Map data update method, device and electronic device

Abstract

The present disclosure provides a method, apparatus and electronic device for updating map data, relating to the technical field of data processing, and specifically to the technical field of map. The specific implementation solution is as follows: obtaining image data including ground roads and a target image accuracy of the image data; performing image recognition processing on the image data to obtain first marking information of target traffic signs on the ground roads; and updating map data related to the target traffic signs according to a target update method corresponding to the target image accuracy, where the map data can construct at least two maps with different road accuracy levels.

Description

Technical Field

[0001] The present disclosure relates to the technical field of data processing, in particular to the technical field of map technology, and specifically relates to a method, device and electronic device for updating map data. Background Art

[0002] In the production operation of map data, when depicting traffic signs such as guiding arrows on ground roads, it is usually necessary to produce map data with different depiction accuracies of traffic signs according to different business and application requirements, etc., in order to produce electronic maps with different road accuracy levels, such as high-precision electronic maps, lane-level electronic maps and standard navigation electronic maps.

[0003] Currently, due to the different accuracies of traffic signs on ground roads in different data sources, in the production operation of map data, usually different data sources with different accuracies are respectively pushed to different production devices for updating map data, in order to produce electronic maps with corresponding road accuracy levels. Summary of the Invention

[0004] The present disclosure provides a method, device and electronic device for updating map data.

[0005] According to a first aspect of the present disclosure, there is provided a method for updating map data, including:

[0006] Obtaining image data including ground roads and a target image accuracy of the image data;

[0007] Performing image recognition processing on the image data to obtain first marking information of a target traffic sign on the ground road;

[0008] Based on the first marking information, updating map data related to the target traffic sign according to a target update method corresponding to the target image accuracy, where the map data can construct at least two maps with different road accuracy levels.

[0009] According to a second aspect of the present disclosure, there is provided a device for updating map data, including:

[0010] An obtaining module, configured to obtain image data including ground roads and a target image accuracy of the image data;

[0011] An image recognition module, configured to perform image recognition processing on the image data to obtain first marking information of a target traffic sign on the ground road;

[0012] An updating module, configured to update map data related to the target traffic sign according to a target update method corresponding to the target image accuracy, where the map data can construct at least two maps with different road accuracy levels.

[0013] According to a third aspect of the present disclosure, there is provided an electronic device, including:

[0014] at least one processor; and

[0015] a memory communicatively connected to the at least one processor; wherein,

[0016] the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute any one of the methods in the first aspect.

[0017] According to a fourth aspect of the present disclosure, there is provided a non-transitory computer-readable storage medium storing computer instructions for causing a computer to execute any one of the methods in the first aspect.

[0018] According to a fifth aspect of the present disclosure, there is provided a computer program product including a computer program which, when executed by a processor, implements any one of the methods in the first aspect.

[0019] The technology according to the present disclosure solves the problem that the update efficiency of map data is relatively poor, and improves the update efficiency of map data.

[0020] It should be understood that the content described in this part is not intended to identify the key or important features of the embodiments of the present disclosure, nor is it used to limit the scope of the present disclosure. Other features of the present disclosure will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] The drawings are used to better understand the solution and do not constitute a limitation to the present disclosure. Among them:

[0022] Figure 1 is a schematic flowchart of a method for updating map data according to a first embodiment of the present disclosure;

[0023] Figure 2 is a schematic flowchart of updating map data related to a target traffic sign based on high-precision data;

[0024] Figure 3 is a schematic flowchart of updating map data related to a target traffic sign based on low-precision data;

[0025] Figure 4 is a schematic structural diagram of a map data updating device according to a second embodiment of the present disclosure;

[0026] Figure 5 is a schematic block diagram of an exemplary electronic device for implementing the embodiments of the present disclosure. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0027] The following describes exemplary embodiments of the present disclosure with reference to the accompanying drawings. Various details of the embodiments of the present disclosure are included to facilitate understanding, and they should be considered merely exemplary. Therefore, those of ordinary skill in the art should recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of the present disclosure. Similarly, descriptions of well-known functions and structures are omitted below for clarity and conciseness.

[0028] First Embodiment

[0029] As Figure 1 shown, the present disclosure provides a method for updating map data, including the following steps:

[0030] Step S101: Obtain image data including ground roads and the target image accuracy of the image data.

[0031] In this embodiment, the map data update method relates to the field of data processing, particularly to the field of map technology, and can be widely applied to the scenario of electronic map production. The map data update method of the embodiments of the present disclosure can be executed by the map data update device of the embodiments of the present disclosure. The map data update device of the embodiments of the present disclosure can be configured in any electronic device to execute the map data update method of the embodiments of the present disclosure. The electronic device can be a server or a terminal device, and no specific limitation is provided here.

[0032] The ground road can refer to a road pre-planned on the ground, and the image data can be data for updating map data, which includes at least one image. Among them, the image data including ground roads can refer to that the images in the image data include the image content of the ground roads.

[0033] According to different sources of the image data, the image data can be divided into photo results collected by a collection vehicle. For example, the image data includes trajectory point photo information collected by the collection vehicle every few meters, and top view results collected by point cloud panoramic collection, which can restore ground information through stitching processing.

[0034] Therefore, in an alternative embodiment, the map data update device can receive image data pushed by other devices, such as receiving image data pushed by a collection vehicle or receiving image data pushed by a point cloud panoramic collection device. In another alternative embodiment, the map data update device can also obtain pre-stored image data, or can also obtain image data collected by its camera device.

[0035] The target image accuracy can be the image accuracy of the image data for updating map data. Among them, the image accuracy of the image data can be divided into two types, namely high accuracy and low accuracy, and the target image accuracy is one of high accuracy and low accuracy.

[0036] Correspondingly, according to the image precision of the image data, the image data can be divided into high-precision data and low-precision data. Usually, the image data collected by the collection vehicle is low-precision data, while the image data obtained by point cloud panoramic collection and stitching has a smaller precision error and is high-precision data. That is, in an alternative embodiment, the target image precision of the image data can be determined according to the source of the image resource. For example, when the source of the image data is the data pushed by the collection vehicle, the target image precision is low precision; when the source of the image data is the data pushed by the point cloud panoramic collection device, the target image precision is high precision.

[0037] In another alternative embodiment, the target image precision associated with the image data can be obtained. That is, when other devices push data, the image data can be associated with the image precision and then pushed to the map data update device. The map data update device can also obtain the target image precision through image processing methods.

[0038] Step S102: Perform image recognition processing on the image data to obtain the first marking information of the target traffic signs on the ground road.

[0039] In this step, the ground road involved in the image data usually includes traffic signs, such as guiding arrow signs, speed limit signs, electronic eye signs, marking signs, etc. The target traffic signs can be all traffic signs on the ground road or preset traffic signs, such as guiding arrow signs. In the following embodiments, the guiding arrow sign will be used as an example to illustrate the target traffic signs.

