map feature, such as any ground arrow on a lane (assumed to be ground arrow A). It is assumed that the map data acquisition device passes by ground arrow A a total of 3 times. However, the map data acquisition device only collects ground arrow A the first time it passes by. The other two times it passes by map feature A, ground arrow A is not collected due to some other reasons (such as ground arrow A being obscured). Therefore, the confidence level of ground arrow A can be calculated as 1 / 3.

[0043] It should be noted that the map data to be processed in this embodiment can be map data used to generate high-precision maps.

[0044] Step 102: Determine a set of map features from the map data to be processed; the set of map features includes a first map feature and a second map feature that are adjacent in position.

[0045] In the embodiments of this specification, "adjacent" means that the distance between the first map element and the second map element is less than or equal to a preset distance threshold, or that the first map element and the second map element at least partially overlap.

[0046] Step 102 may specifically include:

[0047] For any one of the first map features in the map data to be processed, determine the first type of the first map feature.

[0048] If the first type is a first preset type, then from the map data to be processed, map elements that are adjacent to the first map element and belong to the second type are obtained to obtain the second map element.

[0049] If the first type is the second preset type, then from the map data to be processed, map elements that are adjacent to the first map element and belong to the first type are obtained to obtain the second map element.

[0050] The first map element and the second map element are assigned to the map element set.

[0051] Specifically, the first preset type includes ground markings and traffic signs. The second preset type includes ground arrows, traffic signs, and traffic sign poles. In a specific example, if the first map element is a ground marking, then the ground arrow adjacent to the ground marking is obtained from the map data to be processed, and the ground marking and the ground arrow are assigned to the map element set; if the first map element is a first ground arrow, then the second ground arrow adjacent to the ground arrow is obtained from the map data to be processed, and the first ground arrow and the second ground arrow are assigned to the map element set; if the first map element is a first traffic sign pole, then the second traffic sign pole adjacent to the first traffic sign pole is obtained from the map data to be processed, and the first traffic sign pole and the second traffic sign pole are assigned to the map element set.

[0052] It should be noted that when the first map element is of the first preset type, the resulting second map element will not be classified into the same map element set as when the first map element is of the second preset type. An example is provided below.

[0053] If the first map element is a first traffic sign, then when the first traffic sign is of a first preset type, the traffic sign pole adjacent to the first traffic sign is obtained from the map data to be processed, and the first traffic sign and the traffic sign pole are assigned to the first map element set; when the first traffic sign is of a second preset type, the second traffic sign adjacent to the first traffic sign is obtained from the map data to be processed, and the first traffic sign and the second traffic sign are assigned to the second map element set.

[0054] Step 103: Based on the confidence values ​​of the first map element and the second map element, perform map element denoising processing on the map element set to obtain denoised map data.

[0055] The embodiments in this specification employ the above technical solution. After acquiring the map data to be processed, a set of map elements containing adjacent first and second map elements is determined from the map data. Based on the confidence values ​​of the first and second map elements, denoising processing is performed on the set of map elements to obtain denoised map data. Therefore, by performing denoising processing on the set of map elements based on the confidence values ​​of the first and second map elements, the accuracy of the map data processing results is improved.

[0056] based on Figure 1 In addition to the method described in the embodiments of this specification, some specific implementation schemes of the method are also provided, which will be described below.

[0057] Preferably, if the first map element is a ground arrow and the second map element is a ground marking, then step 102 may specifically include:

[0058] Step 1: Determine the ground markings that constitute the preset road segment, as well as the ground arrows located within the preset road segment.

[0059] Step 2: Divide the ground markings that constitute the preset road segment, as well as the ground arrows located within the preset road segment, into the map element set.

[0060] Step 103 may specifically include:

[0061] Step 1001: Determine the set of high-confidence ground markings from the set of map elements.

[0062] Step 1002: Based on the high-confidence ground markings in the high-confidence ground marking set and the preset topological relationship, the ground arrows are denoised to obtain a denoised ground arrow set; the high-confidence ground markings are ground markings with confidence values ​​greater than the preset confidence threshold.

[0063] Step 1003: Based on the ground arrows in the denoised ground arrow set and the preset topological relationship, perform denoising processing on the ground markings in the map element set, excluding the high-confidence ground marking set, to obtain a denoised ground marking set.

[0064] Step 1003: Based on the denoised ground markings in the denoised ground marking set and the high-confidence ground markings in the high-confidence ground marking set, perform denoising processing on the ground arrows in the denoised ground arrow set.

[0065] Specifically, preset topological relationships can include: ground markings and ground arrows do not overlap. In a specific example, such as... Figure 2 As shown, a is the center line of ground marking 21, and ground marking 21 is a high-confidence ground marking. After determining that ground arrow 22 overlaps with ground marking 21 based on the location information of ground marking 21 and ground arrow 22, it can be determined that ground marking 21 and ground arrow 22 do not conform to the preset topological relationship, and ground arrow 22 is removed from the map feature set.

[0066] In the embodiments of this specification, step 1001 may specifically include:

[0067] Identify a first target ground arrow that does not conform to a preset topological relationship with the high-confidence ground markings in the set of high-confidence ground markings.

[0068] Remove the first target ground arrow from the map feature set.

[0069] Step 1002 may specifically include:

[0070] Among the ground markings in the map feature set excluding the high-confidence ground marking set, target ground markings that do not conform to the preset topological relationship with the ground arrows in the denoised ground arrow set are identified.

[0071] Remove the target ground markings from the map feature set.

[0072] Step 1003 may specifically include:

[0073] Based on the denoised ground markings in the denoised ground marking set and the high-confidence ground markings in the high-confidence ground marking set, a lane boundary model is constructed.

[0074] Based on the lane boundary model, a second target ground arrow that does not conform to the preset ground arrow distribution rule is determined from the ground arrows in the set of denoised ground arrows.

[0075] Remove the second target ground arrow from the map feature set.

[0076] Among them, the preset ground arrow distribution rules are the distribution rules for lane boundary lines (ground markings that make up lanes) and ground arrows set by those skilled in the art according to actual needs. For example, ground arrows cannot exceed lane boundary lines, and ground arrows are evenly distributed within the lanes.

