CN117632978A - Road network update method and equipment - Google Patents

Abstract

This application provides a road network update method and equipment, which can be used in the field of data processing technology. The method includes: obtaining image data collected by the vehicle during driving; generating the driving characteristics of the vehicle based on the image data; matching the driving characteristics with the road characteristics of the road network roads in the road network database; updating the successfully matched roads through the image data The corresponding attribute information of network roads in the road network database. This application updates the attribute information of the road network roads in the road network database through the image data collected while the vehicle is driving. The acquisition and analysis of image data requires low costs, especially now that vehicles are equipped with image acquisition devices, such as driving recorders. In this scenario, there is no need to purchase additional image acquisition equipment, so the cost of updating the road network database can be reduced.

Description

Road network update method and equipment

Technical field

This application relates to the field of data processing technology, and in particular to a road network update method and equipment.

Background technique

In some applications, users are provided with map services such as navigation and route planning. These map services rely on a large amount of road information. Within a designated area, the roads are interconnected and intertwined to form a network road system. Therefore, the road network database emerged as the times require and is used to store a large amount of road information so that applications can call the road network database to implement map services.

In the existing technology, the road information in the road network database needs to be updated regularly, and the update is realized by combining satellite remote sensing technology, GPS (Global Positioning System, Global Positioning System), geographic information system and manual methods.

However, the above solution has the problem of high cost.

Contents of the invention

This application provides a road network update method and equipment to reduce the update cost of the road network database.

In the first aspect, this application provides a road network update method, including:

Obtain image data collected while the vehicle is driving;

Generate driving characteristics of the vehicle based on the image data;

Match the driving characteristics with the road characteristics of the road network roads in the road network database;

The corresponding attribute information of the successfully matched road network road in the road network database is updated through the image data.

Optionally, the road features include at least one of: fenced area, road direction, and road height.

Optionally, generating the driving characteristics of the vehicle based on the image data includes:

The driving characteristics of the vehicle are generated according to the collection information corresponding to the image data. The collection information includes at least one of the following: collection address, collection time, height information and camera posture when collecting images.

Optionally, the road direction includes at least one of the following: the pointing direction from the first position to the second position on the road network, and the three-dimensional vector corresponding to the extension direction of the road network;

The driving characteristics include at least one of the following: driving trajectory, driving direction, driving height and the camera attitude;

Matching the driving characteristics with the road characteristics of the road network roads in the road network database includes:

Perform at least one of the following matching processes: matching between the driving trajectory and the fence area, matching between the driving direction and the pointing direction, matching between the driving height and the road height, and the matching between the camera pose and the three-dimensional vector;

If at least one of the matching processes is successful, it is determined that the driving characteristics and the road characteristics are successfully matched.

Optionally, generating the driving characteristics of the vehicle based on the collection information corresponding to the image data includes:

Generate the driving trajectory of the vehicle according to the collection addresses corresponding to the plurality of image data corresponding to the same vehicle;

Performing matching between the driving trajectory and the fence area includes:

From the driving trajectory, determine the number of collection addresses falling into the fence area;

If the number of collected addresses is greater than or equal to the first threshold, it is determined that the driving trajectory matches the fence area successfully.

Optionally, generating the driving characteristics of the vehicle based on the collection information corresponding to the image data includes:

Determine the first time for the vehicle to arrive at the first position and the second time for the vehicle to arrive at the second position based on the collection address and collection time corresponding to at least two frames of the image data corresponding to the same vehicle;

If the first time is less than the second time, determine the traveling direction to be the direction from the first position to the second position;

If the second time is less than the first time, the traveling direction is determined to be the direction from the second position to the first position.

Optionally, generating the driving characteristics of the vehicle based on the collection information corresponding to the image data includes:

According to the collection information of the image data and the picture information of the image data, determine the shape of the driving route of the vehicle, the driving speed and whether there are obstructions above the vehicle;

If the shape of the driving route is a straight line, the driving speed is constant, and there is no obstruction above the vehicle, then the driving height is determined to be the elevated road height;

Otherwise, the driving height is determined to be an ordinary road height.

Optionally, before corresponding attribute information in the road network database, the road network road that is successfully matched through the image data update also includes:

Filter out at least one of the following image data: image data outside the target time period, image data outside the target area, image data collected outside the target weather state, image data with an occlusion area on the driving road greater than or equal to the second threshold , and image data with a sun azimuth angle outside the preset angle range, where the sun azimuth angle includes the angle between the straight line between the sun and the vehicle and the driving direction of the vehicle.

Optionally, filtering out at least one of the following image data includes:

Determine a sunshine time period based on the collection time and collection address of the image data, and the sunshine time period is associated with the sunrise time and sunset time corresponding to the collection address at the collection time;

Generate the target time period according to the sunshine time period and/or a preset time period before the current time;

Filter out image data whose acquisition time is outside the corresponding target time period.

Optionally, filtering out at least one of the following image data includes:

Determine the corresponding weather state according to the collection time and collection address of the image data;

Filter out image data whose weather state is other than the target weather state, where the target weather state includes a weather state with visibility greater than or equal to a third threshold.

Optionally, filtering out at least one of the following image data includes:

According to the collection time and collection address of the image data, determine the straight line formed by the sun and the vehicle, and the driving direction of the vehicle;

Determine the angle between the straight line and the traveling direction as the solar azimuth angle;

Image data whose sun azimuth angle is outside the preset angle range is filtered out.

In the second aspect, this application provides a road network updating device, including:

The image data acquisition module is used to acquire image data collected while the vehicle is driving;

A driving characteristic generation module, used to generate the driving characteristics of the vehicle according to the image data; a matching module, used to match the driving characteristics with the road characteristics of the road network roads in the road network database;

An update module, configured to update the corresponding attribute information of the successfully matched road network road in the road network database through the image data.