[0040] The first marking information can be used to mark the entity-related information of the target traffic signs. The entity-related information can include the information of the target traffic signs themselves, the relationship information between the target traffic signs and other traffic signs, the relationship information between the target traffic signs and the attached road network, etc. Among them, the attached road network can refer to the lanes and / or roads to which the target traffic signs are bound. For example, the relationship information between the target traffic signs and the attached road network can include the lane centerlines and lane numbers to which the target traffic signs are bound.

[0041] The image recognition processing can include an image detection sub-operation and an image recognition sub-operation. The image detection sub-operation is used to detect the entities in the image of the image data to obtain the entity-related information in the image. Among them, the entities in the image can include traffic signs, vehicles, pedestrians, etc. The entity-related information can include the bounding boxes of the entities and the semantic attributes of the entities themselves (such as pedestrians, vehicles, traffic signs). The image recognition sub-operation is used to identify the entity-related information of the entities with semantic attributes of traffic signs and being target traffic signs based on the entity-related information obtained by the image detection sub-operation, so as to obtain the first marking information.

[0042] Step S103: Based on the first marking information, update the map data related to the target traffic sign according to a target update method corresponding to the target image accuracy, where the map data can construct at least two maps with different road accuracy levels.

[0043] Constructing an electronic map usually requires map data of various entities, such as map data of ground roads, surrounding buildings, green belts, etc. The map data related to the target traffic sign refers to the map data that can construct the target traffic sign image content of the ground road in the electronic map.

[0044] It should be noted that when constructing the map image of the ground road in the electronic map, in addition to the map data related to the target traffic sign, other map data may also be required, such as map data related to other traffic signs. For example, when the target traffic sign is a guiding arrow sign, when constructing the map image of the ground road in the electronic map, not only the map data related to the guiding arrow sign is required, but also the map data related to electronic eye signs, speed limit signs, etc.

[0045] The map data related to the target traffic sign can be stored in a preset database. The preset database can store only the map data related to the target traffic sign, or can also store other map data, and the map data related to the target traffic sign is stored separately from other map data to facilitate the extraction of the map data related to the target traffic sign from the preset database.

[0046] The map data related to the target traffic sign can be used to construct at least two maps with different road accuracy levels, such as to construct a high-precision electronic map, a lane-level electronic map, and a standard navigation electronic map. The road accuracy levels of these three electronic maps are all different, and specifically can be used to construct the image content of the target traffic sign of the ground road in at least two maps with different road accuracy levels.

[0047] The target update method can be a method for updating the map data related to the target traffic sign, and can include two situations. The first situation can be: first update the map data used to construct an electronic map with a high road accuracy level (such as a high-precision electronic map, a lane-level electronic map), and then update the map data used to construct an electronic map with a low road accuracy level, such as a standard navigation electronic map. The second situation is related to the first situation, that is, first update the map data used to construct an electronic map with a low road accuracy level, and then update the map data used to construct an electronic map with a high road accuracy level.

[0048] The target update method can correspond to the target image accuracy. For example, when the target image accuracy is high precision, the target update method selects the update method of the first situation. When the target image accuracy is low precision, the target update method selects the update method of the second situation to ensure the timeliness of map data update.

[0049] Based on the first marking information, the map data related to the target traffic sign can be updated according to the target update method. In an optional implementation manner, the difference information between the first marking information and the map data related to the target traffic sign can be compared, and based on this difference information, the map data related to the target traffic sign can be updated according to the target update method.

[0050] In another optional implementation manner, the first difference information between the first marking information and the first target map data can be compared first, and based on this first difference information, the first target map data can be updated. The first target map data can be the data of the first map used to construct the road accuracy level corresponding to the target image accuracy in the map data related to the target traffic sign. For example, when the target image accuracy is high accuracy, the map data for constructing the electronic map with a high road accuracy level is updated first.

[0051] After that, the second difference information between the updated first target map data and the second target map data can be compared, and based on this second difference information, the second target map data can be updated. The second target map data is the data of the second map in the map data related to the target traffic sign. The road accuracy level of the first map is different from that of the second map. For example, the second map is the map data of the electronic map with a low road accuracy level.

[0052] In this embodiment, by obtaining the image data including the ground road and the target image accuracy of the image data; performing image recognition processing on the image data to obtain the first marking information of the target traffic sign on the ground road; based on the first marking information, updating the map data related to the target traffic sign according to the target update method corresponding to the target image accuracy, the map data can construct at least two maps with different road accuracy levels. In this way, according to the target image accuracy of the image data, the corresponding update method can be used to perform integrated update on the map data for constructing at least two maps with different road accuracy levels, avoiding separately pushing the same image data to the data service production lines of maps with different road accuracy levels for update, ensuring the timeliness of map data update, and avoiding the repeated use of resources and the repeated production of data, thereby improving the update efficiency of map data and reducing the resource cost of map data update.

[0053] Optionally, the step S102 specifically includes:

[0054] Performing image detection on the image data to obtain the physical and geometric attribute information of the traffic signs on the ground road;

[0055] Based on the pre-acquired traffic sign coding information and the physical and geometric attribute information, perform semantic parsing on the traffic signs on the ground road to obtain semantic information with the semantic attribute being the target traffic sign;

[0056] Obtain the relationship information between the target traffic sign and the attached road network, where the attached road network is the lane and / or road bound to the target traffic sign;

[0057] Among them, the first marking information includes the physical and geometric attribute information, semantic information, and the relationship information of the target traffic sign.

[0058] In this embodiment, the first marking information may include the physical and geometric attribute information, semantic information, and relationship information of the target traffic sign.

[0059] A deep learning model can be used to perform image detection on image data in an existing or new manner to obtain the bounding boxes of entities on the ground road. Based on the characteristics of traffic signs, the bounding boxes of traffic signs are obtained from the bounding boxes of entities. Based on the bounding boxes, the physical and geometric attribute information of traffic signs on the ground road can be determined. The physical and geometric attribute information can characterize the inherent geometric attribute features of entities, including spatial position, color, volume, shape, material, composition, etc. For example, the physical and geometric attribute information of traffic signs can include the outline, length, width, height, angle, orientation, and position of traffic signs.

[0060] The traffic sign coding information can include the semantic content of traffic signs defined based on national standards, services, application requirements, etc. The map data update device can pre-store this traffic sign coding information. Correspondingly, based on the traffic sign coding information and the physical and geometric attribute information of traffic signs, semantic parsing can be performed on the traffic signs on the ground road to obtain semantic information with the semantic attribute being the target traffic sign.

[0061] Among them, the semantic information of the target traffic sign can include semantic attribute information and semantic relationship information. The semantic attribute information can characterize the key semantic attribute features of entities, which can be the entity semantic definitions extracted based on national standards, services, application requirements, etc., such as guiding arrows, speed limits, electronic eyes, etc.

