Traffic sign recognition method, electronic device, storage medium and program product

Abstract

Embodiments of the present application provide a traffic sign recognition method, an electronic device, a storage medium and a program product. The method comprises: obtaining attribute information of multiple traffic signs on the same road from map update data; determining corner point road coordinates and sign face projection areas of the multiple traffic signs in a road coordinate system according to the attribute information; performing clustering processing on the multiple traffic signs according to the corner point road coordinates and the sign face projection areas, to obtain multiple target clustering clusters; for each target clustering cluster, determining a target type of a target traffic sign based on the corner point road coordinates and the sign face projection areas of the target traffic sign included in the target clustering cluster; and the target type is a side-by-side combined type or a composite combined type. The method is used to improve the recognition accuracy of traffic signs.

Description

Technical Field

[0001] This application relates to the field of map data processing technology, and in particular to a method for recognizing traffic signs, electronic devices, storage media, and program products. Background Technology

[0002] In the road environment, traffic signs, such as speed limit signs, prohibition signs, and instruction signs, are important data supporting safe driving. The accuracy of traffic signs in electronic maps directly affects the ability of autonomous driving and navigation services to understand the road environment.

[0003] Traffic signs often exist in the form of combination signs. In related technologies, traffic signs are identified by sensing data collected by a sensing system. However, the spatial distribution of combination signs is complex and prone to misidentification, resulting in low accuracy in identifying traffic signs. Summary of the Invention

[0004] This application provides a method for recognizing traffic signs, electronic devices, storage media, and program products to improve the recognition accuracy of traffic signs.

[0005] In a first aspect, embodiments of this application provide a method for recognizing traffic signs, including:

[0006] Obtain attribute information of multiple traffic signs on the same road from map update data;

[0007] Based on the attribute information, determine the corner road coordinates and sign projection area of ​​multiple traffic signs in the road coordinate system;

[0008] Based on the corner road coordinates and the sign projection area, multiple traffic signs are clustered to obtain multiple target clusters;

[0009] For each target cluster, the target type of the target traffic sign is determined based on the corner road coordinates and the sign projection area of ​​the target traffic signs included in the target cluster; the target type is either a row-type combination type or a composite combination type.

[0010] Secondly, embodiments of this application provide a traffic sign recognition device, comprising:

[0011] The acquisition module is used to obtain attribute information of multiple traffic signs on the same road from map update data;

[0012] The coordinate transformation and projection module is used to determine the corner road coordinates and the projection area of ​​multiple traffic signs in the road coordinate system based on attribute information.

[0013] The clustering module is used to cluster multiple traffic signs based on the corner road coordinates and the sign projection area to obtain multiple target clusters;

[0014] The type determination module is used to determine the target type of the target traffic sign for each target cluster based on the corner road coordinates and the projection area of ​​the target traffic sign included in the target cluster; the target type is either a row-type combination type or a composite combination type.

[0015] Thirdly, embodiments of this application provide an electronic device, including: a memory and a processor; the memory stores computer-executable instructions; the processor executes the computer-executable instructions stored in the memory, causing the processor to perform the first aspect and / or various possible implementations of the first aspect as described above.

[0016] Fourthly, embodiments of this application provide a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, are used to implement the first aspect and / or various possible implementations of the first aspect.

[0017] Fifthly, embodiments of this application provide a computer program product, including a computer program that, when executed by a processor, implements the first aspect and / or various possible implementations of the first aspect.

[0018] The traffic sign recognition method provided in this application, for multiple traffic signs on the same road, determines the corner road coordinates and sign projection areas of the multiple traffic signs in the road coordinate system based on the attribute information of the multiple traffic signs in the map data. Clustering is then performed based on the corner road coordinates and sign projection areas to obtain multiple target clusters. Then, based on the corner road coordinates and sign projection areas of the target traffic signs in each target cluster, a row-type or composite combination type is determined. By representing the traffic signs in the road coordinate system, the influence of road curvature on traffic sign recognition is eliminated, making the spatial positional relationship of multiple traffic signs on the road more accurate. Clustering based on corner road coordinates and sign projection areas ensures that each target cluster includes target traffic signs with strong spatial correlation. Finally, the target type is determined based on the corner road coordinates and sign projection areas of the target traffic signs in the target cluster, effectively improving the accuracy of traffic sign recognition. Attached Figure Description

[0019] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0020] Figure 1A schematic diagram illustrating the traffic sign recognition method provided in this application;

[0021] Figure 2 A schematic diagram of the road coordinate system provided for this application;

[0022] Figure 3 A schematic diagram of multiple traffic signs of the row-type combination provided in this application;

[0023] Figure 4 A schematic diagram of multiple traffic signs of the composite combination type provided in this application;

[0024] Figure 5 A flowchart illustrating the target type of the target traffic sign provided for this application;

[0025] Figure 6 A schematic diagram of the structure of the traffic sign recognition device provided in this application;

[0026] Figure 7 A schematic diagram of the structure of the electronic device provided in this application.

[0027] The accompanying drawings illustrate specific embodiments of this application, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concept of this application to those skilled in the art through reference to particular embodiments. Detailed Implementation

[0028] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.

[0029] The technical solution of this application and how the technical solution of this application solves the above-mentioned technical problems are described in detail below with specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments. The embodiments of this application will now be described with reference to the accompanying drawings.

[0030] Figure 1 This is a flowchart illustrating the traffic sign recognition method provided in this application. The traffic sign recognition method can be applied to electronic devices, which can be either terminals or servers. Figure 1 As shown, the methods for recognizing traffic signs include:

[0031] S101. Obtain the attribute information of multiple traffic signs on the same road from the map update data.

[0032] Traffic signs are signs that convey road information to road users through graphics, colors, and text. For example, traffic signs include, but are not limited to, warning signs, prohibitory signs, instruction signs, and directional signs.

[0033] The map update data can be determined based on the perception data collected in the actual scene and the electronic map; optionally, the electronic map can be a high-precision map, where "high precision" means that the degree of proximity between the position of the geographic element in the geographic coordinate system and the real world position is measured with centimeter-level precision, and the accuracy of the spatial relationship between map elements is measured with sub-meter or centimeter-level precision.

[0034] In one implementation, the perception data collected in the actual scene is identified to obtain multiple candidate signs, and map update data is determined based on the multiple candidate signs and the electronic map; the attribute information of multiple traffic signs associated with the same road is obtained from the map update data.

[0035] Specifically, the system identifies multiple candidate traffic signs from the sensor data collected in the actual scene, and retrieves multiple original traffic signs from the electronic map. Based on the multiple original and candidate traffic signs, the system determines the attribute information of the multiple traffic signs, and then determines the map update data based on the attribute information of the multiple traffic signs. The multiple traffic signs are the union of the multiple candidate and multiple original traffic signs.