[0077] In this embodiment of the specification, if the first map element is a traffic sign and the second map element is a traffic sign pole, then step 102 may specifically include:

[0078] For any traffic sign in the map data to be processed;

[0079] Determine whether the traffic sign is a gantry traffic sign.

[0080] If the traffic sign is a gantry traffic sign, then other traffic signs located on the same gantry as this traffic sign are identified from the map data to be processed.

[0081] Determine the sets of first traffic sign poles corresponding to both sides of the road where the traffic sign is located.

[0082] For each set of first traffic sign poles, determine the third traffic sign pole that is closest to that traffic sign.

[0083] The traffic sign, other traffic signs, and the two identified third traffic sign poles are assigned to the map element set.

[0084] If the traffic sign is not a gantry traffic sign, then the fourth traffic sign pole closest to the traffic sign is determined from the map data to be processed.

[0085] The traffic sign and the fourth traffic sign pole are assigned to the map element set.

[0086] Step 103 may specifically include:

[0087] Determine whether the traffic sign is a high-confidence traffic sign and whether the traffic sign pole is a low-confidence traffic sign pole; the high-confidence traffic sign is a traffic sign with a confidence value greater than a preset confidence threshold; the low-confidence traffic sign pole is a traffic sign pole with a confidence value less than a preset confidence threshold.

[0088] If the traffic sign is a high-confidence traffic sign and the traffic sign pole is a low-confidence traffic sign pole, then it is prohibited to remove the low-confidence traffic sign pole from the map element set.

[0089] In a specific example, such as Figure 3 As shown, from the map data to be processed, determine as follows: Figure 3 The three traffic signs 32 shown are located on the same gantry. Then, from the map data to be processed, the first sets of traffic sign poles corresponding to both sides of the road where the traffic sign 32 is located are determined. From each first set of traffic sign poles, the traffic sign pole 31 closest to the traffic sign 32 is determined, and the traffic sign pole 31 and the traffic sign 32 are assigned to map element sets. Finally, if it is determined that the traffic sign pole 31 is a low-confidence traffic sign pole and the traffic sign 32 is a high-confidence traffic sign pole, then removing the traffic sign pole 31 from the map element set is prohibited.

[0090] Furthermore, the map data to be processed includes type data and location data of traffic sign elements; the types of traffic sign elements include main traffic signs and sub-traffic signs.

[0091] Prior to step 102, the method in the embodiments of this specification may further include:

[0092] Based on the location data of the traffic sign elements, the parent-child relationship reliability among the various traffic sign elements in the map data to be processed is determined.

[0093] Based on the parent-child relationship reliability, a set of traffic sign elements is determined from the map data to be processed; the set of traffic sign elements includes the main traffic sign and the child traffic sign that have a parent-child relationship as determined by the parent-child relationship reliability.

[0094] Identify duplicate sub-traffic signs among the sub-traffic signs at the set of traffic sign elements.

[0095] Based on the confidence values ​​of each of the repeated sub-traffic signs, the traffic sign element set is subjected to sub-traffic sign denoising processing.

[0096] In the embodiments of this specification, traffic signs include individual traffic signs and composite traffic signs. An individual traffic sign can refer to a traffic sign whose constituent sub-elements in the acquired map data to be processed number one. A composite traffic sign can refer to a traffic sign whose constituent sub-elements have multiple sub-elements with parent-child relationships. (Refer to...) Figure 3 Traffic sign 32 is a composite traffic sign, whose constituent elements include a lane sign, speed limit signs with values ​​of 60 and 80, and arrow symbols. If the traffic sign only contains arrow symbols, it can be a standalone traffic sign. In this embodiment, the parent traffic sign of the composite traffic sign and the standalone traffic sign are defined as the main traffic sign, and the child traffic signs are the various sub-traffic signs of the composite traffic sign.

[0097] During map feature acquisition, the data acquisition results for composite traffic signs include their constituent sub-elements. Continuing with the example above, the data acquisition results for traffic sign 32 may include four constituent sub-elements: a lane sign, speed limit signs with values ​​of 60 and 80, and an arrow sign. The lane sign can be defined as the parent traffic sign, and the other sub-elements as child traffic signs. The data acquisition results for individual traffic signs are for the sign itself.

[0098] In practical applications, after acquiring the data collection results for each traffic sign, the first step is to determine the parent-child relationship reliability between the sub-elements of each traffic sign. Then, each parent-child relationship reliability is compared with a preset parent-child relationship reliability threshold. If the reliability of a parent-child relationship exceeds the preset threshold, a parent-child relationship is defined as existing between the two sub-elements corresponding to that reliability. Based on this, the sub-elements are grouped according to their parent-child relationship reliability, ensuring that sub-elements with a parent-child relationship belong to the same group. To illustrate this method more clearly, an example is provided below.

[0099] Figure 4 This is a schematic diagram illustrating the grouping of sub-elements of a traffic sign, as provided in an embodiment of this specification. Figure 4 As shown, letters A, G, and F represent sub-features. After obtaining the data information of each sub-feature, the reliability of the parent-child relationship between each sub-feature is determined, as follows: Figure 4As shown, the numbers next to the arrows represent the reliability of the parent-child relationship between the corresponding two child elements. Specifically, for example, the reliability of the parent-child relationship between child element A and child element B is 0.1, and the reliability of the parent-child relationship between child element B and child element F is 0.8. Assuming the set threshold for parent-child relationship reliability is 0.1, meaning that when the reliability of the parent-child relationship between two child elements is less than or equal to 0.1, it is determined that the two child elements do not have a parent-child relationship. Based on this logic, [the following is applied to...] Figure 4 The child elements are grouped according to their parent-child relationships, and the grouping results are as follows: Figure 4 As shown, since the parent-child relationship confidence between sub-element A and sub-element B is less than or equal to 0.1, it is determined that sub-element A and sub-element B do not have a parent-child relationship. Therefore, sub-element A is a separate group, and it can be determined that sub-element A is a single traffic sign. Similarly, sub-element B and sub-element F have a parent-child relationship and are grouped together. It can be determined that sub-element B is the parent traffic sign and sub-element F is the child traffic sign. Sub-element B and sub-element F constitute a composite traffic sign. Sub-element E, sub-element D, and sub-element H are grouped together; sub-element C and sub-element G are grouped together.