[0062] The semantic relationship information can include the semantic information of the entity itself. For example, for the guiding arrow at an intersection, the semantic information of the entity itself can express the connected path of the guiding arrow (such as going straight, turning left, turning right, etc.). The semantic relationship information can also include the abstract semantic information between the target traffic sign and other entities. For example, when the lane marking bound to the guiding arrow is a dotted line, it can indicate that the guiding arrow is for guiding buses during the peak commuting hours. Another example is that when the lane marking bound to the guiding arrow at an intersection is a solid line, it can indicate that the guiding arrow has intersection restrictions.

[0063] Based on the physical geometric attribute information of traffic signs, the semantic attribute information of the target traffic sign and the semantic attribute information of other traffic signs can be matched from the traffic sign coding information. Based on the semantic attribute information and physical geometric attribute information of different entities, the abstract semantic information between the target traffic sign and other entities such as other traffic signs can be parsed out.

[0064] In addition, the relationship information between the target traffic sign and the attached road network may include the first relationship information between the target traffic sign and the bound lane, and / or the second relationship information between the target traffic sign and the bound road.

[0065] The first relationship information can characterize the lane attributes bound by the target traffic sign. The lane attributes may include the lane center line, lane number, etc. The second relationship information can characterize the road attributes bound by the target traffic sign. The road attributes may include road lines, road numbers, etc.

[0066] Wherein, a road may include at least one lane. Correspondingly, the road attributes of the target traffic sign are aggregated into groups by the lane attributes of each lane in the road. For example, the lane numbers of each lane in a road can be aggregated into a group.

[0067] Scene recognition can be performed on the image data to identify the lane attributes and / or road attributes of the road network to which the target traffic sign in the image data belongs. Alternatively, the lane attributes and / or road attributes of the road network to which the target traffic sign belongs can be determined by obtaining the acquisition location of the image data. For example, if the acquisition location is Road A, the road attributes of the road network to which the target traffic sign belongs are the attributes of Road A.

[0068] In this embodiment, by performing image detection on the image data, the physical geometric attribute information of the traffic signs on the ground road is obtained; based on the pre-obtained traffic sign coding information and the physical geometric attribute information, semantic parsing is performed on the traffic signs on the ground road to obtain the semantic information whose semantic attribute is the target traffic sign; the relationship information between the target traffic sign and the attached road network is obtained, and the attached road network is the lane and / or road bound by the target traffic sign. In this way, the acquisition of the first marking information of the target traffic sign can be realized, laying a foundation for the data source of subsequent map data update.

[0069] Optionally, the step S103 specifically includes:

[0070] Compare the first marker information with the first target map data to obtain the first update data of the first map. The road accuracy level of the first map corresponds to the target image accuracy. The first target map data is the data in the map data used to construct the first map. The first target map data is located in the first data layer of the preset database, and the preset database is used to store the map data;

[0071] Update the first target map data based on the first update data;

[0072] Based on the lane numbers used to associate different data layers in the preset database, compare the updated first target map data with the second target map data to obtain the second update data of the second map. The second target map data is the data in the map data used to construct the second map. The second target map data is located in the second data layer of the preset database, and the road accuracy level of the second map is different from that of the first map;

[0073] Update the second target map data based on the second update data.

[0074] In this embodiment, when the map data related to the target traffic signs is stored in layers in the preset database, the map data used to construct maps with different road accuracy levels can be integrally updated through two comparison processes.

[0075] Specifically, the preset database may include a first data layer and a second data layer. The first data layer may store the map data related to the target traffic signs in the first map, and the second data layer may store the map data related to the target traffic signs in the second map. The road accuracy levels of the first map and the second map are different.

[0076] The map data related to the target traffic signs in the first map can be updated first, and then the map data related to the target traffic signs in the second map can be updated. The road accuracy level of the first map corresponds to the target image accuracy.

[0077] For example, when the target image accuracy is high, the first map may include a lane-level electronic map and a high-precision electronic map, and the second map may be a standard navigation electronic map. Correspondingly, the first target map data stored in the first data layer is data for constructing the lane-level electronic map and the high-precision electronic map. Among them, compared with the lane-level electronic map, the high-precision electronic map can depict entities such as traffic signs in more detail and specifically. That is, when constructing the high-precision electronic map, on the basis of the map data of the lane-level electronic map, other map data for depicting entities, such as material texture, thickness, etc., can also be used. The second target map data stored in the second data layer is data for constructing the standard navigation electronic map.

[0078] For another example, when the target image accuracy is low, the first map may be a standard navigation electronic map, and the second map may include a high-precision electronic map and a lane-level electronic map. Correspondingly, the first target map data stored in the first data layer is data for constructing the standard navigation electronic map. The second target map data stored in the second data layer is data for constructing the lane-level electronic map and the high-precision electronic map.

[0079] Since the standard navigation electronic map and the lane-level electronic map are essentially the relationship between roads and lanes, and roads can be aggregated from lanes, the first data layer and the second data layer can be associated by lane numbers, so that the preset database can represent map data with different data structures through a data model, thereby constructing maps with different road accuracy levels.

[0080] The first target map data can be updated by comparing the first marking information with the first target map data. Specifically, the first marking information can be compared with the first target map data stored in the first data layer of the preset database to obtain the first update data of the first map. The first update data can represent the first difference information between the first marking information and the first target map data. Correspondingly, the first target map data can be automatically updated based on the first update data.

[0081] For example, the first marking information may include information related to the guiding arrow on lane B. If the first target map data does not include the map data of the guiding arrow on lane B, the information related to the guiding arrow on lane B can be added to the first target map data. If the first target map data includes the map data of the guiding arrow on lane B, but its map data is different from the information related to the guiding arrow on lane B in the first marking information, the map data of the guiding arrow on lane B in the first target map data can be modified according to the first marking information.

[0082] Based on the update of the first target map data, the updated first target map data can be compared with the second target map data to determine lane change information. The lane change information can indicate which lane the target traffic sign, such as a guiding arrow, has changed. Based on the lane numbers used to associate different data layers in the preset database and the lane change information, the second update data of the second map can be determined. The second update data can represent the second difference information between the updated first target map data and the second target map data. Accordingly, the second target map data can be updated based on the second update information.

[0083] Among them, both the first data layer and the second data layer are concepts of virtual data layers, which vary according to different target image accuracies. For example, the data layers in the preset database in terms of physical concepts include a first data sub-layer, a second data sub-layer, a third data sub-layer, and a fourth data sub-layer. The first data sub-layer and the second data sub-layer are used to construct a high-precision electronic map and a lane-level electronic map, and the third data sub-layer and the fourth data sub-layer are used to construct a standard navigation electronic map.