[0036] The attribute information includes location information, shape information, and status information.

[0037] Location information includes the geodetic coordinates of the corner points of the traffic sign, that is, the position of the corner points of the traffic sign in the geodetic coordinate system.

[0038] For example, the first The location information of each traffic sign is represented as follows: , It is the first Multiple corners of a traffic sign It is a positive integer greater than 1; , Corner point longitude, Corner point latitude, Corner point The elevation.

[0039] Shape information refers to the external outline shape of traffic signs, such as rectangles, circles, triangles, etc.

[0040] Status information indicates the difference between the status of traffic signs on the actual road and on the electronic map.

[0041] For example, the status information includes: added status, existing status, and deleted status; added status indicates that the traffic sign is identified as existing in the actual road, but the electronic map does not include the traffic sign, and it may be necessary to add the traffic sign to the electronic map; existing status indicates that the traffic sign is identified as existing in both the actual road and the electronic map; deleted status indicates that the traffic sign is identified as not existing in the actual road, but the electronic map includes the traffic sign, and it may be necessary to delete the traffic sign from the map data.

[0042] It should be noted that the map update data is determined based on the identification results of the perception data and the electronic map. Since the identification results of the perception data may be incorrect, if the electronic map is updated directly based on the attribute information of multiple traffic signs, there may be update errors. The embodiments of this application can be used to identify the target types of multiple traffic signs. Subsequently, multiple traffic signs can be verified based on the target types of multiple traffic signs to correct the error information. Then, the electronic map is updated based on the corrected attribute information of multiple traffic signs to ensure the accuracy of the electronic map update.

[0043] S102. Based on the attribute information, determine the corner road coordinates and sign projection area of ​​multiple traffic signs in the road coordinate system.

[0044] Among them, the road coordinate system is a coordinate system built based on the road, which can be the Frenet coordinate system.

[0045] Optionally, the method further includes: determining road reference lines based on roads associated with multiple traffic signs, and constructing a road coordinate system based on the road reference lines; such as Figure 2 As shown, the origin of the road coordinate system is the position point on the road reference line. The vertical axis of the road coordinate system is the road reference line, and the road driving direction is the positive direction. The coordinate values ​​on the vertical axis represent the distance from the origin along the road reference line. The horizontal axis of the road coordinate system is perpendicular to the road reference line. The coordinate values ​​on the horizontal axis represent the lateral offset from the origin. The vertical axis of the road coordinate system is perpendicular to the road plane formed by the vertical and horizontal axes. The coordinate values ​​on the vertical axis represent the elevation values.

[0046] Among them, the corner road coordinates are the coordinates of the corner of the traffic sign in the road coordinate system; the sign projection area is the projection of the traffic sign onto the horizontal and vertical planes of the road coordinate system. The horizontal and vertical planes are the planes formed by the horizontal and vertical axes of the road coordinate system, and are also the normal planes of the driving direction of the road (link).

[0047] In some embodiments, determining the corner road coordinates and sign projection area of ​​multiple traffic signs in the road coordinate system based on attribute information includes: for each traffic sign, obtaining the corner geodetic coordinates of the target traffic sign from the attribute information of the target traffic sign; performing coordinate transformation on the corner geodetic coordinates to obtain the corner road coordinates of the target traffic sign in the road coordinate system; and determining the sign projection area of ​​the target traffic sign in the horizontal vertical plane based on the corner road coordinates.

[0048] Specifically, for each traffic sign, the location information is obtained from the attribute information of the traffic sign. The location information includes the geodetic coordinates of the corner point of the traffic sign, that is, the coordinate representation of the corner point of the traffic sign in the geodetic coordinate system. According to the mapping relationship between the road coordinate system and the geodetic coordinate system, the geodetic coordinates of the corner point are transformed to the road coordinate system to obtain the road coordinates of the corner point of the traffic sign.

[0049] Based on the road coordinates of the corner points of the traffic sign, determine the corner point projection coordinates, that is, the coordinate representation of the corner points of the traffic sign in the horizontal and vertical planes; obtain the shape information, such as rectangle, circle, triangle, etc., from the attribute information of the traffic sign; based on the corner point projection coordinates and shape information of the traffic sign, construct a closed polygon, that is, the projection area of ​​the sign.

[0050] It should be noted that the attribute information of traffic signs also includes shape information, which includes the shape type, width, height, and orientation of the traffic sign; the shape type can be rectangular, circular, etc.; for example, the first... The shape information of a traffic sign is represented as follows: , Is it shape type, It is the width, It is height, It's the direction.

[0051] For example, the coordinates of the corner road are , The vertical coordinate represents the distance between the corner point and the origin along the road reference line. It is a horizontal coordinate, representing the lateral offset between the corner point and the origin. These are vertical coordinates, representing the elevation values ​​of the corner points; the projected coordinates of the corner points are... .

[0052] Optionally, after obtaining the corner point projection coordinates, for any two traffic signs, the positional relationship between them is determined based on the distance between their longitudinal and lateral projection coordinates, instead of using geodetic coordinates. This method eliminates positional deviations caused by road curvature; for example, for the first... The first traffic sign and the first The distance between traffic signs on their vertical axes is represented as follows: The distance between the horizontal coordinates is expressed as: .

[0053] In the above embodiments, the corner coordinates of the traffic sign are transformed to the road coordinate system to obtain the corner road coordinates. The projection area of ​​the traffic sign is determined based on the corner road coordinates. In the geodetic coordinate system, latitude and longitude may make it difficult to distinguish traffic signs in curved scenarios. In the road coordinate system, the longitudinal and lateral coordinates can clearly distinguish traffic signs in different locations. Through coordinate transformation, the influence of road curvature on traffic sign recognition is eliminated, and the accuracy of traffic sign recognition is improved.

[0054] S103. Based on the corner road coordinates and the sign projection area, multiple traffic signs are clustered to obtain multiple target clusters.

[0055] The target cluster includes at least one target traffic sign. When the target cluster includes at least two target traffic signs, the at least two target traffic signs are located close to each other and have a high degree of overlap between their projection areas.

[0056] In one implementation, the longitudinal coordinates of multiple traffic signs are determined based on the corner road coordinates, the longitudinal coordinates of any two traffic signs are determined, and the longitudinal spacing between any two traffic signs is determined; the regional overlap between any two traffic signs is determined based on the projection area of ​​the sign faces of any two traffic signs; and clustering is performed based on the longitudinal spacing and regional overlap between any two traffic signs to obtain multiple target clusters.

[0057] Specifically, multiple traffic signs are clustered to obtain multiple target clusters, with the constraint that the longitudinal spacing between traffic signs is less than the preset spacing and the regional overlap is greater than the preset overlap. That is, when a target cluster includes at least two target traffic signs, the longitudinal spacing between the at least two target traffic signs is less than the preset spacing and the regional overlap is greater than the preset overlap. The preset spacing and preset overlap can be set according to actual needs.