[0100] After grouping the sub-elements, the grouping for the composite traffic sign can be determined as the traffic sign element set of this embodiment. For example, sub-elements E, D, and H constitute a traffic sign element set. Then, duplicate sub-traffic signs are identified among the sub-traffic signs in the traffic sign element set. Duplicate sub-traffic signs refer to sub-traffic signs with the same element type, such as sub-elements D and H both being straight arrow signs. Finally, the second map element set is denoised based on the confidence values ​​of each duplicate sub-traffic sign. Continuing with the previous example, if sub-elements D and H are duplicate sub-traffic signs, the confidence values ​​of sub-elements D and H are determined, and the sub-elements with lower confidence values ​​are identified as noise elements, and the identified noise elements are deleted from the second map element set.

[0101] In another specific example, such as Figure 5 As shown, the arrows are used to indicate parent-child relationships. When grouping the child elements according to the parent-child relationship, the child element 6 located on the road boundary line can be assigned to the group where the child element 1 is located according to the parent-child relationship. Furthermore, there are two child elements 5 in this group. Therefore, the group can be denoised based on the confidence values ​​of the two child elements 5.

[0102] Furthermore, both the first map element and the second map element can be main traffic signs.

[0103] In the embodiments of this specification, the main traffic sign can be either the main traffic sign in a gantry traffic sign system or the main traffic sign in the traffic signs on both sides of the road. The distribution patterns of the gantry traffic signs and the traffic signs on both sides of the road differ. The gantry traffic signs are evenly distributed; for example, each gantry traffic sign is located at positions A, B, C, D, and E on a certain road, and the distance between A, B, C, D, and E is 5 meters. In contrast, the traffic signs on both sides of the road are irregularly distributed.

[0104] Based on this, in the embodiments of this specification, step 102 may specifically include:

[0105] Determine the installation type of the main traffic sign in the map data to be processed.

[0106] If the installation type indicates that the main traffic sign is a gantry traffic sign, then based on the location information of the road markings and the location information of the main traffic sign, the map element set is determined from the map data to be processed; the map element set includes each main traffic sign located in the same lane of the same road segment.

[0107] Specifically, based on the location information of each main traffic sign in the map data to be processed, the installation type of each main traffic sign is determined. If the installation type indicates that the main traffic sign is a gantry traffic sign, then the main traffic signs above the preset lanes of the preset road segment constitute a set of map elements. (Refer to...) Figure 5 Lane 1 has two main traffic signs (i.e., sub-elements 2 and 7) in the illustrated road segment. Therefore, sub-elements 2 and 7 constitute a set of map elements.

[0108] If the installation type indicates that the main traffic sign is a traffic sign on both sides of the road, then a specific location on one side of the road can be used as the center point to determine all the main signs within a preset distance range from that specific location to form a map element set.

[0109] Step 103 may specifically include:

[0110] Identify the repeating primary traffic signs at the set of map elements.

[0111] Based on the reliability of each repeated main traffic sign, the map element set is subjected to main traffic sign denoising processing.

[0112] Following the example above, after grouping the main traffic signs, for any given set of map features, identify the duplicate main traffic signs among the main traffic signs at that set of map features, such as... Figure 5As shown, the map element set of lane 3 contains two sub-elements 4. These two sub-elements 4 can be considered as duplicate main traffic signs. The confidence values ​​of these two duplicate main traffic signs are determined, and the main traffic signs with lower confidence values ​​are removed from the third map element set.

[0113] Furthermore, after determining the set of traffic sign elements from the map data to be processed based on the parent-child relationship reliability, the method in this embodiment may further include:

[0114] From the map data to be processed, determine the target main traffic sign that is located in the same lane of the same preset road segment as the traffic sign element in the traffic sign element set;

[0115] The traffic sign element set and the target main traffic sign are assigned to the target traffic sign element set;

[0116] Identify the target sub-traffic signs in the target traffic sign element set that are duplicates of the main traffic sign;

[0117] Based on the confidence values ​​of the main traffic sign and the target sub-traffic sign, noise reduction processing is performed on the target traffic sign element set between the main traffic sign and the sub-traffic sign.

[0118] In the embodiments of this specification, Figure 5 For example, in the target traffic sign element set of lane 1, there are two sub-elements 7, one of which is the main traffic sign and the other is the sub-traffic sign. Therefore, by comparing the reliability values ​​of the two sub-elements 7, the sub-element 7 with the lower reliability value is removed from the target traffic sign element set.

[0119] Furthermore, the map data to be processed includes traffic signs; after step 101, the method of this embodiment may further include:

[0120] For any traffic sign in the map data to be processed, determine whether the traffic sign is a high-confidence traffic sign.

[0121] If the traffic sign is a high-reliability traffic sign, then determine whether the map data to be processed contains traffic sign poles that are connected to the traffic sign.

[0122] If the map data to be processed does not contain any traffic sign poles that are connected to the traffic sign, then historical map data is obtained.

[0123] From the historical map data, identify historical traffic sign poles that are connected to the traffic signs.

[0124] Based on the historical traffic sign poles, the map data to be processed is supplemented with traffic sign poles.

[0125] In this embodiment of the specification, determining whether the map data to be processed contains a traffic sign pole that is connected to the traffic sign may include:

[0126] Determine whether there is a traffic sign pole within the area where the traffic sign is located that is less than a preset traffic sign distance threshold; if not, determine that the map data to be processed does not contain any traffic sign poles that are connected to the traffic sign.

[0127] In the embodiments of this specification, historical map data can be map data corresponding to a high-precision map that has already been mapped. If the historical map data includes the traffic sign, and includes historical traffic sign poles that are connected to the traffic sign, then based on the historical traffic sign poles, traffic sign poles that are connected to the traffic sign are generated in the map data to be processed.

[0128] Furthermore, the first map element is a first ground arrow, and the second map element is a second ground arrow.

[0129] The map feature set includes surface arrows located in the same lane of the same preset road segment.

[0130] Step 102 may specifically include:

[0131] A lane model is constructed based on the ground markings in the map data to be processed.

[0132] Based on the lane model, the surface arrows in the same lane of the same preset road segment are divided into map element sets.

[0133] Step 103 may specifically include:

[0134] If the first ground arrow overlaps with the second ground arrow, then all ground arrows in the first and second ground arrows that overlap, except for the ground arrow with the highest confidence value, will be removed from the map feature set.