Optionally, the road features include at least one of: fenced area, road direction, and road height.

Optionally, the matching module is also used to:

The driving characteristics of the vehicle are generated according to the collection information corresponding to the image data. The collection information includes at least one of the following: collection address, collection time, height information and camera posture when collecting images.

Optionally, the road direction includes at least one of the following: the pointing direction from the first position to the second position on the road network, and the three-dimensional vector corresponding to the extension direction of the road network;

The driving characteristics include at least one of the following: driving trajectory, driving direction, driving height and the camera attitude;

The matching module is also used to:

Perform at least one of the following matching processes: matching between the driving trajectory and the fence area, matching between the driving direction and the pointing direction, matching between the driving height and the road height, and the matching between the camera pose and the three-dimensional vector;

If at least one of the matching processes is successful, it is determined that the driving characteristics and the road characteristics are successfully matched.

Optionally, the matching module is also used to:

Generate the driving trajectory of the vehicle according to the collection addresses corresponding to the plurality of image data corresponding to the same vehicle;

From the driving trajectory, determine the number of collection addresses falling into the fence area;

If the number of collected addresses is greater than or equal to the first threshold, it is determined that the driving trajectory matches the fence area successfully.

Optionally, the matching module is also used to:

Determine the first time for the vehicle to arrive at the first position and the second time for the vehicle to arrive at the second position based on the collection address and collection time corresponding to at least two frames of the image data corresponding to the same vehicle;

If the first time is less than the second time, determine the traveling direction to be the direction from the first position to the second position;

If the second time is less than the first time, the traveling direction is determined to be the direction from the second position to the first position.

Optionally, the matching module is also used to:

According to the collection information of the image data and the picture information of the image data, determine the shape of the driving route of the vehicle, the driving speed and whether there are obstructions above the vehicle;

If the shape of the driving route is a straight line, the driving speed is constant, and there is no obstruction above the vehicle, then the driving height is determined to be the elevated road height;

Otherwise, the driving height is determined to be an ordinary road height.

Optionally, the device also includes:

A filtering module configured to filter out at least one of the following image data before the corresponding attribute information of the successfully matched road network road in the road network database through the image data update: image data outside the target time period, Image data outside the target area, image data collected outside the target weather state, image data with an occlusion area on the driving road greater than or equal to the second threshold, and image data with the sun azimuth outside the preset angle range, the The sun azimuth angle includes the angle between the straight line between the sun and the vehicle and the traveling direction of the vehicle.

Optionally, the filtering module is also used to:

Determine a sunshine time period based on the collection time and collection address of the image data, and the sunshine time period is associated with the sunrise time and sunset time corresponding to the collection address at the collection time;

Generate the target time period according to the sunshine time period and/or a preset time period before the current time;

Filter out image data whose acquisition time is outside the corresponding target time period.

Optionally, the filtering module is also used to:

Determine the corresponding weather state according to the collection time and collection address of the image data;

Filter out image data whose weather state is other than the target weather state, where the target weather state includes a weather state with visibility greater than or equal to a third threshold.

Optionally, the filtering module is also used to:

According to the collection time and collection address of the image data, determine the straight line formed by the sun and the vehicle, and the driving direction of the vehicle;

Determine the angle between the straight line and the traveling direction as the solar azimuth angle;

Image data whose sun azimuth angle is outside the preset angle range is filtered out.

In a third aspect, this application provides an electronic device, including a memory and at least one processor;

Among them, the memory stores computer execution instructions;

At least one processor executes computer execution instructions stored in the memory, so that the electronic device implements the method of the first aspect.

In a fourth aspect, the present application provides a computer-readable storage medium. Computer-executable instructions are stored in the computer-readable storage medium. When executed by a processor, the computer-executable instructions are used to implement the method as described in the first aspect. .

In a fifth aspect, this application provides a computer program product for implementing the method of the first aspect.

The road network update method and equipment provided by the embodiments of the present application can obtain image data collected by the vehicle during driving; generate the driving characteristics of the vehicle based on the image data; and combine the driving characteristics with the road characteristics of the road network roads in the road network database. Matching; update the corresponding attribute information of successfully matched road network roads in the road network database through image data. It can be seen that the embodiment of the present application updates the attribute information of the road network roads in the road network database through the image data collected while the vehicle is driving. Among them, the cost of collecting and analyzing image data is relatively low, especially now that vehicles are equipped with image acquisition devices, such as driving recorders. In this scenario, the embodiment of the present application can update the attribute information of the road network roads in the road network database without purchasing additional image acquisition equipment. Therefore, the cost of updating the road network database can be effectively reduced.

Description of drawings

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

Figure 1 is a schematic diagram of road network database update provided by the prior art;

Figure 2 is a step flow chart of a road network update method provided by an embodiment of the present application;

Figure 3 is a schematic diagram of the relationship between a driving trajectory and a fenced area of a road network provided by an embodiment of the present application;

Figure 4 is a schematic diagram of a driving route shape provided by an embodiment of the present application;

Figure 5 is a schematic diagram of the relationship between driving speed and image data provided by an embodiment of the present application;

Figure 6 is a schematic diagram of the blocking area on a driving road provided by an embodiment of the present application;

Figure 7 is a schematic diagram of a solar direction angle provided by an embodiment of the present application;

Figure 8 is a detailed flow chart of a road network update process provided by an embodiment of the present application;

Figure 9 is a structural block diagram of a road network update device provided by an embodiment of the present application;

Figure 10 is a structural block diagram of an electronic device provided by an embodiment of the present application.

Through the above-mentioned drawings, clear embodiments of the present application have been shown, which will be described in more detail below. These drawings and text descriptions are not intended to limit the scope of the present application's concepts in any way, but are intended to illustrate the application's concepts for those skilled in the art with reference to specific embodiments.