[0058] In one implementation, multiple traffic signs are clustered based on corner road coordinates and sign projection areas to obtain multiple target clusters. This includes: determining the longitudinal coordinates of the multiple traffic signs based on corner road coordinates; clustering the multiple traffic signs based on the longitudinal coordinates to obtain multiple initial clusters; for each initial cluster, determining the regional overlap between any two initial traffic signs included in the target initial cluster based on the sign projection areas; and performing secondary clustering on the multiple initial traffic signs included in the target initial cluster based on the regional overlap to obtain at least one target cluster.

[0059] Each traffic sign includes multiple corner road coordinates. In actual scenarios, the traffic sign is parallel to the horizontal and vertical planes of the road coordinate system. Therefore, the vertical coordinates of the multiple corner road coordinates of the traffic sign are the same.

[0060] Among them, the degree of regional overlap refers to the degree to which the projected areas of two traffic signs overlap.

[0061] The initial cluster includes at least one initial traffic sign. When the initial cluster includes at least two initial traffic signs, the longitudinal coordinates of the at least two initial traffic signs are similar.

[0062] For each traffic sign, obtain the longitudinal coordinate of any corner point of the traffic sign in the road coordinate system, and perform greedy clustering based on the longitudinal coordinates of multiple traffic signs to obtain multiple initial clusters.

[0063] Specifically, multiple traffic signs are sorted in ascending order according to their vertical coordinates to obtain a traffic sign sequence. The traffic signs are then traversed in the order of arrangement. The traffic signs obtained through this traversal are used as signs to be processed. Based on the vertical coordinates of the signs to be processed and the average vertical coordinates of the nearest initial clusters, it is determined whether the signs to be processed meet the clustering criteria. If the signs to be processed meet the clustering criteria, they are added to the nearest initial clusters, and the average vertical coordinates of the initial clusters are recalculated. If the signs to be processed do not meet the clustering criteria, a new initial cluster including the signs to be processed is created, and the step of traversing the traffic sign sequence in the order of arrangement continues until all traffic signs have been traversed and multiple initial clusters have been obtained.

[0064] For example, the sign s1 obtained in the first traversal is the first traffic sign in the sequence. Since there is no initial cluster yet, sign s1 does not meet the clustering criteria. Therefore, an initial cluster G1 including sign s1 is created, and the average vertical coordinate of the initial cluster G1 is determined. The sign s2 obtained in the second traversal is the second traffic sign in the sequence. The absolute value of the difference between the average vertical coordinate of sign s2 and the initial cluster G1 is determined. If the absolute value of the difference is less than a preset distance difference, sign s2 meets the clustering criteria and is added to the initial cluster G1. If the absolute value of the difference is not less than the preset distance difference, sign s2 does not meet the clustering criteria, and an initial cluster G2 including sign s2 is created. This process is repeated until all traffic signs are traversed, resulting in multiple initial clusters. .

[0065] Among them, the absolute value of the difference between the longitudinal coordinate of the sign to be processed and the average longitudinal coordinate of the nearest initial cluster is determined, as shown in formula (1).

[0066] Formula (1): ;

[0067] in, It is the vertical coordinate of the k-th sign to be processed during the traversal. It is the average vertical coordinate of the initial cluster Gm that is closest to the k-th sign to be processed. It is the absolute value of the difference.

[0068] For example, if the preset distance difference is 5m, for the signs to be processed obtained through the traversal, if the absolute value of the calculated difference is... If the difference is less than 5m, the sign to be processed is added to the current initial cluster. If the absolute value of the difference is... If the value is not less than 5m, then a new initial cluster is created.

[0069] The traffic signs included in the initial cluster are used as the initial traffic signs; for each initial cluster, if the number of initial traffic signs included in the initial cluster is one, then the type of the initial traffic sign is determined to be an independent type.

[0070] If the initial cluster includes at least two initial traffic signs, the overlap between any two initial traffic signs is determined based on the area of ​​the projected area of ​​any two initial traffic signs in the initial cluster. Based on the overlap, a secondary clustering is performed on the at least two initial traffic signs included in the initial cluster to obtain at least one target cluster.

[0071] Specifically, for ease of explanation, any two initial traffic signs are referred to as the first initial traffic sign and the second initial traffic sign; the difference between the longitudinal coordinates of the center points of the first initial traffic sign and the second initial traffic sign is taken as the longitudinal distance between any two initial traffic signs.

[0072] Based on the projection area of ​​the first initial traffic sign and the projection area of ​​the second initial traffic sign, the intersecting area is determined; among the projection areas of the first initial traffic sign and the second initial traffic sign, the candidate initial traffic sign with the smallest area is determined, and the ratio between the area of ​​the intersecting area and the area of ​​the candidate initial traffic sign is used as the area overlap; for example, as shown in formula (2).

[0073] Formula (2): ;

[0074] in, It is the area of ​​the projected area of ​​the first initial traffic sign. It is the area of ​​the projected area of ​​the second initial traffic sign. It is the area of ​​the intersecting region. It is the degree of overlap between the first initial traffic sign and the second initial traffic sign.

[0075] For the initial cluster of multiple initial traffic signs, connectivity clustering is performed. Specifically, each initial traffic sign is treated as an independent cluster. If the regional overlap between two independent clusters meets a threshold condition, the two independent clusters are merged to obtain a new independent cluster. This process is repeated for any two independent clusters until no two independent clusters meet the regional overlap condition. The current independent cluster is then used as the target cluster.

[0076] Among them, the region overlap between two independent clusters meets the threshold condition, which can be that the region overlap between two independent clusters is greater than a first threshold; the specific value of the first threshold can be set according to actual needs, for example, the first threshold can be 0.5.

[0077] In one alternative approach, the degree of regional overlap between any two initial traffic signs included in the target initial cluster is determined based on the sign projection area, including: obtaining the orientation information of the initial traffic signs in the attribute information for the initial traffic signs included in the target initial cluster; removing initial traffic signs with opposite orientation information; and determining the degree of regional overlap between any two initial traffic signs based on the sign projection area.

[0078] Specifically, the orientation information is the orientation angle. If the orientation angle falls within a preset angle range, the orientation information is determined to be positive; if the orientation angle does not fall within the preset angle range, the orientation information is determined to be negative. The specific value of the preset angle range can be set according to actual needs.

[0079] For example, the preset angle range is If the initial traffic sign orientation information Not satisfied: If the initial traffic sign is removed, it will not participate in the subsequent secondary clustering. By removing the reverse initial traffic sign, the amount of data participating in the secondary clustering is reduced, and the number of traffic signs for subsequent type recognition is also reduced, thus improving recognition efficiency.