[0135] Furthermore, after step 102, the method in the embodiments of this specification may further include:

[0136] If the first ground arrow and the second ground arrow have the same vertical relative position, then the ground arrows that are not located at the preset road position in the map element set are removed from the map element set.

[0137] The following is based on Figure 6 The above solution will be explained in detail with an example. Figure 6 As shown, dashed line a represents the centerline of lane 1, dashed lines b and c are the dividing lines of lane 2 (i.e., dashed lines b and c divide lane 2 into three equal sub-lanes), and dashed lines d and e are the dividing lines of lane 3. After obtaining the data of the surface arrows on the road, the road can be divided into multiple segments according to the preset road segmentation rules. Figure 6 Two road segments (segment 1 and segment 2) are shown. It should be noted that the lengths of segment 1 and segment 2 can be arbitrary, and their lengths can be the same or different. Then, all ground arrows within the same lane of the same road segment are identified as a set of map features. Figure 6 For example, Figure 6 The dashed boxes are used to group the ground arrows in each location, with ground arrow groups 11-16 representing a map feature set.

[0138] After determining each fourth map element set, for any map element set, denoising processing is performed on the map element set according to the preset ground arrow logic relationship. For example, for ground arrow group 11, this ground arrow group contains two straight arrows, and both straight arrows are located on the center line a. However, the longitudinal position (road direction) of these two straight arrows is different. Therefore, denoising can be performed on them according to the confidence of the two straight arrows, and the straight arrows with lower confidence values ​​are removed from ground arrow group 11.

[0139] For ground arrow group 14, which contains two straight arrows that completely overlap, it can be determined that there is a noisy straight arrow among these two arrows. Similarly, noise can be removed based on the confidence values ​​of these two arrows, eliminating the arrows with lower confidence values ​​from ground arrow group 14. It should be noted that alternatively, the map feature set could contain two partially overlapping ground arrows of the same type, and noise can be removed based on the confidence values ​​of these two partially overlapping ground arrows of the same type.

[0140] In practical applications, for any given lane, if there is only one ground arrow pointing in the direction of its cross-section, that ground arrow should be located on the centerline of the lane. Figure 6 In lane 1, the straight arrow should be located on the center line of lane 1. If there are two parallel ground arrows in the cross-sectional direction of this lane, the two parallel ground arrows should be evenly distributed within the lane, such as... Figure 6In lanes 2 and 3, the straight-ahead arrow should be located on dashed line b, and the right-turn arrow should be located on dashed line c. Based on this, for ground arrow group 13, which contains one straight-ahead arrow and one right-turn arrow, since the right-turn arrow in ground arrow group 13 is not located on dashed line e, it can be determined that the right-turn arrow is a noise arrow, and therefore, the noise arrow is removed from ground arrow group 13. Based on the same principle, it can be seen that ground arrow groups 15 and 16 may not contain any noise ground arrows.

[0141] Furthermore, prior to step 102, the method in the embodiments of this specification may also include:

[0142] Identify each high-confidence ground arrow in the map data to be processed; the high-confidence ground arrow is a ground arrow with a confidence value greater than a preset confidence threshold.

[0143] Based on the location information of the high-confidence ground arrows and the location information of road markings, each fifth map element set is determined from the map data to be processed; the vertical relative positions of all ground arrows in any fifth map element set are the same, and each set contains at least one high-confidence ground arrow.

[0144] From the map data to be processed, a set of low-confidence ground arrows is determined; the low-confidence ground arrows are ground arrows with a confidence value less than or equal to a preset confidence threshold.

[0145] For any target low-confidence ground arrow in the set of low-confidence ground arrows, determine whether the target low-confidence ground arrow meets the preset extraction conditions.

[0146] If the judgment result indicates that the target low-confidence ground arrow meets the preset extraction conditions, then it is prohibited to remove the target low-confidence ground arrow from the map data to be processed.

[0147] In the embodiments of this specification, the conditions for extracting low-confidence ground arrows are as follows:

[0148] Condition 1: The target low-confidence ground arrow is located in any of the fifth map feature sets, or at a specified distance from a specified fifth map feature set; the specified fifth map feature set is the fifth map feature set closest to the target low-confidence ground arrow; the specified distance is the average distance between adjacent fifth map feature sets.

[0149] Condition 2: The target low-confidence ground arrow does not exceed the left and right boundary lines of the corresponding lane;

[0150] Condition 3: The total number of times the target low-confidence ground arrows are collected exceeds the preset minimum reporting threshold;

[0151] If the judgment result indicates that the target low-confidence ground arrow meets the above three conditions (condition 1, condition 2 and condition 3), then it is prohibited to remove the target low-confidence ground arrow from the map data to be processed.

[0152] In the embodiments of this specification, the following formula can be used to determine whether the target low-confidence ground arrows meet the above-mentioned low-confidence ground arrow extraction conditions:

[0153] min(abs(Index-groupIndexi),(i=1,…,m))=AveIndexDiff±indexBuffer(1)

[0154]

[0155]

[0156] ReportCount ≥ MinReportcount (4)

[0157] Where, Index is the longitudinal relative position of the target low-confidence ground arrow; groupIndex is the average longitudinal relative position of each ground arrow in the i-th fifth map feature set, m is equal to the number of the fifth map feature sets; AveIndexDiff is the average distance between each adjacent fifth map feature set; indexBuffer takes a first preset value to represent the buffer value corresponding to the longitudinal offset of the target low-confidence arrow; LateralOffset is the lateral offset of the target low-confidence arrow relative to the standard reference line; Width is the lateral width of the target low-confidence ground arrow; LaneRightOffset j LaneLeftOffset j These represent the lateral offsets of the left and right boundary lines of the j-th lane relative to the standard reference line; offsetBuffer takes a second preset value to represent the buffer value corresponding to the longitudinal offset of the target low-confidence arrow; ReportCount is the number of times the target low-confidence ground arrow is reported; MinReportcount is a preset minimum reporting threshold.