Detailed ways

Exemplary embodiments will be described in detail herein, examples of which are illustrated in the accompanying drawings. When the following description refers to the drawings, the same numbers in different drawings refer to the same or similar elements unless otherwise indicated. The implementations described in the following exemplary embodiments do not represent all implementations consistent with this application. Rather, they are merely examples of apparatus and methods consistent with aspects of the application as detailed in the appended claims.

This application is suitable for the update scenario of road network database. Figure 1 is a schematic diagram of road network database update provided by the prior art. Referring to Figure 1, updates can be achieved through geographic information systems, GPS, satellite remote sensing technology and manual methods. However, this method requires a lot of resources and is costly.

In order to solve the above problems, this application updates the attribute information of the road network roads in the road network database through the image data collected while the vehicle is driving. The acquisition and analysis of image data requires low costs, especially now that vehicles are equipped with image acquisition devices, such as driving recorders. In this scenario, there is no need to purchase additional image acquisition equipment, so the cost of updating the road network database can be reduced. In addition, the analysis and matching of image data can be realized by computers, and the cost is low.

It should be noted that the above image data is collected for the driving road, from which the attribute information of the driving road can be analyzed. Therefore, the image data can be matched with the road network roads in the road network database. When the matching is successful, the driving road of the representative vehicle is the road network road in the road network database. At this time, the attribute information of the road network roads can be updated through image data.

In addition, the image data in this application are collected from vehicles driving on the road, realizing crowdsourcing updates. Since vehicles are spread across various roads, image data for each road can be obtained, which helps expand the update range.

Moreover, image processing technology is already very mature. Therefore, the process of matching and calculating image data takes a short time, which can effectively increase the update speed.

The technical solution of the present application and how the technical solution of the present application solves the above technical problems will be described in detail below with specific embodiments. The following 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 the present application will be described below with reference to the accompanying drawings.

Figure 2 is a step flow chart of a road network update method provided by an embodiment of the present application. Referring to Figure 2, the road network update method of this application may include:

S201: Obtain image data collected while the vehicle is driving.

The vehicle here can be at least any vehicle, and the image data corresponds to one or more frames of images, usually corresponding to multiple frames of images. In the embodiment of the present application, a multi-frame image of a vehicle is taken as an example to illustrate. When the current driving road of the vehicle is a road network road in the road network database, the road network road is updated through the multi-frame image. Attribute information in the road network database.

It can be understood that the image data may include: picture information and collection information. The picture information includes collected environmental pictures, such as road pictures, sky pictures, some vehicle pictures, etc. The collection information includes some relevant information when collecting the image data, such as collection time, collection location, elevation information, camera pose, etc. Among them, the camera pose is the pose when the image acquisition device collects image data, and is used to represent the shooting direction of the image acquisition device, and the shooting direction can be represented by a three-dimensional vector.

The above image data is collected by an image acquisition device installed on the vehicle. The image acquisition device can be any device capable of image acquisition. In order to improve the update accuracy as much as possible, a high-definition image acquisition device can be used. In the embodiment of the present application, a DVR (Digital Video Recorder) with a corresponding interface reserved can be used to collect image data, but the embodiment of the present application does not limit the type of the image collection device.

In some implementations, the vehicle may collect image data when receiving a road network update task. In other words, the vehicle in S201 is the vehicle that received the road network update task. The vehicle can upload the collected image data to the image database to obtain the image data from the image database when executing S201.

The above road network update task is a task initiated for at least one or more road network roads in the road network database. When initiating the road network update task, you can specify a road network road to update the attribute information of the specified road network road, and/or you can specify a target area to update the attribute information of all road network roads in the target area. , and/or, a preset time period before the current time can be specified to update the attribute information of the road network roads through the image data within the preset time period, and/or, the number of execution times of the road network update task can be specified, Execute the road network update task once or multiple times.

The above-mentioned target area may be an administrative area, for example, it may be a city, province, county, etc. The embodiment of this application does not limit the target area.

In practical applications, there may be a lag in the time when image data is entered into the image database. In this way, when we perform the road network update task, the required image data has not yet been entered into the image database. In this scenario, it will cause the update Bugs or even missing updates. In order to solve this problem, embodiments of the present application can execute the road network update task multiple times to reduce update omissions caused by image data lag.

In some implementations, when a road network update task is initiated, it may be temporarily stored in the task database for execution one by one.

S202: Generate the driving characteristics of the vehicle based on the image data.

Among them, the driving characteristics are used to represent the characteristics of the vehicle during driving. According to the driving characteristics, the driving road of the vehicle can be determined. Different driving roads correspond to different driving characteristics. Since the image data is collected while the vehicle is driving, the image data can reflect the driving characteristics of the vehicle.

S203: Match the driving characteristics with the road characteristics of the road network roads in the road network database.

Considering that the matching process of S203 depends on the driving characteristics of S202, steps S202 and S203 will be jointly described below.

Among them, the road network database is pre-created, which includes road characteristics and attribute information of one or more road network roads, such as the shape of the road network roads, intersection information, road types, etc. Therefore, the attribute information of at least one road network road can be updated. The specific updated road network roads need to be specified according to the road network update task.

It can be understood that the purpose of matching here is to determine whether the image data is collected for the road network roads in the road network database, which includes two aspects: First, determine that the vehicle's driving road is the road network road in the road network database. ; The second aspect is to determine whether the shooting direction of the image collection device is directed to the road network road. When the image data and the road network road are successfully matched, it is determined that the image data is collected for the road network road with which the matching is successful.

In some embodiments, the above matching process may include: first, generating the driving characteristics of the vehicle according to the image data to match the driving characteristics with the road characteristics of the above-mentioned road network roads in the road network database.