[0084] When the target image accuracy is high-precision, the first data layer includes the first data sub-layer and the second data sub-layer, and the second data layer includes the third data sub-layer and the fourth data sub-layer. When the target image accuracy is low-precision, the first data layer includes the third data sub-layer and the fourth data sub-layer, and the second data layer includes the first data sub-layer and the second data sub-layer.

[0085] Based on the update of the first target map data and the second target map data, the preset database can be synchronously maintained so that the map data in the preset database is updated synchronously.

[0086] In this embodiment, by comparing the first marking information with the first target map data, the first update data of the first map is obtained; based on the first update data, the first target map data is updated; based on the lane numbers used to associate different data layers in the preset database, the updated first target map data is compared with the second target map data to obtain the second update data of the second map; based on the second update data, the second target map data is updated. In this way, when the map data related to the target traffic sign in the preset database is stored in layers, it is convenient to extract and convert the map data of different road accuracy levels, and thus, through two comparison processes, the integrated update of the map data used to construct different road accuracy levels can be achieved.

[0087] Optionally, the image accuracy includes a first accuracy and a second accuracy, and the first accuracy is greater than the second accuracy. When the target image accuracy is the first accuracy, the relationship information includes the lane information to which the target traffic sign belongs, and the first update data includes the physical geometric attribute information and semantic information of the target traffic sign on the first target lane obtained by comparing based on the lane information. The first target lane is the lane where the target traffic sign has changed.

[0088] Updating the first target map data based on the first update data includes:

[0089] Updating the physical geometric attribute information and semantic information corresponding to the lane number of the first target lane in the first target map data based on the physical geometric attribute information and semantic information of the target traffic sign on the first target lane.

[0090] In this embodiment, the image data is high-precision data, such as data obtained by point cloud panoramic acquisition. When the image data is high-precision data, the first marking information of the target traffic sign can be obtained more accurately based on this image data, including the physical geometric attribute information, semantic information, and relationship information with the attached road network of the target traffic sign. The relationship information may include the lane information bound to the target traffic sign, such as the lane center line and lane number.

[0091] Based on this lane information, comparing the first marking information with the first target map data to obtain lane change information. The lane change information includes the first target lane. The first target lane is the lane where the target traffic sign has changed, such as a lane with a newly added guiding arrow, or a lane where the guiding arrow changes from straight to left. The lane change information may also include the physical geometric attribute information and semantic information of the target traffic sign on the first target lane. Among them, the physical geometric attribute information and semantic information of the target traffic sign on the first target lane are the physical geometric attribute information and semantic information of the target traffic sign identified based on the image data.

[0092] Correspondingly, the physical geometric attribute information and semantic information corresponding to the lane number of the first target lane in the first target map data can be updated based on the physical geometric attribute information and semantic information of the target traffic sign on the first target lane.

[0093] For example, when the first target lane is a lane with a newly added guiding arrow, the lane number of the first target lane can be added to the first target map data, and the map data associated with the lane number is added. The map data includes the physical geometric attribute information and semantic information of the target traffic sign on the first target lane.

[0094] For another example, when the first target lane is a lane where the guiding arrow changes from going straight to turning left, the physical and geometric attribute information and semantic information of the lane number corresponding to the first target lane in the first target map data can be modified based on the physical and geometric attribute information and semantic information of the target traffic sign on the first target lane.

[0095] The following is a detailed description of the process for updating map data related to guiding arrows based on high-precision data.

[0096] Figure 2 It is a schematic diagram of the process for updating map data related to target traffic signs based on high-precision data. As Figure 2 shown, for the push of high-precision data, due to its high precision, when updating map data related to target traffic signs, the physical and geometric attribute information and semantic information of the guiding arrow on the lane, which is an independently stored object, can be updated first to synchronously maintain map data of different road precision levels.

[0097] Specifically, the pushed image data can be recognized to obtain the first marking information of the guiding arrow on the ground road. The first marking information can record the physical and geometric attribute information, semantic information, and lane information (including the lane center line) of the guiding arrow on the lane. By associating and comparing the first marking information with the lane-level map arrow data in the preset database, the physical and geometric attribute information and semantic information of the guiding arrow on the lane in the lane-level map arrow data can be automatically updated based on the first update data obtained from the comparison.

[0098] After that, specific lane change information is obtained and mapped to the standard navigation map arrow data through the lane number for differential comparison, and the standard navigation map arrow data is automatically updated based on the second update information obtained from the comparison.

[0099] The lane-level map arrow data and the standard navigation map arrow data are updated, and the preset database is synchronously maintained so that the map data in the preset database is updated synchronously.

[0100] In this embodiment, when the image data is high-precision data, by first updating the physical and geometric attribute information and semantic information of the guiding arrow on the lane, which is an independently stored object, the map data of different road precision levels can be synchronously maintained, ensuring the timeliness of the update of map data of different road precision levels.

[0101] Optionally, the image precision includes a first precision and a second precision, where the first precision is greater than the second precision. When the target image precision is the second precision, the relationship information includes the road information to which the target traffic sign is attached; the first update data includes the position information of the target traffic sign on the target road obtained by comparing based on the road information, and the target road is the road where the target traffic sign has changed.

[0102] Updating the first target map data based on the first update data includes:

[0103] Updating the map data corresponding to the road serial number of the target road in the first target map data based on the position information of the target traffic sign on the target road.

[0104] In this embodiment, the image data is low-precision data, such as the data collected by a collection vehicle. When the image data is low-precision data, since its confidence level is relatively low compared to high-precision data, therefore, when updating the map data related to the target traffic sign, the map data of the road, which is the object of aggregated storage, can be updated first to ensure the timeliness of map data update.

[0105] Among them, the attributes of the road, which is the object of aggregated storage, can be obtained by aggregating the attributes of the lanes, which are the objects of independent storage. In implementation, the information of each lane in the road can be aggregated together according to the lane serial number to form a record, and this record can be the association information between the lane and the road. The preset database can store this record so that the preset database can represent the map data of different data structures through a data model, thereby constructing maps of different road precision levels. At the same time, this data model can also include the map data representing the road for constructing a standard navigation electronic map.

[0106] The first marking information of the target traffic sign can be obtained, including the physical geometric attribute information, semantic information of the target traffic sign, and the relationship information with the attached road network. This relationship information can include the road information bound by the target traffic sign, such as road lines and road serial numbers. The physical geometric attribute information of the target traffic sign can include the position information of the target traffic sign on the road.

[0107] Based on this road information, comparing the first marking information with the first target map data to obtain road change information. This road change information can include the target road, where the target road is the road where the target traffic sign has changed. This road change information can also include the position information of the target traffic sign on the target road, that is, at which position on the target road the target traffic sign has changed.

[0108] Correspondingly, based on the position information of the target traffic signs on the target road, the map data corresponding to the road number of the target road in the first target map data can be updated. For example, at the left position on Road A, if the guiding arrow changes from turning left to going straight, the relevant data of the target traffic signs at the left position on Road A in the first target map data can be updated to update the physical geometric attribute information and semantic information of the target traffic signs at the left position on Road A.