[0158] In this embodiment of the specification, formula (1) is used to determine whether any of the target low-confidence ground arrows is located in any of the fifth map element sets, or whether it is located at a specified distance from a specified fifth map element set; the specified fifth map element set is the fifth map element set closest to the target low-confidence ground arrow; the specified distance, namely AveIndexDiff, is the average distance between each adjacent fifth map element set; formulas (2) and (3) are used to determine whether the target low-confidence ground arrow exceeds the left and right boundary lines of the corresponding lane; formula (4) is used to determine whether the total number of collections of the target low-confidence ground arrow exceeds the preset minimum reporting threshold. When the determination result indicates that the target low-confidence ground arrow simultaneously meets the above low-confidence ground arrow extraction conditions, the low-confidence ground arrow can be labeled with "low-confidence extraction" so that the system prohibits the removal of the low-confidence ground arrow from the map data based on the label.

[0159] The following is based on Figure 7 The above solution will be explained in detail with an example. Figure 7 As shown, a1 and a2 represent the centerlines of lane 1 and lane 2, respectively; assuming Z1-Z6 represent the longitudinal relative positions of ground arrows 1-6, groupIndex i The possible values ​​are as follows:

[0160] When i = 1 (corresponding to ground arrow group 71), groupIndex i =1 / 2*(Z1+Z2); when i=2 (corresponding to ground arrow group 72), groupIndex i =1 / 2*(Z3+Z4); when i=3 (corresponding to ground arrow group 73), groupIndex i =1 / 2*(Z5+Z6); m=3.

[0161] The values ​​of AveIndexDiff are as follows:

[0162]

[0163] LateralOffset is the lateral offset of the target low-confidence arrow position relative to the standard reference line; the standard reference line can be the center line of the lane, and in this example, the standard reference line is the center line a1.

[0164] When j = 1 (corresponding to lane 1), LaneRightOffset j LaneLeftOffset jThese represent the lateral offsets of the left and right boundary lines of lane 1 relative to the center line a1, respectively; when j = 2 (corresponding to lane 2), LaneRightOffset... j LaneLeftOffset j These are the lateral offsets of the left and right boundary lines of lane 2 relative to the center line a2, respectively; n = 2.

[0165] Following the example above, assume that ground arrows 1, 3, 4, 5, and 6 are identified as high-confidence ground arrows, and ground arrows 2, 7, and 8 as low-confidence ground arrows. Based on the location information of the high-confidence ground arrows and the location information of road markings, the fifth map element sets are determined from the map data to be processed, resulting in three fifth map element sets: ground arrow group 71, ground arrow group 72, and ground arrow group 73. In the specific extraction process of low-confidence ground arrows, it can be determined that ground arrow 2 is within the ground arrow group containing high-confidence ground arrows (i.e., ground arrow group 71), does not exceed the left and right boundary lines of lane 2, and its total collection count exceeds the preset minimum reporting threshold. Therefore, ground arrow 2 can be labeled "low-confidence extraction" to prevent it from being removed from the map data to be processed. Similarly, it can be determined that ground arrow 7 and ground arrow 8 are both located at AveIndexDiff at a distance of 73 from ground arrow group 73, and both are beyond the left and right boundary lines of the corresponding lanes, and the total number of collections exceeds the preset minimum reporting threshold. Therefore, ground arrow 7 and ground arrow 8 can be labeled "low confidence extraction" to prevent them from being removed from the map data to be processed.

[0166] Furthermore, the first map element is a first traffic sign pole, and the second map element is a second traffic sign pole. The first traffic sign pole and the second traffic sign pole are located on the same side of the preset road.

[0167] Step 102 may specifically include:

[0168] For any traffic sign pole in the map data to be processed, determine the second set of traffic sign poles corresponding to the road side where the traffic sign pole is located from the map data to be processed.

[0169] From the second set of traffic sign poles, target traffic sign poles within a preset distance range from the traffic sign poles are determined.

[0170] The traffic sign poles and the target traffic sign poles are assigned to a map feature set.

[0171] Step 103 may specifically include:

[0172] Determine whether the first traffic sign pole and the second traffic sign pole are duplicate traffic sign poles.

[0173] If the first traffic sign pole and the second traffic sign pole are duplicates, then all traffic sign poles except the one with the lowest confidence level will be removed from the map feature set.

[0174] In the embodiments of this specification, after acquiring the map data to be processed, each traffic sign pole can be identified from the map data. Then, based on the location information of the road centerline and the location information of each traffic sign pole, the traffic sign poles can be divided into two groups. Each group of traffic sign poles belongs to the same side of the road. Alternatively, they can be divided into two groups based on the two boundary lines of the road and the location information of each traffic sign pole; this is not limited here. Next, for any group of traffic sign poles, based on the location of a specific pole, traffic sign poles whose distance from that specific pole is less than a preset traffic sign pole distance threshold can be grouped together to obtain a map element set. Duplicate traffic sign poles are identified among the traffic sign poles in the map element set, where duplicate traffic sign poles refer to poles of the same type and with the same location coordinates. Finally, the confidence value of each duplicate traffic sign pole is determined, and duplicate traffic sign poles with lower confidence values ​​are removed from the map element set.

[0175] Based on the same idea, embodiments of this specification also provide apparatus corresponding to the above methods. Figure 8 This is a schematic diagram of a map data processing device provided in an embodiment of this specification. Figure 8 As shown, the device may include:

[0176] The acquisition module 801 is used to acquire map data to be processed.

[0177] The determination module 802 is used to determine a set of map elements from the map data to be processed; the set of map elements includes a first map element and a second map element that are adjacent in position.

[0178] The denoising module 803 is used to perform map element denoising processing on the map element set based on the confidence values ​​of the first map element and the second map element to obtain denoised map data.

[0179] Furthermore, the first map element is a ground arrow, and the second map element is a ground marking.

[0180] Module 802 can be specifically used for:

[0181] For any one of the first map features in the map data to be processed, determine the first type of the first map feature.

[0182] If the first type is a first preset type, then from the map data to be processed, map elements that are adjacent to the first map element and belong to the second type are obtained to obtain the second map element.

[0183] If the first type is the second preset type, then from the map data to be processed, map elements that are adjacent to the first map element and belong to the first type are obtained to obtain the second map element.

[0184] The first map element and the second map element are assigned to the map element set.

[0185] The 803 noise reduction module can be used specifically for:

[0186] From the set of map elements, a set of high-confidence ground markings is determined.