Among them, the driving characteristics are used to represent the characteristics of the vehicle during driving. According to the driving characteristics, the driving road of the vehicle can be determined. Different driving roads correspond to different driving characteristics. Road features are used to represent the characteristics of roads in a road network and are used to uniquely represent a road. Different road network roads have different corresponding road features.

Specifically, the driving characteristics may include at least one of the following: driving trajectory, driving direction, driving height, and camera attitude; the road characteristics may include at least one of the following: fence areas, road directions, and road heights. The road direction includes at least one of the following: the pointing direction from the first position to the second position on the road network, and the three-dimensional vector corresponding to the extension direction of the road network.

Among them, the above-mentioned fence area is used to represent the area where the road network roads are located, the road direction refers to the allowed driving direction on the road network roads, and the road height is used to represent the height of the road network roads compared to the ground. The road network roads in this application are actual roads split or spliced according to preset lengths. After the road network is obtained, it is surrounded by a fence. The area corresponding to the fence is called the fence area of the road network and is used to determine whether the vehicle has entered the road network. Specifically, when a vehicle enters a fenced area of a road network, it can be considered that the vehicle has entered the road network; otherwise, it is considered that the vehicle has not entered the road network.

The above preset length can be flexibly set according to the actual application scenario. For example, it can be set to 600 meters to 800 meters, and the width of the fence area can be 20 meters. When the preset length is 600 meters to 800 meters, many road network roads are formed with a length ranging from 600 meters to 800 meters, and the corresponding fence areas are 600 meters to 800 meters long and 20 meters wide.

Therefore, matching the driving characteristics and the road characteristics includes performing at least one of the following matching processes: matching between the driving trajectory and the fence area (which can be called the first matching), and the matching between the driving direction and the pointing direction (which can be called the first matching). is the second matching), the matching between driving height and road height (can be called the third matching), and the matching between the camera attitude and the three-dimensional vector (can be called the fourth matching); if at least one of the above matching processes If the matching is successful, it is determined that the driving characteristics and the road characteristics are successfully matched.

Of course, in order to improve the accuracy of matching, it can be determined that the matching of driving characteristics and road characteristics is successful when all the above four matching processes are successful; otherwise, it is determined that the matching of driving characteristics and road characteristics fails.

The driving trajectory is a trajectory formed by multiple passing positions of the vehicle during driving. The passing positions are consistent with the collection positions provided by the image acquisition device when collecting image data, and the driving trajectory is non-directional. When the vehicle's driving trajectory is consistent with the road shape of the road network, it means that the first matching process is successful, and the vehicle's driving path may be a road network road; when the vehicle's driving trajectory is inconsistent with the road shape of the road network, it means that the third matching process is successful. The matching process fails and the vehicle's driving road is not a road network road.

The traveling direction is used to indicate that the vehicle travels from a first position to a second position, or from a second position to a first position. The first position and the second position are two positions on the road network. The pointing direction of the road network can be designated as the first position to the second position. Therefore, it can be determined whether the driving direction and the pointing direction are consistent. When the driving direction is consistent with the pointing direction, it means that the second matching process is successful, and the vehicle's driving road may be a road network road; when the driving direction is inconsistent with the pointing direction, it means that the second matching process fails, and the vehicle's driving road is not. Road network roads.

Ride height is used to indicate the height of the vehicle when driving, which is consistent with the height of the road the vehicle is traveling on. When the driving height is consistent with the road height, it means that the third matching process is successful, and the vehicle's driving road may be a road network road; when the driving direction is inconsistent with the road height, it means that the third matching process fails, and the vehicle's driving road is not. Road network roads.

The camera posture is used to represent the shooting direction of the image acquisition device when collecting image data, and can be represented by a three-dimensional vector. When the camera posture is consistent with the three-dimensional vector corresponding to the extension direction of the road network, it means that the fourth matching process is successful. The shooting direction of the image acquisition device is directed at the road network; when the camera posture is consistent with the extension direction of the road network, When the corresponding three-dimensional vectors are inconsistent, it means that the fourth matching process fails, and the shooting direction of the image acquisition device is not aimed at the roads of the road network.

In some implementations, the above driving characteristics can be generated in the following manner: generating the driving characteristics of the vehicle based on the collected information corresponding to the image data. The collection information includes at least one of the following: collection address, collection time, height information, and camera posture when collecting images.

Among them, the collection information is auxiliary information when collecting image data, and is used to represent the scene when collecting image data. The above collection address is the address of the vehicle when collecting image data, which can be represented by GPS coordinates. The collection time is the time when image data is collected, which can be expressed in GPS time. Height information is the height of the vehicle when image data is collected, which can be represented by GPS elevation information. For example, the height of a vehicle is different when it is driving on an elevated highway or on a road under a bridge.

Specifically, the above driving trajectory can be generated in the following manner: generating the driving trajectory of the vehicle based on the collection addresses corresponding to the multiple frames of image data corresponding to the same vehicle. That is to say, the driving trajectory includes collection addresses corresponding to the multiple frames of image data, and the collection addresses are used as the passing addresses of the vehicle during driving.

Correspondingly, the above-mentioned first matching process includes: determining the number of collection addresses falling into the fenced area from the above-mentioned driving trajectory. If the number of collected addresses is greater than or equal to the first threshold, it is determined that the driving trajectory and the fence area are successfully matched, that is, the first matching process is successful; otherwise, it is determined that the driving trajectory and the fence area fail to match, that is, the first matching process fails.

Wherein, the above-mentioned first threshold may be a fixed value, for example, 60. The first threshold can also be a value that changes dynamically with the total number of collection addresses in the driving trajectory. Specifically, it can be the product of the total number of collection addresses in the driving trajectory and the preset proportion. That is to say, the driving trajectory falls into the fenced area. Whether the proportion of the number of collection addresses within is greater than or equal to the preset proportion. For example, the total number of collected addresses in the driving trajectory × 70%. Of course, when the above-mentioned first threshold dynamically changes with the total number of collected addresses in the driving trajectory, the matching accuracy of the first matching process is higher.