[0080] In the above embodiments, the positional relationship between traffic signs can be intuitively determined by the longitudinal coordinates of the traffic signs and the degree of overlap between the projected areas of any two traffic signs. A first clustering is performed using the longitudinal coordinates, and a second clustering is performed based on the degree of overlap. This ensures that traffic signs with strong spatial correlation are aggregated into a target cluster. Subsequent identification is performed based on the target cluster, which improves the accuracy of traffic sign identification.

[0081] S104. For each target cluster, based on the corner road coordinates and sign projection area of ​​the target traffic signs included in the target cluster, determine the target type of the target traffic signs; the target type is either a row-type combination type or a composite combination type.

[0082] Among them, the row-type combination type refers to multiple traffic signs arranged horizontally or vertically on the same supporting structure, with installation gaps between the signs. In practical applications, the row-type combination type of traffic signs is also called a string sign; string signs may be mistaken for independent signs, such as... Figure 3 As shown, in actual roads, a combination of three traffic signs with the content "Gxxx Area A", "Gyyy Area B", and "C Direction" may be mistakenly identified as an independent sign.

[0083] Composite combination type refers to multiple traffic signs combined in a primary and secondary form to constitute a complete traffic instruction. There are no installation gaps between composite combination type traffic signs. In practical application, composite combination type traffic signs are also called composite signs. Composite signs may be mistaken for independent signs, such as... Figure 4 As shown, in actual roads, the main sign is an icon and text indicating the section speed measurement, and the auxiliary sign is the end point. This composite sign, consisting of the main and auxiliary signs, may be mistaken for an independent sign.

[0084] In one alternative approach, if the number of target traffic signs included in the target cluster is not greater than a second quantity threshold, then the target type of the target traffic sign is determined to be an independent type; the second quantity threshold is 2, that is, when the target cluster includes 1 target traffic sign, the target traffic sign belongs to the independent type and the target traffic sign is an independent traffic sign.

[0085] In one alternative approach, if the number of target traffic signs included in the target cluster is greater than a second quantity threshold, then the positional relationship between the target traffic signs is determined based on the corner road coordinates, the combination relationship between the target traffic signs is determined based on the sign projection area, and the target traffic signs are determined to belong to a row-type combination type or a composite combination type based on the positional and combination relationships between the target traffic signs.

[0086] Specifically, determining the positional relationship between target traffic signs based on the corner road coordinates can be achieved by determining the coordinate differences of the target traffic signs in the longitudinal, lateral, and vertical directions based on the corner road coordinates. If the coordinate differences of the target traffic signs in the longitudinal, lateral, and vertical directions satisfy the lateral rowing condition or the vertical rowing condition, then the positional relationship is determined to be a rowed positional relationship. If the coordinate differences of the target traffic signs in the longitudinal, lateral, and vertical directions do not satisfy the lateral rowing condition or the vertical rowing condition, then the positional relationship is determined to be a non-rowed positional relationship.

[0087] Among them, the projection area of ​​the sign determines the combination relationship between target traffic signs. This can be done by determining the largest reference sign among the target traffic signs based on the area of ​​the projection area of ​​the sign, and then determining the combination relationship based on the area of ​​the reference sign and the total area of ​​the target traffic signs. For example, if the ratio between the area of ​​the reference sign and the total area is less than a preset ratio, the combination relationship is a row combination relationship; if the ratio between the area of ​​the reference sign and the total area is not less than the preset ratio, the combination relationship is a composite combination relationship.

[0088] For example, Figure 3 The signs with the content "Gxxx" in area A are the baseline signs; the signs with the content "Gyyy" in area B and the signs with the content "C" indicating the direction are the supplementary signs. Figure 4 The signs indicating section speed measurement are the baseline signs, while those indicating the destination are the supplementary signs.

[0089] For example, if the positional relationship between the target traffic signs is a row positional relationship, then determine whether the combination relationship between the target traffic signs is a row combination relationship. If the combination relationship between the target traffic signs is a row combination relationship, then the target traffic signs belong to the row combination type.

[0090] If the positional relationship between the target traffic signs is not a row-together relationship, then determine whether the combination relationship between the target traffic signs is a composite combination relationship. If the combination relationship between the target traffic signs is a composite combination relationship, then the target traffic signs belong to the composite combination type.

[0091] Optionally, if the positional relationship between the target traffic signs is not a row-together positional relationship, and the combination relationship between the target traffic signs is not a composite combination relationship, then the target traffic signs belong to the independent type.

[0092] The traffic sign recognition method provided in this application, for multiple traffic signs on the same road, determines the corner road coordinates and sign projection areas of the multiple traffic signs in the road coordinate system based on the attribute information of the multiple traffic signs in the map data. Clustering is then performed based on the corner road coordinates and sign projection areas to obtain multiple target clusters. Then, based on the corner road coordinates and sign projection areas of the target traffic signs in each target cluster, a row-type or composite combination type is determined. By representing the traffic signs in the road coordinate system, the influence of road curvature on traffic sign recognition is eliminated, making the spatial positional relationship of multiple traffic signs on the road more accurate. Clustering based on corner road coordinates and sign projection areas ensures that each target cluster includes target traffic signs with strong spatial correlation. Finally, the target type is determined based on the corner road coordinates and sign projection areas of the target traffic signs in the target cluster, effectively improving the accuracy of traffic sign recognition.

[0093] In some embodiments, the target type of the target traffic sign is determined based on the corner road coordinates and the projected area of ​​the target traffic sign included in the target cluster, including:

[0094] Based on the projection area of ​​the target traffic signs included in the target cluster, determine the base sign and additional signs among the target traffic signs;

[0095] Determine the intersection ratio based on the projection areas of the base sign and the supplementary sign;

[0096] If the number of target traffic signs exceeds the first quantity threshold, the positional relationship is determined based on the corner road coordinates of the base sign and the additional sign, and the target type of the target traffic signs is determined to be a row-type combination type based on the positional relationship and the intersection ratio.

[0097] If the number of target traffic signs is not greater than the first quantity threshold but greater than the second quantity threshold, or if it is determined that the target type of the target traffic signs is not a row-type combination type, the target type is determined to be a composite combination type based on the traffic ratio.

[0098] The target cluster includes multiple target traffic signs, where "multiple" means at least two.

[0099] The baseline sign is the target traffic sign with the largest projection area; the supplementary sign is the target traffic sign other than the baseline sign among multiple target traffic signs.

[0100] The overlap ratio is determined based on the union of the projection areas of the base sign and the projection areas of the supplementary sign, and is used to reflect the degree of overlap between the union of the projection areas of the base sign and the projection areas of the supplementary sign.