[0187] Based on the high-confidence ground markings in the high-confidence ground marking set and the preset topological relationship, the ground arrows are denoised to obtain a denoised ground arrow set; the high-confidence ground markings are ground markings with confidence values ​​greater than the preset confidence threshold.

[0188] Based on the ground arrows in the denoised ground arrow set and the preset topological relationship, the ground markings in the map element set, excluding the high-confidence ground marking set, are denoised to obtain a denoised ground marking set.

[0189] Based on the denoised ground markings in the denoised ground marking set and the high-confidence ground markings in the high-confidence ground marking set, the ground arrows in the denoised ground arrow set are denoised.

[0190] Furthermore, the first map element is a traffic sign, and the second map element is a traffic sign pole.

[0191] The 803 noise reduction module can be used specifically for:

[0192] Determine whether the traffic sign is a high-confidence traffic sign and whether the traffic sign pole is a low-confidence traffic sign pole; the high-confidence traffic sign is a traffic sign with a confidence value greater than a preset confidence threshold; the low-confidence traffic sign pole is a traffic sign pole with a confidence value less than a preset confidence threshold.

[0193] If the traffic sign is a high-confidence traffic sign and the traffic sign pole is a low-confidence traffic sign pole, then it is prohibited to remove the low-confidence traffic sign pole from the map element set.

[0194] Furthermore, the map data to be processed includes type data and location data of traffic sign elements; the types of traffic sign elements include main traffic signs and sub-traffic signs.

[0195] The apparatus in the embodiments of this specification may further include a sub-traffic sign noise reduction processing module, used for:

[0196] Based on the location data of the traffic sign elements, the parent-child relationship reliability among the various traffic sign elements in the map data to be processed is determined.

[0197] Based on the parent-child relationship reliability, a set of traffic sign elements is determined from the map data to be processed; the set of traffic sign elements includes the main traffic sign and the child traffic sign that have a parent-child relationship as determined by the parent-child relationship reliability.

[0198] Identify duplicate sub-traffic signs among the sub-traffic signs at the set of traffic sign elements.

[0199] Based on the reliability of each repeated sub-traffic sign, the set of traffic sign elements is subjected to sub-traffic sign denoising processing.

[0200] Furthermore, the first map element and the second map element are different main traffic signs located in the same lane of the same preset road segment.

[0201] The 803 noise reduction module can be used specifically for:

[0202] Identify the repeating primary traffic signs at the location of the map feature set;

[0203] Based on the reliability of each repeated main traffic sign, the map element set is subjected to main traffic sign denoising processing.

[0204] Furthermore, the apparatus in the embodiments of this specification may also include a parent-child traffic sign noise reduction module, which is used for:

[0205] From the map data to be processed, determine the target main traffic sign that is located in the same lane of the same preset road segment as the traffic sign element in the traffic sign element set.

[0206] The traffic sign element set and the target main traffic sign are assigned to the target traffic sign element set.

[0207] Identify the target sub-traffic signs in the target traffic sign element set that are duplicates of the main traffic sign.

[0208] Based on the reliability of the main traffic sign and the target sub-traffic sign, noise reduction processing is performed on the target traffic sign element set between the main traffic sign and the sub-traffic sign.

[0209] Furthermore, the map data to be processed includes traffic signs. The apparatus in this embodiment may further include a traffic sign pole supplementation processing module, which is used for:

[0210] For any traffic sign in the map data to be processed, determine whether the traffic sign is a high-confidence traffic sign.

[0211] If the traffic sign is a high-reliability traffic sign, then determine whether the map data to be processed contains traffic sign poles that are connected to the traffic sign.

[0212] If the map data to be processed does not contain any traffic sign poles that are connected to the traffic sign, then historical map data is obtained.

[0213] From the historical map data, identify historical traffic sign poles that are connected to the traffic signs.

[0214] Based on the historical traffic sign poles, the map data to be processed is supplemented with traffic sign poles.

[0215] Furthermore, the first map element is a first ground arrow, and the second map element is a second ground arrow. The map element set includes ground arrows located in the same lane of the same preset road segment.

[0216] The 803 noise reduction module can be used specifically for:

[0217] If the first ground arrow overlaps with the second ground arrow, then all ground arrows in the first and second ground arrows that overlap, except for the ground arrow with the highest confidence value, will be removed from the map feature set.

[0218] Furthermore, the apparatus in the embodiments of this specification may further include a parallel ground arrow noise reduction module, the parallel ground arrow noise reduction module being used for:

[0219] If the first ground arrow and the second ground arrow have the same vertical relative position, then the ground arrows that are not located at the preset road position in the map element set are removed from the map element set.

[0220] Furthermore, the apparatus in the embodiments of this specification may also include a low-confidence ground arrow extraction module, used for:

[0221] Identify each high-confidence ground arrow in the map data to be processed; the high-confidence ground arrow is a ground arrow with a confidence value greater than a preset confidence threshold.

[0222] Based on the location information of the high-confidence ground arrows and the location information of road markings, each fifth map element set is determined from the map data to be processed; the vertical relative positions of all ground arrows in any fifth map element set are the same, and each set contains at least one high-confidence ground arrow.

[0223] From the map data to be processed, a set of low-confidence ground arrows is determined; the low-confidence ground arrows are ground arrows with a confidence value less than or equal to a preset confidence threshold.

[0224] For any target low-confidence ground arrow in the set of low-confidence ground arrows, determine whether the target low-confidence ground arrow meets the preset extraction conditions.

[0225] If the judgment result indicates that the target low-confidence ground arrow meets the preset extraction conditions, then it is prohibited to remove the target low-confidence ground arrow from the map data to be processed.

[0226] Furthermore, the first map element is a first traffic sign pole, and the second map element is a second traffic sign pole. The first traffic sign pole and the second traffic sign pole are located on the same side of the preset road.

[0227] The 803 noise reduction module can be used specifically for:

[0228] Determine whether the first traffic sign pole and the second traffic sign pole are duplicate traffic sign poles.

[0229] If the first traffic sign pole and the second traffic sign pole are duplicates, then all traffic sign poles except the one with the lowest confidence level will be removed from the map feature set.

[0230] Based on the same idea, this specification also provides devices corresponding to the above methods in its embodiments.