Figure 3 is a schematic diagram of the relationship between a driving trajectory and a fenced area of a road network provided by an embodiment of the present application. Referring to Figure 3, the driving trajectory of the vehicle includes a total of 10 collection addresses D1, D2, D3, D4, D5, D6, D7, D8, D9, and D10. These 10 collection addresses can be used to collect 10 frames of image data for the vehicle. corresponding collection addresses respectively. As can be seen from Figure 3, 9 collection addresses in the driving trajectory fall within the fenced area W of the road network, and one collection address in the driving trajectory falls outside the fenced area W of the road network.

When the above-mentioned first threshold is the total number of collection addresses in the driving trajectory × 70%, for the scenario shown in Figure 3, the specific value of the first threshold is 10 × 70% = 7. Since the number 9 of collected addresses falling into the fenced area W is greater than the above-mentioned first threshold 7, it can be determined that the driving trajectory of the vehicle matches the fenced area W of the road network.

To sum up, the embodiments of the present application can realize the above-mentioned first matching process based on the number of collected addresses that fall into the fenced area in the driving trajectory, and can ensure accurate matching between the driving trajectory and the fenced area of the road network.

For the above second matching process, matching can be achieved through the following process: according to the collection address and collection time corresponding to at least two frames of image data corresponding to the same vehicle, determine the first time when the vehicle reaches the first position, and the first time when the vehicle reaches the second position. the second time; then, determine the driving direction based on the relationship between the first time and the second time. Specifically, if the first time is less than the second time, the traveling direction is determined to be the direction from the first position to the second position; if the second time is less than the first time, the traveling direction is determined to be the direction from the second position to the first position. .

Wherein, the above-mentioned first time is the time when the vehicle reaches the first position, and the second time is the time when the vehicle reaches the second position. The determination strategies of the first time and the second time are the same. Determining the first time is used as an example for explanation. The first time can be determined through the following multiple strategies.

In the first time determination strategy, the first position and the collection addresses corresponding to at least two frames of image data can be matched, so that the collection time corresponding to the collection address consistent with the first position is used as the first time.

In the second time determination strategy, at least one collection address closest to the first location may be selected, and then the average value of the collection times corresponding to the at least one closest collection address may be used as the first time.

In the third time determination strategy, the vehicle's driving route can be simulated based on the collection address and collection time. The driving route is a continuous route, and each location on it corresponds to a time, which is used to represent the time when the vehicle arrives at that location. , and then find the first location on the first driving route, and use the time corresponding to the first location as the first time.

When the above-mentioned first time determination strategy fails, the second time determination strategy or the third time determination strategy may be adopted.

When the driving direction is obtained through at least one of the above time strategies, the driving direction can be matched with the pointing direction of the road network road. When the driving direction and the pointing direction of the road network road are the same, it is determined that the above second matching process is successful; when the driving direction and the pointing direction of the road network road are not the same, it is determined that the above second matching process fails.

In other embodiments, after determining the first time and the second time, the first time and the second time may be used as driving characteristics to match them with the pointing direction of the road network road. Taking the pointing direction as the first position to the second position as an example, when the first time is less than the second time, it is determined that the driving direction and the pointing direction of the road network are successfully matched; when the first time is greater than or equal to the first time, it is determined The driving direction and the pointing direction of the road network failed to match.

For the aforementioned third matching process, when the image acquisition device provides elevation information at the same time when collecting image data, the elevation information can be used as the height information in the driving characteristics. When the above-mentioned image acquisition device does not provide elevation information, the height information in the driving characteristics can be determined through the following process: First, according to the acquisition information of the image data and the screen information of the image data, determine the shape of the vehicle's driving route, the driving speed and the height above the vehicle. Whether there are obstructions; then, the driving height is determined based on the shape of the vehicle's driving route, driving speed, and whether there are obstructions above the vehicle. Specifically, if the shape of the driving route is a straight line, the driving speed is constant, and there are no obstructions above the vehicle, it means that the vehicle has been moving forward at a constant speed, and it can be considered that the vehicle has not passed the intersection. At this time, since there are no obstructions above the vehicle, then it can If the vehicle is considered to be driving on a viaduct, the driving height is determined to be the elevated road height; otherwise, the vehicle can be considered to be driving on an ordinary road under the viaduct, and the driving height is determined to be the ordinary road height.

Among them, the ordinary road height is the height of roads other than elevated roads, which can usually be 0. The shape of the driving route can be determined based on the driving trajectory, and the process of determining the driving trajectory can refer to the foregoing description. The driving route shape may include but is not limited to: a straight line, a left-turn curve, a right-turn curve, etc. The driving route shape is used to determine whether the vehicle is going straight. When the shape of the driving route corresponds to a straight line, it means that the vehicle is going straight; otherwise, it means that the vehicle is turning.

Figure 4 is a schematic diagram of a driving route shape provided by an embodiment of the present application. Referring to FIG. 4 , when the vehicle travels from position A1 to position A2, the shape of the traveling route is a straight line. When the vehicle travels from position A1 to position A3, the shape of the driving route is a left-turn curve. When the vehicle travels from position A1 to position A4, the shape of the driving route is a right-turn curve.

Whether there is an obstruction above the vehicle can be obtained by analyzing the screen information corresponding to the image data. In the embodiment of this application, the obstruction mainly refers to the viaduct.

The driving speed is the speed of the vehicle during driving, and it can be multiple. A driving speed can be calculated between the collection of any two frames of image data. The driving speed can be calculated based on the distance between the collection addresses of the two frames of image data and the time difference between the collection times, specifically the ratio of the distance to the time difference.