[0109] In this embodiment, when the image data is low-precision data, by first updating the map data of the object aggregated and stored, that is, the road, the timeliness of the map data update is guaranteed.

[0110] Optionally, the target road includes at least one lane, and the attribute information of the target traffic signs on the target road is aggregated from the attribute information of the target traffic signs on the at least one lane. The attribute information includes physical geometric attribute information and semantic information;

[0111] The comparison of the updated first target map data with the second target map data related to the target traffic signs based on the lane numbers used to associate different data layers in the preset database to obtain the second update data of the second map includes:

[0112] Comparing the map data corresponding to the road number of the target road in the updated first target map data with the map data corresponding to the lane numbers of the at least one lane in the second target map data to obtain the physical geometric attribute information and semantic information of the target traffic signs on the second target lane. The second target lane is the lane where the target traffic signs change. The second update data includes the physical geometric attribute information and semantic information of the target traffic signs on the second target lane.

[0113] In this embodiment, the attributes of the object aggregated and stored, that is, the road, can be aggregated from the attributes of the independently stored objects, that is, the lanes. In implementation, according to the lane numbers, the information of each lane in the road can be aggregated together to form a record. This record can be the association information between the lane and the road, and this record can be stored in the preset database.

[0114] In this way, based on the map data corresponding to the road number of the target road in the first target map data and the map data corresponding to the lane numbers of at least one lane of the target road in the second target map data, comparison can be made to obtain the physical geometric attribute information and semantic information of the target traffic signs on the second target lane, where the second target lane is the lane where the target traffic signs have changed. Thus, through the road change information and the lane numbers of the associated roads and lanes, the lane change information can be obtained to further update the map data of the high-precision electronic map and the lane-level electronic map.

[0115] Optionally, the updating of the second target map data based on the second update data includes:

[0116] When the accuracy of the target image reaches the preset confidence accuracy, determining an update operation matching the accuracy of the target image;

[0117] Based on the second update data, updating the second target map data according to the update operation.

[0118] In this embodiment, the map data of the lane-level electronic map and the high-precision electronic map can be maintained asynchronously. The asynchronous method may refer to selecting whether to update the map data of the lane-level electronic map and the high-precision electronic map according to the accuracy confidence strategy.

[0119] Specifically, when the accuracy of the target image reaches the preset confidence accuracy, an update operation matching the accuracy of the target image can be determined. The accuracy confidence strategy may include the preset confidence accuracy, which can be set according to the actual situation. For example, when the accuracy of the target image is low accuracy, the low-accuracy level can be further divided into sub-levels, such as 10 sub-levels. When the accuracy of the target image is greater than or equal to a certain sub-level, that is, when the preset confidence accuracy is reached, the map data of the high-precision electronic map and the lane-level electronic map can be updated; otherwise, the map data of the high-precision electronic map and the lane-level electronic map is not updated.

[0120] Among them, the update operation can match the accuracy of the target image. The higher the accuracy of the target image, the larger the range of the update operation can be. For example, when the accuracy of the target image is the first sub-level (the first sub-level is the highest sub-level), all update operations, such as adding, modifying, deleting, etc., can be allowed. Another example is that when the accuracy of the target image is the third sub-level, partial update operations, such as adding and modifying, can be allowed, while the deletion operation is not allowed.

[0121] Correspondingly, based on the second update data, the second target map data is updated according to the update operation. For example, when the second update data indicates that map data needs to be added and the update operation allows addition, the second target map data can be updated based on the second update data. Another example is that when the second update data indicates that map data needs to be deleted but the update operation does not allow deletion, the second target map data is not updated at this time. In this way, by maintaining the map data of the lane-level electronic map and the high-precision electronic map asynchronously, the update accuracy of the map data can be controlled.

[0122] The following details the process of updating the map data related to the guiding arrow based on the low-precision data.

[0123] Figure 3 is a schematic diagram of the process of updating the map data related to the target traffic sign based on the low-precision data. For example, Figure 3 As shown, for the push of the low-precision data, since its accuracy is relatively low, when updating the map data related to the target traffic sign, the map data of the road, which is the aggregated storage object, can be updated first.

[0124] Specifically, the pushed image data can be recognized to obtain the first marking information of the guiding arrow on the ground road. The first marking information can record the physical and geometric attribute information, semantic information, and road information (including road lines and road numbers) of the guiding arrow on the road. By associating, that is, comparing the first marking information with the standard navigation map arrow data in the preset database, the physical and geometric attribute information and semantic information of the guiding arrow at the corresponding position on the road in the standard navigation map arrow data are automatically updated based on the obtained first update data.

[0125] After that, based on the road change information and the lane number, the standard navigation map arrow data and the lane-level map arrow data are compared to obtain the lane change information. Based on the target image accuracy, a constraint judgment is made according to the accuracy confidence strategy to obtain the update operation. When the update of the lane-level map arrow data is allowed, based on the second update information, the lane-level map arrow data is automatically updated according to the update operation.

[0126] After updating the standard navigation map arrow data, the preset database is synchronously maintained. After updating the lane-level map arrow data, the preset database is synchronously maintained, so that the map data in the preset database is updated.

[0127] Optionally, before the step S101, the method further includes:

[0128] Construct a preset database, which includes a first data sub-layer, a second data sub-layer, a third data sub-layer, and a fourth data sub-layer. The first data sub-layer is used to store the physical and geometric attribute information of the target traffic signs on the lane. The second data sub-layer is used to store the semantic information of the target traffic signs on the lane. The third data sub-layer is used to store the association relationship information between the lane and the road through the lane number. The fourth data sub-layer is used to store the attribute information of the target traffic signs on the road;

[0129] Among them, the first data sub-layer and the second data sub-layer are used to construct a map of the first road accuracy level. The third data sub-layer and the fourth data sub-layer are used to construct a map of the second road accuracy level. The first road accuracy level is higher than the second road accuracy level. The third data sub-layer and the fourth data sub-layer are used to construct a map of the second road accuracy level. The first road accuracy level is higher than the second road accuracy level.

[0130] In this embodiment, before updating the map data, it is necessary to construct a preset database, which can use a data model to store map data that can construct maps of different road accuracy levels. This data model can store map data related to target traffic signs in a hierarchical modeling manner to meet data extraction and conversion at different data levels, so that map data for constructing maps of different road accuracy levels can be stored integrally, and then the integrated update of map data of different road accuracy levels can be realized based on this preset database.