[0101] The first and second quantity thresholds can be set according to actual needs. For example, the first quantity threshold can be 3 and the second quantity threshold can be 2.

[0102] The positional relationship can be either a row-together relationship or a non-row-together relationship; a row-together relationship refers to multiple target traffic signs being arranged continuously in the horizontal or vertical direction.

[0103] Specifically, for the target cluster including multiple target traffic signs, the base sign and supplementary signs are determined based on the area of ​​the projection area of ​​the multiple traffic signs. The base sign is the target traffic sign with the largest area, and the supplementary signs are the target traffic signs other than the base sign.

[0104] The intersection ratio is determined based on the area of ​​the baseline sign and the combined area of ​​the baseline signs.

[0105] If the number of target traffic signs exceeds the first quantity threshold, then based on the corner road coordinates, determine the lateral, longitudinal, and vertical coordinate differences between the base sign and each additional sign. Based on the lateral, longitudinal, and vertical coordinate differences between the base sign and each additional sign, determine whether the positional relationship of the multiple target traffic signs is a row-like or non-row-like relationship. Based on the positional relationship and the intersection ratio, determine whether the target type of the target traffic signs is a row-like type.

[0106] If the number of target traffic signs is not greater than the first quantity threshold but greater than the second quantity threshold, or if the target type is determined not to be a row-type combination type after the above process, then the target type of the target traffic signs is determined to be a composite combination type based on the traffic ratio.

[0107] In the above embodiments, the determination process for entering the row type or the composite combination type is determined based on the number of target traffic signs. The determination of whether it is the row type is determined based on the positional relationship and the intersection ratio, and the determination of whether it is the composite combination type is determined based on the intersection ratio, thereby improving the recognition efficiency of traffic signs.

[0108] In some embodiments, determining the positional relationship based on the corner road coordinates of the reference sign and the supplementary sign includes: determining the positional relationship based on the corner road coordinates of the reference sign and the supplementary sign, including: determining the coordinate difference between the reference sign and the supplementary sign based on the corner road coordinates of the reference sign and the supplementary sign; determining the target supplementary sign that meets the row position condition among the supplementary signs based on the coordinate difference; and determining whether the positional relationship is a row positional relationship or a non-row positional relationship based on the number of target supplementary signs.

[0109] Specifically, based on the corner road coordinates of the base sign, the first center point road coordinates of the base sign are determined; for each additional sign, the second center point road coordinates of the additional sign are determined; based on the first center point road coordinates and the second center point road coordinates, the elevation difference (i.e., vertical coordinate difference) and the lateral coordinate difference are determined.

[0110] Among them, the first center point road coordinates are the representation of the center point of the reference sign in the road coordinate system; the second center point road coordinates are the representation of the center point of the supplementary sign in the road coordinate system; taking the calculation of the first center point road coordinates as an example, the average of the road coordinates of multiple corner points of the reference sign is used to obtain the first center point road coordinates of the reference sign.

[0111] Obtain the lateral coordinates and elevation value of the first center point from the road coordinates of the first center point; obtain the lateral coordinates and elevation value of the second center point from the road coordinates of the second center point; calculate the difference between the lateral coordinates of the first center point and the second center point, and the difference between the elevation values ​​of the first center point and the second center point.

[0112] For each additional sign, if the elevation difference between the additional sign and the base sign is less than or equal to the elevation threshold, and the lateral coordinate difference between the additional sign and the base sign is greater than or equal to the lateral offset threshold, then the additional sign and the base sign meet the lateral alignment condition, and the additional sign is designated as the first target additional sign.

[0113] For example, the elevation threshold is 0.5m, and the lateral offset threshold is 0.3m; if ,and, If the additional sign j and the base sign i satisfy the horizontal row position condition, then the additional sign j and the base sign i satisfy the horizontal row position condition.

[0114] If the number of additional signs attached to the first target is greater than the first preset number, the positional relationship is determined to be a row positional relationship. If the number of additional signs attached to the first target is less than the first preset number, the positional relationship is determined to be a non-row positional relationship. The first preset number can be set according to actual needs, for example, the first preset number is 2.

[0115] For each additional sign, if the elevation difference between the additional sign and the base sign is greater than the elevation threshold, and the lateral coordinate difference between the additional sign and the base sign is less than the lateral offset threshold, then the additional sign and the base sign meet the vertical row position condition, and the additional sign is used as the second target additional sign.

[0116] For example, the elevation threshold is 0.5m, and the lateral offset threshold is 0.3m; if ,and, If the additional sign j and the base sign i satisfy the vertical alignment condition, then the additional sign j and the base sign i satisfy the vertical alignment condition.

[0117] If the number of additional signs for the second target is greater than the first preset number, the positional relationship is determined to be a row positional relationship; if the number of additional signs for the second target is less than the first preset number, the positional relationship is determined to be a non-row positional relationship.

[0118] In the above embodiments, based on the coordinate difference between the base sign and the additional sign in the road coordinate system, it is determined whether the additional sign meets the row position conditions, thereby determining the positional relationship and improving the accuracy of the determined positional relationship.

[0119] In some embodiments, determining the intersection ratio based on the projection areas of the base sign and the supplementary sign includes: determining a sign to be processed in the supplementary sign with a shape consistent with the base sign based on attribute information; and determining the intersection ratio based on the combined area of ​​the projection areas of the sign to be processed and the main area of ​​the base sign.

[0120] Specifically, from the attribute information of multiple traffic signs, the shape information of the baseline sign and the supplementary sign is obtained, and the supplementary sign with the same shape information as the baseline sign is taken as the sign to be processed.

[0121] When the number of signs to be processed is greater than the first preset number, a combination area is determined based on the projection area of ​​the sign to be processed. The combination area is the union of the projection areas of the sign to be processed. The area where the combination area overlaps with the projection area of ​​the reference sign is the intersection area.

[0122] The minimum area is determined based on the area of ​​the combined area and the area of ​​the projection area of ​​the reference sign; the ratio between the area of ​​the intersecting area and the minimum area is taken as the intersection ratio.

[0123] In the above embodiments, a sign to be processed with the same shape as the reference sign is identified in the additional sign. The intersection ratio is determined based on the combined area of ​​the sign to be processed and the projection area of ​​the reference sign. The intersection ratio can reflect the spatial overlap of the target traffic signs, obtain a more accurate spatial positional relationship between the signs, and improve the accuracy of the target type determined based on the intersection ratio.

[0124] In some embodiments, determining whether the target type of the target traffic sign is a row-type combination type based on the positional relationship and the intersection ratio includes: determining that the target type of the target traffic sign is a row-type combination type when the positional relationship is a row-type positional relationship and the intersection ratio belongs to a first intersection ratio range; determining whether the target type of the target traffic sign is a composite combination type based on the intersection ratio includes: determining that the target type of the target traffic sign is a composite combination type when the intersection ratio belongs to a second intersection ratio range.