A calculation process for a driving speed is given below. First, for continuous multiple frames of image data corresponding to the same vehicle, image sampling is performed according to the frame number interval N-1 to obtain M frames of images. For example, when N is 10, 10 frames of images can be obtained by sampling from 100 frames of image data. Then, for the sampled M frames of images, a driving speed can be calculated based on each two adjacent image data. In this way, M-1 driving speeds can be obtained: V_1 to V_M-1.

When multiple driving speeds are obtained, it can be determined whether the driving speed is uniform, including: first, calculating the average speed; then, judging the difference between M-1 driving speeds: V_1 to V_M-1 and the average speed V_A , if the differences are less than the preset threshold, it is determined to be a uniform speed, otherwise, it is determined to be a non-uniform speed, including: acceleration, deceleration, etc.

The above-mentioned strategies for calculating average speed can include multiple methods.

In the first average speed calculation strategy, the average speed V_A of M-1 driving speeds can be calculated.

In the second calculation strategy of average speed, the above-mentioned consecutive multiple frames of image data are obtained from the vehicle entering the road network to exiting the road network. Therefore, the collection time of the first frame of image data can be regarded as the entry time, and the last frame of image data can be regarded as the entry time. The collection time of frame image data is used as the exit time to calculate the driving time. The distance between the collection address corresponding to the first frame of image data and the collection address corresponding to the last frame of image data is regarded as the road length of the road network. The average speed can be calculated based on the road length and driving time.

When the vehicle is traveling at high speed, although the acquisition time of the first frame of image lags behind the time it enters the road network for a short period of time, the vehicle has already traveled a certain distance on the road network. In the same way, after the last frame of image collection, the vehicle still travels on the road network for a certain distance before exiting the road network. Therefore, it is necessary to add a reserved length to the road length so that the distance used in calculating the average speed is as close as possible to the actual length. The average speed can be calculated by the following formula: (road length + reserved length) / driving time.

The above reserved length is related to the vehicle's driving speed and the collection time difference of two adjacent frames of image data. After setting the driving speed and collection time difference according to empirical values, the reserved length can be calculated.

Figure 5 is a schematic diagram of the relationship between driving speed and image data provided by an embodiment of the present application. Referring to FIG. 5 , there are 10 frames of image data P1 to P10, which respectively correspond to the collection addresses D1 to D10 and the collection times T1 to T10, and T1 to T10 gradually increase. Therefore, the first driving speed V_1=D(D1, D2)/(T2-T1) can be obtained based on the distance D(D1, D2) and T2-T1 between the collection addresses D1 and D2 of the image data P1 and P2. . In the same way, we can get the second to ninth driving speeds V_2=D(D2, D3)/(T3-T2), V_3=D(D3, D4)/(T4-T3), V_4=D(D4, D5 )/(T5-T4), V_5＝D(D5, D6)/(T6-T5), V_6＝D(D6, D7)/(T7-T6), V_7＝D(D7, D8)/(T8- T7), V_8=D(D8, D9)/(T9-T8), V_9=D(D9, D10)/(T10-T9).

In addition, it can be seen from Figure 5 that the distance between the collection address of the first frame image P1 and the collection address of the last frame image P10 is the actual length. There is a semicircle with a radius of reserved length/2 before the first frame image P1 and after the last frame image P10, so as to set the reserved length/2 before the vehicle enters the road network, and when the vehicle exits the road network Then set the reserved length/2.

S204: Update the corresponding attribute information of the successfully matched road network road in the road network database through the image data.

Among them, the attribute information can be any attribute that may change corresponding to the road network, including but not limited to: the number of intersections, intersection traffic light duration, sidewalk information, etc.

It can be understood that the above attribute information can be obtained by analyzing the picture information in the image data. When the attribute information obtained by parsing the picture information in the image data is inconsistent with the corresponding attribute information in the road network database, the attribute information in the road network database can be updated with the parsed attribute information.

In some implementations, in order to improve the accuracy of updating attribute information of road network roads as much as possible, it is necessary to ensure the quality of image data. In this embodiment of the present application, some image data with poor quality can also be filtered out before updating the corresponding attribute information of the road network roads in the road network database. Image data of poor quality may include at least one of the following image data: image data outside the target time period, image data outside the target area, image data collected outside the target weather state, and the occlusion area on the driving road is greater than or equal to the second The image data of the threshold, and the image data of the sun azimuth angle outside the preset angle range, the sun azimuth angle includes the angle between the straight line between the sun and the vehicle and the driving direction of the vehicle.

The above-mentioned target time period may be a time period with better lighting and/or a closer time. On the one hand, it can ensure the illumination of the image data as much as possible to avoid poor image clarity caused by too dark an image; on the other hand, it can also ensure that the image data is as up-to-date as possible, which helps to ensure the path of image data analysis. The attribute information of the network roads is the latest, which can improve the accuracy of the attribute information of the road network roads.

For image data outside the above target time period, the following process can be used to filter out: First, determine the sunshine time period based on the collection time and collection address of the image data. The sunrise time and the sunrise time corresponding to the collection time period and the collection address at the collection time are The sunset time is associated; then, the target time period is generated based on the sunshine time period and/or the preset time period before the current time; finally, the image data whose acquisition time is outside the corresponding target time period is filtered out.

The above-mentioned sunrise time and sunset time can be read from the sunrise and sunset database according to the collection time and collection address. Since the sunrise and sunset times are different at different times and addresses, the above solution of this application can ensure that the sunrise time and sunset time are accurate times that match the collection time and collection address, which helps to improve the screening accuracy of image data. This improves the accuracy of updating attribute information of roads in the road network.