[0131] This data model needs to meet entity layering, attribute layering, and relationship layering. Among them, entity layering refers to the physical and geometric attributes and semantic layering of entities. For example, this preset database can include a first data sub-layer and a second data sub-layer. The first data sub-layer is used to store the physical and geometric attribute information of the target traffic signs on the lane. The second data sub-layer is used to store the semantic information of the target traffic signs on the lane. That is, entity layering is realized through the first data sub-layer and the second data sub-layer. The first data sub-layer and the second data sub-layer can store map data of high-precision electronic maps and lane-level electronic maps.

[0132] Attribute layering refers to the layering of the minimum semantic attributes and the aggregated semantic attributes. The minimum semantic attributes can refer to the attributes of the lane, and the aggregated semantic attributes can refer to the attributes of the road. For example, this preset database can include a second data sub-layer and a third data sub-layer. The second data sub-layer is used to store the semantic information of the target traffic signs on the lane. The third data sub-layer is used to store the association relationship information between the lane and the road through the lane number. That is, attribute layering is realized through the second data sub-layer and the third data sub-layer.

[0133] The relationship stratification may include meta-semantic content and abstract semantic stratification. The meta-semantic content may refer to the semantic information of the entity itself, while the abstract semantics may refer to the semantics derived from the relationship between the target traffic sign and other entities. The second data sub-layer can be further divided into two sub-layers, which are respectively used to store the meta-semantic content and abstract semantics of the target traffic sign, so as to realize the stratification of the meta-semantic content and abstract semantics.

[0134] The preset database may further include a fourth data sub-layer, which is used to store the attribute information of the target traffic signs on the road. The attribute information may include the location information, semantic information, etc. of the target traffic sign. The third data sub-layer and the fourth data sub-layer can store the map data of the standard navigation electronic map.

[0135] In addition, the first data sub-layer, the second data sub-layer, the third data sub-layer, and the fourth data sub-layer can all store the relationship information between the target traffic sign and the attached road network to identify the geographical location of the target traffic sign.

[0136] Second Embodiment

[0137] As Figure 4 shown, the present disclosure provides a map data updating device 400, including:

[0138] An obtaining module 401, configured to obtain image data including ground roads and the target image accuracy of the image data;

[0139] An image recognition module 402, configured to perform image recognition processing on the image data to obtain first marking information of the target traffic signs on the ground road;

[0140] An updating module 403, configured to update the map data related to the target traffic sign according to a target updating method corresponding to the target image accuracy based on the first marking information. The map data can construct at least two maps with different road accuracy levels.

[0141] Optionally, the image recognition module 402 is specifically configured to:

[0142] Perform image detection on the image data to obtain the physical geometric attribute information of the traffic signs on the ground road;

[0143] Based on the pre-obtained traffic sign coding information and the physical geometric attribute information, perform semantic parsing on the traffic signs on the ground road to obtain semantic information with the semantics of the target traffic sign;

[0144] Obtain the relationship information between the target traffic sign and the attached road network, where the attached road network is the lane and / or road bound to the target traffic sign;

[0145] Among them, the first marker information includes the physical geometric attribute information, semantic information, and the relationship information of the target traffic sign.

[0146] Optionally, the update module 403 includes:

[0147] A first comparison unit, configured to compare the first marker information with first target map data to obtain first update data of a first map, where the road accuracy level of the first map corresponds to the target image accuracy, the first target map data is the map data used to construct the first map, the first target map data is located in a first data layer of a preset database, and the preset database is used to store the map data;

[0148] A first update unit, configured to update the first target map data based on the first update data;

[0149] A second comparison unit, configured to compare the updated first target map data with second target map data based on a lane serial number used to associate different data layers in the preset database to obtain second update data of a second map, where the second target map data is the map data used to construct the second map, the second target map data is located in a second data layer of the preset database, and the road accuracy level of the second map is different from that of the first map;

[0150] A second update unit, configured to update the second target map data based on the second update data.

[0151] Optionally, the image accuracy includes a first accuracy and a second accuracy, the first accuracy is greater than the second accuracy. When the target image accuracy is the first accuracy, the relationship information includes lane information to which the target traffic sign is attached, the first update data includes the physical geometric attribute information and semantic information of the target traffic sign on a first target lane obtained by comparison based on the lane information, and the first target lane is the lane where the target traffic sign has changed;

[0152] The first update unit is specifically configured to update the physical geometric attribute information and semantic information corresponding to the lane serial number of the first target lane in the first target map data based on the physical geometric attribute information and semantic information of the target traffic sign on the first target lane.

[0153] Optionally, the image precision includes a first precision and a second precision, where the first precision is greater than the second precision. When the target image precision is the second precision, the relationship information includes the road information to which the target traffic sign belongs; the first update data includes the position information of the target traffic sign on the target road obtained by comparing based on the road information, and the target road is the road where the target traffic sign has changed;

[0154] The first update unit is specifically configured to update the map data corresponding to the road serial number of the target road in the first target map data based on the position information of the target traffic sign on the target road.

[0155] Optionally, the target road includes at least one lane, and the attribute information of the target traffic sign on the target road is aggregated from the attribute information of the target traffic sign on the at least one lane. The attribute information includes physical and geometric attribute information and semantic information;

[0156] The second comparison unit is specifically configured to compare the map data corresponding to the road serial number of the target road in the updated first target map data with the map data corresponding to the lane serial number of the at least one lane in the second target map data to obtain the physical and geometric attribute information and semantic information of the target traffic sign on the second target lane. The second target lane is the lane where the target traffic sign has changed, and the second update data includes the physical and geometric attribute information and semantic information of the target traffic sign on the second target lane.

[0157] Optionally, the second update unit is specifically configured to:

[0158] When the target image precision reaches the preset confidence precision, determine the update operation matching the target image precision;

[0159] Based on the second update data, update the second target map data according to the update operation.

[0160] Optionally, the device further includes:

[0161] A construction module for constructing a preset database, which includes a first data sub-layer, a second data sub-layer, a third data sub-layer, and a fourth data sub-layer. The first data sub-layer is used to store the physical and geometric attribute information of the target traffic sign on the lane, the second data sub-layer is used to store the semantic information of the target traffic sign on the lane, the third data sub-layer is used to store the association relationship information between the lane and the road through the lane serial number, and the fourth data sub-layer is used to store the attribute information of the target traffic sign on the road;

[0162] Wherein, the first data sub-layer and the second data sub-layer are used to construct a map of the first road accuracy level, the third data sub-layer and the fourth data sub-layer are used to construct a map of the second road accuracy level, and the first road accuracy level is higher than the second road accuracy level.

[0163] The map data update device 400 provided by the present disclosure can implement all the processes implemented by the embodiments of the map data update method and can achieve the same beneficial effects. To avoid repetition, details are not described herein again.

[0164] In the technical solution of the present disclosure, the collection, storage, use, processing, transmission, provision, and disclosure of user personal information comply with the provisions of relevant laws and regulations and do not violate public order and good customs.

[0165] According to the embodiments of the present disclosure, the present disclosure also provides an electronic device, a readable storage medium, and a computer program product.