[0125] The first and second intersection percentage intervals can be set according to actual needs. For example, the first intersection percentage interval is (0.75, 1] ​​and the second intersection percentage interval is (0.85, 1).

[0126] Specifically, if the number of target traffic signs exceeds the first quantity threshold, the positional relationship is a row-by-row positional relationship, and the intersection ratio belongs to the first intersection ratio range, then the target type of the target traffic signs is determined to be a row-by-row combination type.

[0127] Optionally, if the target type of the target traffic sign is determined to be a row-type combination type, a row-type combination type label is added to the target traffic sign. For example, the row-type combination type label is TAG_SERIES.

[0128] If the number of target traffic signs is no greater than the first quantity threshold but greater than the second quantity threshold, and the traffic ratio falls within the second traffic ratio range, then the target type of the target traffic sign is determined to be a composite combination type.

[0129] Optionally, if the target type of the target traffic sign is determined to be a composite type, a composite type label is added to the target traffic sign. For example, the composite type label is TAG_COMPOUND.

[0130] In a specific instance, such as Figure 5As shown, for multiple target traffic signs included in a target cluster, the baseline sign and supplementary signs are determined based on the projection area of ​​the sign face; specifically, the target traffic sign with the largest projection area is selected as the baseline sign based on the area of ​​the target traffic sign face. Other target traffic signs are treated as supplementary signs; the supplementary signs constitute a subset: .

[0131] Determine if the number of multiple target traffic signs in the target cluster is greater than 3; if the number of multiple target traffic signs in the target cluster is greater than 3, proceed to the process of determining whether it is a row-type combination; if the number of multiple target traffic signs in the target cluster is not greater than 3, determine that the target traffic signs are not a row-type combination and proceed to the process of determining whether it is a composite combination.

[0132] The process for determining the row-type combination type: Determine the benchmark signboard The coordinate difference between the additional sign and each additional sign in the subset is used to determine whether the additional sign meets the rowing position conditions. For example, if the elevation difference is less than or equal to the elevation threshold and the lateral coordinate difference is greater than or equal to the lateral offset threshold, then the lateral rowing position conditions are met; if the elevation difference is greater than the elevation threshold and the lateral coordinate difference is less than the lateral offset threshold, then the vertical rowing position conditions are met.

[0133] Determine if the number of additional signs that meet the row location conditions is greater than 2. If so, continue with the subsequent determination. If the number of additional signs that meet the row location conditions is not greater than 2, determine that the target traffic sign is not a row combination type and proceed to the determination process of whether it is a composite combination type.

[0134] In the supplementary signs, identify the signs to be processed that have the same shape as the base sign. Determine if the number of signs to be processed is greater than 2. If so, continue with the subsequent determination. Otherwise, determine if the target traffic sign is not a row-type combination and proceed to the determination process of whether it is a composite combination type.

[0135] The combination area is determined based on the projection area of ​​the sign to be processed and the projection area of ​​the reference sign, and the intersection ratio is determined; it is determined whether the intersection ratio is greater than 0.75; if the intersection ratio is greater than 0.75, the target type is determined to be a row-type combination type; otherwise, the target traffic sign is determined not to be a row-type combination type, and the process of determining whether it is a composite combination type is initiated.

[0136] The process for determining the composite combination type: If the number of multiple target traffic signs in the target cluster is greater than 2, then determine if the intersection ratio is greater than 0.85. If so, then determine that the target type is a composite combination type; otherwise, determine that the target type is not a composite combination type.

[0137] Optionally, if the target type is neither a row-type combination type nor a composite combination type, then multiple target traffic signs in the target cluster are removed.

[0138] In the above embodiments, the determination process for entering the row type or the composite combination type is determined based on the number of target traffic signs. Then, the row type is determined based on the positional relationship and the intersection ratio, and the composite combination type is determined based on the intersection ratio. By verifying multiple dimensions such as quantity constraints, row positional relationship, shape consistency, and intersection ratio, the combination type can be accurately identified. This effectively avoids the problem of row type traffic signs being misidentified as single large signs and composite combination type traffic signs being misjudged as independent signs. This ensures that the type of combination sign is consistent with the actual layout in the road environment, significantly improving the recognition accuracy of combination signs.

[0139] In some embodiments, the method further includes: determining a reference sign and an additional sign in the target traffic signs according to the sign projection area of ​​the target traffic signs included in the target cluster; obtaining the reference state of the reference sign and the additional state of the additional sign according to attribute information; verifying the reference sign, the additional sign, the reference state and the additional state according to the target type to obtain the verified traffic sign and the verified state; and updating the electronic map according to the verified traffic sign and the verified state.

[0140] The specific process for determining the base sign and supplementary sign among the target traffic signs can be found in the previous explanation on determining the base sign and supplementary sign among the target traffic signs.

[0141] In the attribute information, obtain the baseline status of the baseline sign and the additional status of the additional sign; the baseline status can be one of addition, deletion, or existence; the additional status can be one of addition, deletion, or existence.

[0142] Specifically, when the target type is a row-type combination type, if the baseline state is new and the additional state is not new, the baseline sign is removed; if part of the baseline state or the additional state exists and the other part is deleted, the deletion is changed to existence.

[0143] For example, the status of target traffic signs of the row combination type should be the same. If they are not the same, the following verification scenarios may apply:

[0144] (1) Remove falsely detected target traffic signs; if the baseline state and the additional state are different, and the baseline state is newly added, the baseline sign is determined to be falsely detected, resulting in the addition of a large outer envelope sign. The baseline sign is then removed and will not participate in subsequent electronic map updates.

[0145] (2) Correct the status of the target traffic sign that was missed; if part of the baseline status and the additional status is present and the other part is deleted, it is determined that a traffic sign with the status of deleted was missed, and the deletion is changed to existence.

[0146] Specifically, when the target type is a composite combination type, if the baseline state is deletion and the additional state is addition, the additional sign is removed; if the baseline state is deletion and the baseline state is existence, the baseline state is corrected to existence.

[0147] For example, the status of target traffic signs of composite combination types should be the same. If they are not the same, the following verification scenarios may apply:

[0148] (1) Remove false positives for additional signs; additional signs should exist in relation to the base sign. If the base sign is deleted and the additional sign is added, the added additional sign has no factual basis and is judged as a false positive. The additional sign is removed and will not participate in subsequent electronic map updates.

[0149] (2) Correct the status of the missed benchmark signs; the benchmark status is deleted and the additional status is present. That is, the additional signs identified on the actual road also exist in the electronic map, but the benchmark signs that exist in the electronic map are not identified on the actual road. It is determined that the benchmark signs are missed and the benchmark status is changed to present.