After obtaining the above sunrise time and sunset time, the sunshine period can be determined. The period between sunrise time and sunset time can be directly used as the sunshine period. However, considering that the illumination is also low after sunrise or before sunset, the period between sunrise time + T and sunset time - T can be regarded as the sunshine period. Among them, T can be set flexibly based on experience, for example, it can be set to 30 minutes.

The length of the above-mentioned preset time period can also be set flexibly. For example, it can be set to 3 days. In this case, the preset time period is the last 3 days.

After determining the sunshine period and/or the above-mentioned preset time period, the target time period may be determined. The target time period can be a sunshine time period, a preset time period, or an overlapping time period of the sunshine time period and the preset time period. For example, the target time period can be the time period between 7:30 am and 7:30 pm, or the time period in the past 3 days, or it can also include the time period between 7:30 am and 7:30 pm in the past 3 days. Multiple time periods.

For image data outside the above target area, filtering can be achieved through the following process: determine the collection address of the image data to determine whether the collection address belongs to the target area. If not, the image data is filtered out, otherwise, the image data is retained. Among them, the target area can be set when starting the update task. The embodiments of the present application can update the road network in some areas by filtering the image data of the target area, and can also avoid the interference of image data in other areas, thereby improving the update accuracy.

For image data other than the above target weather conditions, filtering can be achieved through the following process: first, determine the corresponding weather condition based on the collection time and collection address of the image data to ensure the accuracy of the weather condition; then, filter out the weather condition is image data outside the target weather state. The target weather conditions include weather conditions with visibility greater than or equal to the third threshold.

It can be understood that the weather state is associated with the collection time and collection address, and different collection times and/or different collection addresses may correspond to different weather states. Specifically, the corresponding weather status can be obtained from the weather network based on the collection time and collection address. In order to improve accuracy, hour-level weather status can be obtained based on the collection time.

The weather conditions in which the visibility is greater than or equal to the third threshold may include: sunny days, or weather with light haze, or light rain or snow. In weather conditions with greater visibility, the clarity of the image data collected is higher; in weather conditions with lower visibility, the clarity of the image data collected is lower. The third threshold can be set based on experience, for example, it can be set to 500. When the visibility is greater than or equal to 500, the clarity of the collected image can meet the requirements for recognition accuracy.

This application can filter out image data collected under weather conditions with visibility less than the third threshold, thus ensuring the clarity of the image data and improving the update accuracy of the road network.

For the image data whose occlusion area on the driving road is greater than or equal to the second threshold, filtering can be achieved through the following process: first, identify the occlusion area on the driving road from the image data, including but not limited to: vehicle occlusion, snow occlusion, Occlusion by other objects, etc.; then, filter out image data whose occlusion area is greater than or equal to the second threshold.

The second threshold may be a fixed value, for example, 10 square meters. The second threshold can also change with the road area, for example, it can be the road area × the preset area ratio. For example, when the preset area ratio is 30%, the second threshold is 30% of the road area.

Figure 6 is a schematic diagram of the blocking area on a driving road provided by an embodiment of the present application. Referring to Figure 6, the road area is S, including the blocking areas S1 to S3, S1=6 square meters, S2=4 square meters, S3=1 square meter, and the road area S is 50 square meters. Therefore, when the second threshold is 30%=50×30%=15 of the road area S, the image data can be retained due to the occlusion area S1+S2+S3=6+4+1=11.

When the above-mentioned driving road or blocking area is a regular graphic, the length and width of the driving road can be identified from the image data and calculated through the area formula corresponding to the regular graphic. For example, driving roads are usually rectangular, so the road area S can be obtained based on the product of length and width.

This application can filter out image data with an excessively large occluded area to ensure that the attribute information of the road network roads can be accurately identified, which helps to improve the accuracy of updating the attribute information of the road network roads.

For image data whose sun azimuth angle is outside the preset angle range, filtering can be achieved through the following process: First, according to the collection time and collection address of the image data, determine the straight line formed by the sun and the vehicle, as well as the driving direction of the vehicle. direction; then, determine the angle between the straight line and the driving direction as the sun azimuth angle; finally, filter out image data whose sun azimuth angle is outside the preset angle range.

Figure 7 is a schematic diagram of a sun direction angle provided by an embodiment of the present application. Referring to FIG. 7 , the position X of the vehicle and the position Y of the sun form a straight line L, and the angle J formed by it and the traveling direction F of the vehicle is the sun direction angle.

The above vehicle location X is the collection address corresponding to the image data. The sun position Y can be determined based on the collection time and collection address of the image data. When the sun's azimuth angle is within the preset angle range, the image data is not backlit when collecting it, so the image data will not be overexposed. When the sun's azimuth angle is outside the preset angle range, the image data is backlit when it is collected. , therefore, the image data may be overexposed. For example, the image acquisition device collects image data in the direction of the vehicle while the vehicle is driving to obtain picture information including the driving road. Therefore, the preset angle range can be 0 to 20 and 40 to 180 degrees. is 20 degrees to 40 degrees. In this scenario, when a vehicle is driving toward the sun during a certain time period in the morning or a certain time period in the afternoon, the image data collected on the road ahead will be overexposed, but the problem will not occur during the rest of the time.

Embodiments of the present application can filter out image data that may have exposure problems based on the sun's azimuth angle to further improve the quality of the image data, thereby improving the accuracy of updating attribute information of roads in the road network.

Figure 8 is a detailed flow chart of a road network update process provided by an embodiment of the present application. Referring to FIG. 8 , the detailed flow of the above road network update process includes the following steps S301 to S322.

S301: Obtain image data collected while the vehicle is driving.

After executing S301, four types of matching can be performed directly, including: the first matching process of S302 to S306, the second matching process of S307 to S311, the third matching process of S312 to S318, and the fourth matching process of S319. These four matching processes can be performed simultaneously or sequentially, and the embodiments of this application do not limit their order.