[0166] Figure 5 A schematic block diagram of an exemplary electronic device that can be used to implement the embodiments of the present disclosure is shown. The electronic device is intended to represent various forms of digital computers, such as, a laptop computer, a desktop computer, a workbench, a personal digital assistant, a server, a blade server, a mainframe computer, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as, a personal digital processor, a cellular phone, a smart phone, a wearable device, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are only examples and are not intended to limit the implementation of the present disclosure described and / or claimed herein.

[0167] As Figure 5 shown, the device 500 includes a computing unit 501, which can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 502 or the computer program loaded from the storage unit 508 into the random access memory (RAM) 503. In the RAM 503, various programs and data required for the operation of the device 500 can also be stored. The computing unit 501, the ROM 502, and the RAM 503 are connected to each other through a bus 504. The input / output (I / O) interface 505 is also connected to the bus 504.

[0168] Multiple components in device 500 are connected to I / O interface 505, including: input unit 506, such as a keyboard, mouse, etc.; output unit 507, such as various types of displays, speakers, etc.; storage unit 508, such as a disk, optical disc, etc.; and communication unit 509, such as a network card, modem, wireless communication transceiver, etc. Communication unit 509 allows device 500 to exchange information / data with other devices via a computer network such as the Internet and / or various telecommunication networks.

[0169] Computing unit 501 can be various general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of computing unit 501 include but are not limited to a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. Computing unit 501 executes the various methods and processes described above, such as the map data update method. For example, in some embodiments, the map data update method can be implemented as a computer software program, which is tangibly contained in a machine-readable medium, such as storage unit 508. In some embodiments, part or all of the computer program can be loaded and / or installed onto device 500 via ROM 502 and / or communication unit 509. When the computer program is loaded into RAM 503 and executed by computing unit 501, one or more steps of the map data update method described above can be executed. Alternatively, in other embodiments, computing unit 501 can be configured to execute the map data update method in any other suitable manner (e.g., by means of firmware).

[0170] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SOCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments can include: implemented in one or more computer programs, the one or more computer programs can be executed and / or interpreted on a programmable system including at least one programmable processor, the programmable processor can be a special or general-purpose programmable processor, can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit the data and instructions to the storage system, the at least one input device, and the at least one output device.

[0171] The program code for implementing the methods of the present disclosure may be written in any combination of one or more programming languages. These program codes may be provided to a processor or controller of a general purpose computer, a special purpose computer, or other programmable data processing device, such that the program codes, when executed by the processor or controller, cause the functions / operations specified in the flowchart and / or block diagram to be implemented. The program code may be executed entirely on the machine, partially on the machine, as a stand-alone software package partially on the machine and partially on a remote machine, or entirely on a remote machine or server.

[0172] In the context of the present disclosure, a machine-readable medium may be a tangible medium that can contain or store a program for use by or in connection with an instruction execution system, apparatus, or device. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of a machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0173] In order to provide interaction with a user, the systems and techniques described herein may be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or a trackball) by which the user can provide input to the computer. Other kinds of devices may also be used to provide interaction with the user; for example, the feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form (including acoustic input, voice input, or tactile input).

[0174] The systems and techniques described herein can be implemented in a computing system including backend components (e.g., as a data server), or a computing system including middleware components (e.g., an application server), or a computing system including frontend components (e.g., a user computer having a graphical user interface or a web browser through which a user can interact with an implementation of the systems and techniques described herein), or a computing system including any combination of such backend components, middleware components, or frontend components. The components of the system can be interconnected to each other by digital data communication in any form or medium (e.g., a communication network). Examples of communication networks include: local area network (LAN), wide area network (WAN), and the Internet.

[0175] A computer system can include a client and a server. The client and the server are generally far from each other and typically interact through a communication network. The client - server relationship is created by computer programs running on the respective computers and having a client - server relationship with each other. The server can be a cloud server, a server of a distributed system, or a server incorporating blockchain.

[0176] It should be understood that various forms of the processes shown above can be used, with steps reordered, added, or deleted. For example, the steps recited in this disclosure can be executed in parallel, sequentially, or in a different order, as long as the desired results of the technical solutions disclosed in this disclosure can be achieved, and this is not limited herein.

[0177] The above - described specific embodiments do not limit the protection scope of this disclosure. Those skilled in the art should understand that various modifications, combinations, sub - combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this disclosure shall be included within the protection scope of this disclosure.

Claims

1. A method for updating map data, comprising: Obtaining image data including ground roads and a target image accuracy of the image data; Performing image recognition processing on the image data to obtain first marker information of target traffic signs on the ground roads; Comparing the first marker information with first target map data to obtain first update data of a first map, wherein the road accuracy level of the first map corresponds to the target image accuracy, the first target map data is the map data related to the target traffic signs and is used to construct the first map, the first target map data is located in a first data layer in a preset database, the preset database is used to store the map data, and the map data can construct at least two maps with different road accuracy levels; Updating the first target map data based on the first update data; Based on the lane numbers used to associate different data layers in the preset database, comparing the updated first target map data with second target map data to obtain second update data of a second map, wherein the second target map data is the map data used to construct the second map, the second target map data is located in a second data layer in the preset database, and the road accuracy level of the second map is different from that of the first map; Updating the second target map data based on the second update data.

2. The method according to claim 1, wherein The performing image recognition processing on the image data to obtain first marker information of target traffic signs on the ground roads includes: Performing image detection on the image data to obtain physical and geometric attribute information of traffic signs on the ground roads; Based on pre-obtained traffic sign coding information and the physical and geometric attribute information, performing semantic parsing on the traffic signs on the ground roads to obtain semantic information with a semantic attribute of the target traffic signs; Obtaining relationship information between the target traffic signs and the attached road network, where the attached road network is the lane and / or road bound by the target traffic signs; Wherein, the first marker information includes the physical and geometric attribute information, semantic information, and the relationship information of the target traffic signs.

3. The method according to claim 2, wherein The image accuracy includes a first accuracy and a second accuracy, the first accuracy is greater than the second accuracy. When the target image accuracy is the first accuracy, the relationship information includes the lane information attached to the target traffic signs, and the first update data includes the physical and geometric attribute information and semantic information of the target traffic signs on a first target lane obtained by comparison based on the lane information, and the first target lane is the lane where the target traffic signs have changed; The updating the first target map data based on the first update data includes: Based on the physical and geometric attribute information and semantic information of the target traffic signs on the first target lane, updating the physical and geometric attribute information and semantic information corresponding to the lane number of the first target lane in the first target map data.

4. The method according to claim 2, wherein The image accuracy includes a first accuracy and a second accuracy, where the first accuracy is greater than the second accuracy. When the target image accuracy is the second accuracy, the relationship information includes the road information to which the target traffic sign is attached; the first update data includes the position information of the target traffic sign on the target road obtained by comparing based on the road information, and the target road is the road where the target traffic sign has changed; The updating of the first target map data based on the first update data includes: Updating the map data corresponding to the road number of the target road in the first target map data based on the position information of the target traffic sign on the target road.