[0231] Figure 9 This is a schematic diagram of the structure of a map data processing device provided in an embodiment of this specification.

[0232] like Figure 9 As shown, device 900 may include:

[0233] At least one processor 910; and a memory 930 communicatively connected to the at least one processor; wherein the memory 930 stores instructions 920 executable by the at least one processor 910, the instructions being executed by the at least one processor 910 to enable the at least one processor 910 to:

[0234] Obtain the map data to be processed.

[0235] From the map data to be processed, a set of map elements is determined; the set of map elements includes a first map element and a second map element that are adjacent in position.

[0236] Based on the confidence values ​​of the first map element and the second map element, map element denoising processing is performed on the map element set to obtain denoised map data.

[0237] The various embodiments in this specification are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on its differences from other embodiments. In particular, for... Figure 9 As the device shown is basically similar to the method embodiment, the description is relatively simple, and relevant parts can be found in the description of the method embodiment.

[0238] In the 1990s, improvements to a technology could be clearly distinguished as either hardware improvements (e.g., improvements to the circuit structure of diodes, transistors, switches, etc.) or software improvements (improvements to the methodology). However, with technological advancements, many methodological improvements today can be considered direct improvements to the hardware circuit structure. Designers almost always obtain the corresponding hardware circuit structure by programming the improved methodology into the hardware circuit. Therefore, it cannot be said that a methodological improvement cannot be implemented using hardware physical modules. For example, a Programmable Logic Device (PLD) (such as a Field Programmable Gate Array (FPGA)) is such an integrated circuit whose logic function is determined by the user programming the device. Designers can program a digital system themselves to "integrate" it onto a PLD, without needing chip manufacturers to design and manufacture dedicated integrated circuit chips. Furthermore, nowadays, instead of manually manufacturing integrated circuit chips, this programming is mostly implemented using "logic compiler" software. Similar to the software compiler used in program development, the original code before compilation must also be written in a specific programming language, called a Hardware Description Language (HDL). There are many HDLs, such as ABEL (Advanced Boolean Expression Language), AHDL (Altera Hardware Description Language), Confluence, CUPL (Cornell University Programming Language), HDCal, JHDL (Java Hardware Description Language), Lava, Lola, MyHDL, PALASM, and RHDL (Ruby Hardware Description Language). Currently, the most commonly used are VHDL (Very-High-Speed ​​Integrated Circuit Hardware Description Language) and Verilog. Those skilled in the art should also understand that by simply performing some logic programming on the method flow using one of these hardware description languages ​​and programming it into an integrated circuit, the hardware circuit implementing the logical method flow can be easily obtained.

[0239] The controller can be implemented in any suitable manner. For example, it can take the form of a microprocessor or processor and a computer-readable medium storing computer-readable program code (e.g., software or firmware) executable by the (micro)processor, logic gates, switches, application-specific integrated circuits (ASICs), programmable logic controllers, and embedded microcontrollers. Examples of controllers include, but are not limited to, the following microcontrollers: ARC 625D, Atmel AT91SAM, Microchip PIC18F26K20, and Silicon Labs C8051F320. A memory controller can also be implemented as part of the control logic of the memory. Those skilled in the art will also recognize that, in addition to implementing the controller in purely computer-readable program code form, the same functionality can be achieved by logically programming the method steps to make the controller take the form of logic gates, switches, application-specific integrated circuits, programmable logic controllers, and embedded microcontrollers. Therefore, such a controller can be considered a hardware component, and the means included therein for implementing various functions can also be considered as structures within the hardware component. Alternatively, the means for implementing various functions can be considered as both software modules implementing the method and structures within the hardware component.

[0240] The systems, devices, modules, or units described in the above embodiments can be implemented by computer chips or entities, or by products with certain functions. A typical implementation device is a computer. Specifically, a computer can be, for example, a personal computer, a laptop computer, a cellular phone, a camera phone, a smartphone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or any combination of these devices.

[0241] For ease of description, the above devices are described separately by function as various units. Of course, in implementing this application, the functions of each unit can be implemented in one or more software and / or hardware.

[0242] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0243] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0244] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0245] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0246] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0247] Memory may include non-persistent storage in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0248] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can store information using any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital character versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0249] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0250] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0251] This application can be described in the general context of computer-executable instructions, such as program modules, that are executed by a computer. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform a specific task or implement a specific abstract data type. This application can also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via a communication network. In distributed computing environments, program modules can reside in local and remote computer storage media, including storage devices.

[0252] The above description is merely an embodiment of this application and is not intended to limit this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principle of this application should be included within the scope of the claims of this application.

Claims

1. A method for processing map data, characterized in that, include: Obtain the map data to be processed; From the map data to be processed, determine the set of map features; The map feature set includes a first map feature and a second map feature that are adjacent in location; Based on the confidence values ​​of the first map element and the second map element, map element denoising processing is performed on the map element set to obtain denoised map data. Specifically, determining the map element set from the map data to be processed includes: for any first map element in the map data to be processed, determining a first type of the first map element; if the first type is a first preset type, then obtaining map elements adjacent to the first map element and belonging to a second preset type from the map data to be processed to obtain a second map element; wherein the first preset type includes ground markings and traffic signs; the second preset type includes ground arrows, traffic signs, and traffic sign poles; if the first type is a second preset type, then obtaining map elements adjacent to the first map element and belonging to the first type from the map data to be processed to obtain a second map element; and classifying the first map element and the second map element into the map element set.

2. The method according to claim 1, characterized in that, The first map element is a ground marking, and the second map element is a ground arrow; The step of performing map feature denoising processing on the map feature set based on the confidence values ​​of the first map feature and the second map feature includes: From the set of map features, a set of high-confidence ground markings is determined; Based on the high-confidence ground markings in the high-confidence ground marking set and the preset topological relationship, the ground arrows are denoised to obtain a denoised ground arrow set; the high-confidence ground markings are ground markings with confidence values ​​greater than a preset confidence threshold. Based on the ground arrows in the denoised ground arrow set and the preset topological relationship, the ground markings in the map element set, excluding the high-confidence ground marking set, are denoised to obtain a denoised ground marking set. Based on the denoised ground markings in the denoised ground marking set and the high-confidence ground markings in the high-confidence ground marking set, the ground arrows in the denoised ground arrow set are denoised.