[0150] In the above embodiments, based on the update characteristics of the row-type and composite-type traffic signs, the baseline signs, additional signs, baseline status, and additional status are verified. This can effectively correct errors caused by false detections and missed detections. For example, it can reduce invalid or erroneous map update data such as adding falsely detected signs, deleting missed signs, and incorrect sign statuses. This reduces the cost of manually correcting recognition results and status data. The electronic map is updated based on the verified traffic signs and verified statuses, avoiding distortion of the updated electronic map and improving its accuracy.

[0151] The traffic sign recognition method provided in this application, for multiple traffic signs on the same road, determines the corner road coordinates and sign projection areas of the multiple traffic signs in the road coordinate system based on the attribute information of the multiple traffic signs in the map data. Clustering is then performed based on the corner road coordinates and sign projection areas to obtain multiple target clusters. Then, based on the corner road coordinates and sign projection areas of the target traffic signs in each target cluster, a row-type or composite combination type is determined. By representing the traffic signs in the road coordinate system, the influence of road curvature on traffic sign recognition is eliminated, making the spatial positional relationship of multiple traffic signs on the road more accurate. Clustering based on corner road coordinates and sign projection areas ensures that each target cluster includes target traffic signs with strong spatial correlation. Finally, the target type is determined based on the corner road coordinates and sign projection areas of the target traffic signs in the target cluster, effectively improving the accuracy of traffic sign recognition.

[0152] It should be understood that although the steps in the flowcharts of the above embodiments are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the above embodiments may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0153] Figure 6 A schematic diagram of the structure of the traffic sign recognition device provided in this application is shown below. Figure 6 As shown, this embodiment provides a traffic sign recognition device 60, which includes:

[0154] The acquisition module 601 is used to obtain attribute information of multiple traffic signs on the same road from map update data;

[0155] The coordinate transformation and projection module 602 is used to determine the corner road coordinates and the projection area of ​​multiple traffic signs in the road coordinate system based on attribute information.

[0156] Clustering module 603 is used to cluster multiple traffic signs based on the corner road coordinates and the sign projection area to obtain multiple target clusters;

[0157] The type determination module 604 is used to determine the target type of the target traffic sign for each target cluster based on the corner road coordinates and the projection area of ​​the target traffic sign included in the target cluster; the target type is either a row-type combination type or a composite combination type.

[0158] In some embodiments, the clustering module 603 is used to determine the longitudinal coordinates of multiple traffic signs based on the corner road coordinates; cluster the multiple traffic signs based on the longitudinal coordinates to obtain multiple initial clusters; for each initial cluster, determine the regional overlap between any two initial traffic signs included in the target initial cluster based on the sign projection area; and perform secondary clustering on the multiple initial traffic signs included in the target initial cluster based on the regional overlap to obtain at least one target cluster.

[0159] In some embodiments, the type determination module 604 is configured to determine a base sign and an additional sign in the target traffic signs based on the sign projection area of ​​the target traffic signs included in the target cluster; determine the intersection ratio based on the sign projection areas of the base sign and the additional sign; if the number of target traffic signs is greater than a first quantity threshold, determine the positional relationship based on the corner road coordinates of the base sign and the additional sign, and determine whether the target type of the target traffic signs is a row-type combination type based on the positional relationship and the intersection ratio; if the number of target traffic signs is not greater than the first quantity threshold but greater than a second quantity threshold, or if the target type is determined not to be a row-type combination type, determine whether the target type of the target traffic signs is a composite combination type based on the intersection ratio.

[0160] In some embodiments, the type determination module 604 is configured to determine, based on attribute information, a sign to be processed that has the same shape as the reference sign in the additional sign; determine a combined area based on the sign projection area of ​​the sign to be processed; determine the intersection area between the combined area and the sign projection area of ​​the reference sign; and determine the intersection ratio based on the intersection area, the combined area, and the sign projection area of ​​the reference sign.

[0161] In some embodiments, the type determination module 604 is used to determine the coordinate difference between the reference sign and the additional sign based on the corner road coordinates of the reference sign and the additional sign; to determine the target additional sign that meets the row position condition in the additional sign based on the coordinate difference; and to determine the position relationship as a row position relationship or a non-row position relationship based on the number of target additional signs.

[0162] In some embodiments, the type determination module 604 is used to determine that the target type of the target traffic sign is a row-type combination type when the positional relationship is a row-type positional relationship and the intersection ratio belongs to a first intersection ratio interval; and to determine whether the target type of the target traffic sign is a composite combination type based on the intersection ratio, including: determining that the target type of the target traffic sign is a composite combination type when the intersection ratio belongs to a second intersection ratio interval.

[0163] In some embodiments, the coordinate transformation and projection module 602 is used to obtain the corner geodetic coordinates of each traffic sign from the attribute information of the traffic sign; perform coordinate transformation on the corner geodetic coordinates to obtain the corner road coordinates of the corner of the traffic sign in the road coordinate system; and determine the sign projection area of ​​the traffic sign in the horizontal vertical plane based on the corner road coordinates.

[0164] In some embodiments, the apparatus further includes: a map update module, configured to determine a reference sign and an additional sign in the target traffic signs based on the sign projection area of ​​the target traffic signs included in the target cluster; obtain the reference state of the reference sign and the additional state of the additional sign based on attribute information; verify the reference sign, the additional sign, the reference state, and the additional state according to the target type to obtain the verified traffic signs and the verified state; and update the electronic map based on the verified traffic signs and the verified state.

[0165] The traffic sign recognition device provided in this embodiment can perform the traffic sign recognition method provided in the above method embodiment. Its implementation principle and technical effect are similar, and will not be described in detail here.

[0166] Figure 7 A schematic diagram of the structure of the electronic device provided in this application. Figure 7 As shown, the electronic device 70 provided in this embodiment includes at least one processor 701 and a memory 702. Optionally, the device 70 further includes a communication component 703. The processor 701, memory 702, and communication component 703 are connected via a bus.

[0167] In a specific implementation, at least one processor 701 executes computer execution instructions stored in memory 702, causing at least one processor 701 to perform the above-described method.

[0168] The specific implementation process of processor 701 can be found in the above method embodiments, and its implementation principle and technical effect are similar. It will not be repeated here.

[0169] In the above embodiments, it should be understood that the processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), etc. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the method disclosed in this invention can be directly implemented by a hardware processor, or implemented by a combination of hardware and software modules within the processor.

[0170] The memory may include random access memory (RAM) and may also include non-volatile memory (NVM), such as at least one disk storage device.

[0171] The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. Buses can be categorized as address buses, data buses, control buses, etc. For ease of illustration, the buses shown in the accompanying drawings are not limited to a single bus or a single type of bus.

[0172] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the above-described method.