5. The method according to claim 4, wherein The target road includes at least one lane, and the attribute information of the target traffic sign on the target road is aggregated from the attribute information of the target traffic sign on the at least one lane. The attribute information includes physical and geometric attribute information and semantic information; The comparing the updated first target map data with the second target map data related to the target traffic sign based on the lane numbers used to associate different data layers in the preset database to obtain the second update data of the second map includes: Comparing the map data corresponding to the road number of the target road in the updated first target map data with the map data corresponding to the lane numbers of the at least one lane in the second target map data to obtain the physical and geometric attribute information and semantic information of the target traffic sign on the second target lane. The second target lane is the lane where the target traffic sign has changed, and the second update data includes the physical and geometric attribute information and semantic information of the target traffic sign on the second target lane.

6. The method according to claim 4, wherein The updating of the second target map data based on the second update data includes: Determining an update operation matching the target image accuracy when the target image accuracy reaches the preset confidence accuracy; Updating the second target map data according to the update operation based on the second update data.

7. According to the method described in claim 1, before obtaining the image data of the ground road and the target image accuracy of the image data, it further includes: Constructing a preset database, which includes a first data sub-layer, a second data sub-layer, a third data sub-layer, and a fourth data sub-layer. The first data sub-layer is used to store the physical and geometric attribute information of the target traffic sign on the lane, the second data sub-layer is used to store the semantic information of the target traffic sign on the lane, the third data sub-layer is used to store the association relationship information between the lane and the road through the lane number, and the fourth data sub-layer is used to store the attribute information of the target traffic sign on the road; Among them, the first data sub-layer and the second data sub-layer are used to construct a map of the first road accuracy level, the third data sub-layer and the fourth data sub-layer are used to construct a map of the second road accuracy level, the first road accuracy level is greater than the second road accuracy level, the third data sub-layer and the fourth data sub-layer are used to construct a map of the second road accuracy level, and the first road accuracy level is greater than the second road accuracy level.

8. A map data update device, comprising: An acquisition module, configured to acquire image data including ground roads and the target image accuracy of the image data; An image recognition module, configured to perform image recognition processing on the image data to obtain first marker information of target traffic signs on the ground roads; An update module, configured to update map data related to the target traffic signs in a target update manner corresponding to the target image accuracy based on the first marker information, where the map data can construct at least two maps of different road accuracy levels; The update module includes: A first comparison unit, configured to compare the first marker information with first target map data to obtain first update data of a first map, where the road accuracy level of the first map corresponds to the target image accuracy, the first target map data is data in the map data used to construct the first map, the first target map data is located in a first data layer in a preset database, and the preset database is used to store the map data; A first update unit, configured to update the first target map data based on the first update data; A second comparison unit, configured to compare the updated first target map data with second target map data based on lane numbers used to associate different data layers in the preset database to obtain second update data of a second map, where the second target map data is data in the map data used to construct the second map, the second target map data is located in a second data layer in the preset database, and the road accuracy level of the second map is different from that of the first map; A second update unit, configured to update the second target map data based on the second update data.

9. The apparatus according to claim 8, wherein, The image recognition module is specifically configured to: Perform image detection on the image data to obtain physical geometric attribute information of traffic signs on the ground roads; Perform semantic parsing on the traffic signs on the ground roads based on pre-acquired traffic sign coding information and the physical geometric attribute information to obtain semantic information whose semantic attribute is the target traffic sign; Obtain relationship information between the target traffic sign and the attached road network, where the attached road network is the lane and / or road bound by the target traffic sign; Among them, the first marker information includes the physical geometric attribute information, semantic information, and relationship information of the target traffic sign.

10. The device according to claim 9, wherein, The image precision includes a first precision and a second precision, and the first precision is greater than the second precision. When the target image precision is the first precision, the relationship information includes the lane information to which the target traffic sign belongs. The first update data includes the physical geometric attribute information and semantic information of the target traffic sign on the first target lane obtained by comparing based on the lane information. The first target lane is the lane where the target traffic sign has changed; The first update unit is specifically configured to update the physical geometric attribute information and semantic information corresponding to the lane number of the first target lane in the first target map data based on the physical geometric attribute information and semantic information of the target traffic sign on the first target lane.

11. The device according to claim 9, wherein, The image precision includes a first precision and a second precision, and the first precision is greater than the second precision. When the target image precision is the second precision, the relationship information includes the road information to which the target traffic sign belongs; the first update data includes the position information of the target traffic sign on the target road obtained by comparing based on the road information. The target road is the road where the target traffic sign has changed; The first update unit is specifically configured to update the map data corresponding to the road number of the target road in the first target map data based on the position information of the target traffic sign on the target road.

12. The device according to claim 11, wherein, The target road includes at least one lane, and the attribute information of the target traffic sign on the target road is aggregated from the attribute information of the target traffic sign on the at least one lane. The attribute information includes physical geometric attribute information and semantic information; The second comparison unit is specifically configured to compare the map data corresponding to the road number of the target road in the updated first target map data with the map data corresponding to the lane numbers of the at least one lane in the second target map data to obtain the physical geometric attribute information and semantic information of the target traffic sign on the second target lane. The second target lane is the lane where the target traffic sign has changed. The second update data includes the physical geometric attribute information and semantic information of the target traffic sign on the second target lane.

13. The apparatus according to claim 11, wherein, The second update unit is specifically configured to: Determine an update operation matching the target image precision when the target image precision reaches the preset confidence precision; Update the second target map data according to the update operation based on the second update data.

14. The device according to claim 8, further comprising: A building module for building a preset database, the preset database including a first data sub-layer, a second data sub-layer, a third data sub-layer and a fourth data sub-layer, the first data sub-layer being used to store physical geometric attribute information of the target traffic signs on the lane, the second data sub-layer being used to store semantic information of the target traffic signs on the lane, the third data sub-layer being used to store the association relationship information between the lane and the road by lane number, and the fourth data sub-layer being used to store attribute information of the target traffic signs on the road; Wherein, the first data sub-layer and the second data sub-layer are used to build a map of the first road accuracy level, and the third data sub-layer and the fourth data sub-layer are used to build a map of the second road accuracy level, and the first road accuracy level is greater than the second road accuracy level.

15. An electronic device, comprising: At least one processor; And A memory communicatively connected to the at least one processor; wherein, The memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the method according to any one of claims 1-7.

16. A non-transitory computer-readable storage medium storing computer instructions, wherein, The computer instructions are used to cause the computer to execute the method according to any one of claims 1-7.

17. A computer program product, comprising a computer program which, when executed by a processor, implements the method according to any one of claims 1-7.

Images