3. The method according to claim 1, characterized in that, The first map element is a traffic sign, and the second map element is a traffic sign pole; The step of performing map feature denoising processing on the map feature set based on the confidence values ​​of the first map feature and the second map feature specifically includes: Determine whether the traffic sign is a high-confidence traffic sign and whether the traffic sign pole is a low-confidence traffic sign pole; the high-confidence traffic sign is a traffic sign with a confidence value greater than a preset confidence threshold; the low-confidence traffic sign pole is a traffic sign pole with a confidence value less than a preset confidence threshold. If the traffic sign is a high-confidence traffic sign and the traffic sign pole is a low-confidence traffic sign pole, then it is prohibited to remove the low-confidence traffic sign pole from the map element set.

4. The method according to claim 1, characterized in that, The map data to be processed includes type data and location data of traffic sign elements; the types of traffic sign elements include main traffic signs and sub-traffic signs. Before determining the set of map features from the map data to be processed, the process further includes: Based on the location data of the traffic sign elements, determine the parent-child relationship reliability between each of the traffic sign elements in the map data to be processed; Based on the parent-child relationship reliability, a set of traffic sign elements is determined from the map data to be processed; the set of traffic sign elements includes the main traffic sign and the child traffic sign that have a parent-child relationship as determined by the parent-child relationship reliability. Identify duplicate sub-traffic signs among the sub-traffic signs at the set of traffic sign elements; Based on the confidence values ​​of each of the repeated sub-traffic signs, the traffic sign element set is subjected to sub-traffic sign denoising processing.

5. The method according to claim 4, characterized in that, Both the first map element and the second map element are the main traffic sign; The step of performing map feature denoising processing on the map feature set based on the confidence values ​​of the first map feature and the second map feature specifically includes: Identify the repeating primary traffic signs at the location of the map feature set; Based on the confidence values ​​of each of the repeated main traffic signs, the map element set is subjected to main traffic sign denoising processing.

6. The method according to claim 4, characterized in that, After determining the set of traffic sign elements from the map data to be processed based on the parent-child relationship reliability, the process further includes: From the map data to be processed, determine the target main traffic sign that is located in the same lane of the same preset road segment as the traffic sign element in the traffic sign element set; The traffic sign element set and the target main traffic sign are assigned to the target traffic sign element set; Identify the target sub-traffic signs in the target traffic sign element set that are duplicates of the main traffic sign; Based on the confidence values ​​of the main traffic sign and the target sub-traffic sign, noise reduction processing is performed on the target traffic sign element set between the main traffic sign and the sub-traffic sign.

7. The method according to claim 1, characterized in that, The map data to be processed includes traffic signs; After acquiring the map data to be processed, the process also includes: For any traffic sign in the map data to be processed, determine whether the traffic sign is a high-confidence traffic sign; If the traffic sign is a high-reliability traffic sign, then determine whether the map data to be processed contains traffic sign poles that are connected to the traffic sign; If the map data to be processed does not contain any traffic sign poles that are connected to the traffic sign, then historical map data is obtained; Identify historical traffic sign poles that are connected to the traffic signs from the historical map data; Based on the historical traffic sign poles, the map data to be processed is supplemented with traffic sign poles.

8. The method according to claim 1, characterized in that, The first map element is a first ground arrow, and the second map element is a second ground arrow; The map feature set includes surface arrows located in the same lane of the same preset road segment; The step of performing map feature denoising processing on the map feature set based on the confidence values ​​of the first map feature and the second map feature specifically includes: If the first ground arrow overlaps with the second ground arrow, then all ground arrows in the first and second ground arrows that overlap, except for the ground arrow with the highest confidence value, will be removed from the map feature set.

9. The method according to claim 8, characterized in that, Before determining the set of map features from the map data to be processed, the process further includes: Identify each high-confidence ground arrow in the map data to be processed; the high-confidence ground arrow is a ground arrow with a confidence value greater than a preset confidence threshold; Based on the location information of the high-confidence ground arrows and the location information of road markings, each fifth map element set is determined from the map data to be processed; the vertical relative positions of all ground arrows in any fifth map element set are the same, and each set contains at least one high-confidence ground arrow. From the map data to be processed, a set of low-confidence ground arrows is determined; the low-confidence ground arrows are ground arrows with a confidence value less than or equal to a preset confidence threshold. For any target low-confidence ground arrow in the set of low-confidence ground arrows, determine whether the target low-confidence ground arrow meets the preset extraction conditions; If the judgment result indicates that the target low-confidence ground arrow meets the preset extraction conditions, then it is prohibited to remove the target low-confidence ground arrow from the map data to be processed.

10. The method according to claim 1, characterized in that, The first map element is a first traffic sign pole, and the second map element is a second traffic sign pole; The first traffic sign pole and the second traffic sign pole are located on the same side of the preset road; The step of performing map feature denoising processing on the map feature set based on the confidence values ​​of the first map feature and the second map feature specifically includes: Determine whether the first traffic sign pole and the second traffic sign pole are duplicate traffic sign poles; If the first traffic sign pole and the second traffic sign pole are duplicates, then all traffic sign poles except the one with the lowest confidence value will be removed from the map feature set.

11. A map data processing apparatus, characterized in that, include: The acquisition module is used to acquire map data to be processed; The determination module is used to determine a set of map features from the map data to be processed; The map feature set includes a first map feature and a second map feature that are adjacent in location; The denoising module is used to perform map element denoising processing on the map element set based on the confidence values ​​of the first map element and the second map element to obtain denoised map data. Specifically, the determining module is used to determine a first type of any first map element in the map data to be processed; if the first type is a first preset type, then obtain map elements adjacent to the first map element and belonging to a second preset type from the map data to be processed to obtain a second map element; wherein the first preset type includes ground markings and traffic signs; the second preset type includes ground arrows, traffic signs and traffic sign poles; if the first type is a second preset type, then obtain map elements adjacent to the first map element and belonging to the first type from the map data to be processed to obtain a second map element; and classify the first map element and the second map element into the map element set.

12. A map data processing device, characterized in that, include: At least one processor; as well as, A memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor to enable the at least one processor to perform the method as described in any one of claims 1-10.

Images