[0173] This application also provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, implement the above-described method.

[0174] The aforementioned readable storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. The readable storage medium can be any available medium accessible to a general-purpose or special-purpose computer.

[0175] An exemplary readable storage medium is coupled to a processor, enabling the processor to read information from and write information to the readable storage medium. Of course, the readable storage medium can also be a component of the processor. The processor and the readable storage medium can reside in an Application Specific Integrated Circuit (ASIC). Alternatively, the processor and the readable storage medium can exist as discrete components in the device.

[0176] The division of units is merely a logical functional division; in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be indirect coupling or communication connection through some interfaces, devices, or units, and may be electrical, mechanical, or other forms.

[0177] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0178] In addition, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0179] If a function is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0180] Those skilled in the art will understand that all or part of the steps of the above-described method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When executed, the program performs the steps of the above-described method embodiments; and the aforementioned storage medium includes various media capable of storing program code, such as ROM, RAM, magnetic disks, or optical disks.

[0181] Finally, it should be noted that other embodiments of the invention will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This invention is intended to cover any variations, uses, or adaptations of the invention that follow the general principles of the invention and include common knowledge or customary techniques in the art not disclosed herein, and is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of the invention is limited only by the appended claims.

Claims

1. A method for recognizing traffic signs, characterized in that, include: Obtain attribute information of multiple traffic signs on the same road from map update data; Based on the attribute information, determine the corner road coordinates and sign projection area of ​​the multiple traffic signs in the road coordinate system; Based on the corner road coordinates and the sign projection area, the multiple traffic signs are clustered to obtain multiple target clusters; For each target cluster, the target type of the target traffic sign is determined based on the corner road coordinates and the sign projection area of ​​the target traffic sign included in the target cluster; the target type is either a row-type combination type or a composite combination type.

2. The method according to claim 1, characterized in that, The traffic signs are clustered based on the corner road coordinates and the sign projection area to obtain multiple target clusters, including: Based on the corner road coordinates, determine the longitudinal coordinates of the multiple traffic signs; Based on the vertical coordinates, the multiple traffic signs are clustered to obtain multiple initial clusters; For each initial cluster, the degree of regional overlap between any two initial traffic signs included in the target initial cluster is determined based on the projected area of ​​the sign face. Based on the degree of regional overlap, a secondary clustering is performed on the multiple initial traffic signs included in the initial cluster to obtain at least one target cluster.

3. The method according to claim 1, characterized in that, The process of determining the target type of the target traffic sign based on the corner road coordinates and the sign projection area of ​​the target traffic sign included in the target cluster includes: Based on the projection area of ​​the target traffic signs included in the target cluster, a base sign and an additional sign are determined in the target traffic signs; The intersection ratio is determined based on the projection area of ​​the base sign and the additional sign. If the number of target traffic signs is greater than a first quantity threshold, the positional relationship is determined based on the corner road coordinates of the base sign and the additional sign, and the target type of the target traffic signs is determined to be a row-type combination type based on the positional relationship and the intersection ratio. If the number of target traffic signs is not greater than the first quantity threshold but greater than the second quantity threshold, or if it is determined that the target type is not a row-type combination type, the target type of the target traffic sign is determined to be a composite combination type based on the traffic ratio.

4. The method according to claim 3, characterized in that, Determining the intersection ratio based on the projection areas of the reference sign and the additional sign includes: Based on the attribute information, a sign to be processed that has the same shape as the reference sign is identified in the additional signage; The combined area is determined based on the projection area of ​​the sign face of the sign to be processed; Determine the intersection area between the combined area and the projection area of ​​the reference sign; The intersection ratio is determined based on the intersecting area, the combined area, and the projection area of ​​the reference sign.

5. The method according to claim 3, characterized in that, Determining the positional relationship based on the corner road coordinates of the reference sign and the additional sign includes: Based on the corner road coordinates of the reference sign and the additional sign, determine the coordinate difference between the reference sign and the additional sign; Based on the coordinate difference, determine the target additional sign that meets the row position conditions among the additional signboards; Based on the number of additional signs attached to the target, the positional relationship is determined to be either a row-together positional relationship or a non-row-together positional relationship.

6. The method according to claim 3, characterized in that, The step of determining whether the target type of the target traffic sign is a row-type combination type based on the positional relationship and the intersection ratio includes: If the positional relationship is a row positional relationship and the intersection ratio belongs to the first intersection ratio interval, the target type of the target traffic sign is determined to be a row combination type; The step of determining whether the target type of the target traffic sign is a composite combination type based on the intersection ratio includes: If the intersection ratio falls within the second intersection ratio range, the target type of the target traffic sign is determined to be a composite combination type.

7. The method according to any one of claims 1 to 6, characterized in that, The step of determining the corner road coordinates and sign projection area of ​​the plurality of traffic signs in the road coordinate system based on the attribute information includes: For each traffic sign, obtain the geodetic coordinates of its corner points from the attribute information of the traffic sign. The geodetic coordinates of the corner point are transformed to obtain the corner point road coordinates of the traffic sign in the road coordinate system; Based on the road coordinates of the corner points, determine the projection area of ​​the traffic sign in the horizontal and vertical plane.

8. The method according to any one of claims 1 to 6, characterized in that, The method further includes: Based on the projection area of ​​the target traffic signs included in the target cluster, a base sign and an additional sign are determined in the target traffic signs; Based on the attribute information, obtain the baseline state of the baseline sign and the additional state of the additional sign; Based on the target type, the baseline sign, the additional sign, the baseline state, and the additional state are verified to obtain the verified traffic sign and the verified state. The electronic map is updated based on the verified traffic signs and the verified status.

9. A traffic sign recognition device, characterized in that, include: The acquisition module is used to obtain attribute information of multiple traffic signs on the same road from map update data; The coordinate transformation and projection module is used to determine the corner road coordinates and the projection area of ​​the multiple traffic signs in the road coordinate system based on the attribute information. The clustering module is used to perform clustering processing on the multiple traffic signs based on the corner road coordinates and the sign projection area to obtain multiple target clusters; The type determination module is used to determine the target type of the target traffic sign for each target cluster based on the corner road coordinates and the projection area of ​​the sign face of the target traffic sign included in the target cluster; the target type is a row-type combination type or a composite combination type.

10. An electronic device / computer-readable storage medium / computer program product, characterized in that, The electronic device includes: a processor, and a memory communicatively connected to the processor; the memory stores computer-executable instructions; the processor executes the computer-executable instructions stored in the memory to implement the method as described in any one of claims 1 to 8; The computer-readable storage medium stores computer-executable instructions, which, when executed by a processor, are used to implement the method as described in any one of claims 1 to 8; The computer program product includes computer execution instructions that, when executed by a processor, implement the method as described in any one of claims 1 to 8.

Images