S302: Generate the driving trajectory of the vehicle based on the collection addresses corresponding to the multiple image data corresponding to the same vehicle.

S303: Determine the number of collection addresses falling into the fenced area from the driving trajectory.

S304: Determine whether the number of collection addresses is greater than or equal to the first threshold.

If yes, go to S305, otherwise, go to S306.

S305: It is determined that the driving trajectory and the fence area are successfully matched.

S306: It is determined that the driving trajectory and the fence area fail to match.

S307: Determine the first time for the vehicle to arrive at the first position and the second time for it to arrive at the second position based on the collection addresses and collection times corresponding to at least two frames of image data corresponding to the same vehicle.

S308: Determine whether the first time is less than the second time.

If yes, go to S309; otherwise, go to S310.

S309: Determine the traveling direction as the direction from the first position to the second position.

S310: Determine the traveling direction as the direction from the second position to the first position.

S311: Match the driving direction and pointing direction.

S312: Determine whether the image data corresponds to elevation information.

If yes, go to S313; otherwise, go to S314.

S313: Use the elevation information as the driving height.

S314: Based on the collection information of the image data and the screen information of the image data, determine the shape of the vehicle's driving route, the driving speed and whether there are obstructions above the vehicle.

S315: Determine whether the shape of the driving route is a straight line, whether the driving speed is uniform, and whether there are no obstructions above the vehicle.

If yes, go to S316; otherwise, go to S317.

S316: Determine the driving height to be the elevated road height.

S317: Determine the driving height to be the normal road height.

S318: Match ride height with road height.

S319: Match the camera posture when collecting images with the three-dimensional vector corresponding to the extension direction of the road network.

S320: Determine whether all the above four matching processes are successful.

If so, proceed to S320, otherwise the image data will not be used to match the attribute information of the roads in the road network.

S321: Filter out at least one of the following image data: image data outside the target time period, image data outside the target area, image data collected outside the target weather state, image data with an occlusion area on the driving road greater than or equal to the second threshold. , and image data whose sun azimuth angle is outside the preset angle range.

S322: Update the corresponding attribute information of the successfully matched road network road in the road network database through the image data.

It should be noted that the above S301 to S322 may refer to the foregoing description of the corresponding positions, and will not be described again here. In addition, the order of the above-mentioned S301 to S322 can be flexibly adjusted on the basis of mutual independence, and this application does not limit the order.

Figure 9 is a structural block diagram of a road network updating device provided by an embodiment of the present application. Referring to Figure 9, the above-mentioned road network updating device 400 includes:

The image data acquisition module 401 is used to acquire image data collected while the vehicle is driving.

The driving characteristic generation module 402 is used to generate the driving characteristic of the vehicle according to the image data.

The matching module 403 is used to match the driving characteristics with the road characteristics of the road network roads in the road network database.

The update module 404 is configured to update the attribute information corresponding to the successfully matched road network road in the road network database through the image data.

Optionally, the road features include at least one of: fenced area, road direction, and road height.

Optionally, the matching module 402 is also used to:

The driving characteristics of the vehicle are generated according to the collection information corresponding to the image data. The collection information includes at least one of the following: collection address, collection time, height information and camera posture when collecting images.

Optionally, the road direction includes at least one of the following: the pointing direction from the first position to the second position on the road network, and the three-dimensional vector corresponding to the extension direction of the road network;

The driving characteristics include at least one of the following: driving trajectory, driving direction, driving height and the camera attitude;

The matching module 402 is also used to:

Perform at least one of the following matching processes: matching between the driving trajectory and the fence area, matching between the driving direction and the pointing direction, matching between the driving height and the road height, and the matching between the camera pose and the three-dimensional vector;

If at least one of the matching processes is successful, it is determined that the driving characteristics and the road characteristics are successfully matched.

Optionally, the matching module 402 is also used to:

Generate the driving trajectory of the vehicle according to the collection addresses corresponding to the plurality of image data corresponding to the same vehicle;

From the driving trajectory, determine the number of collection addresses falling into the fence area;

If the number of collected addresses is greater than or equal to the first threshold, it is determined that the driving trajectory matches the fence area successfully.

Optionally, the matching module 402 is also used to:

Determine the first time for the vehicle to arrive at the first position and the second time for the vehicle to arrive at the second position based on the collection address and collection time corresponding to at least two frames of the image data corresponding to the same vehicle;

If the first time is less than the second time, determine the traveling direction to be the direction from the first position to the second position;

If the second time is less than the first time, the traveling direction is determined to be the direction from the second position to the first position.

Optionally, the matching module 402 is also used to:

According to the collection information of the image data and the picture information of the image data, determine the shape of the driving route of the vehicle, the driving speed and whether there are obstructions above the vehicle;

If the shape of the driving route is a straight line, the driving speed is constant, and there is no obstruction above the vehicle, then the driving height is determined to be the elevated road height;

Otherwise, the driving height is determined to be an ordinary road height.

Optionally, the device also includes:

A filtering module configured to filter out at least one of the following image data before the corresponding attribute information of the successfully matched road network road in the road network database through the image data update: image data outside the target time period, Image data outside the target area, image data collected outside the target weather state, image data with an occlusion area on the driving road greater than or equal to the second threshold, and image data with the sun azimuth outside the preset angle range, the The sun azimuth angle includes the angle between the straight line between the sun and the vehicle and the traveling direction of the vehicle.

Optionally, the filtering module is also used to:

Determine a sunshine time period based on the collection time and collection address of the image data, and the sunshine time period is associated with the sunrise time and sunset time corresponding to the collection address at the collection time;

Generate the target time period according to the sunshine time period and/or a preset time period before the current time;

Filter out image data whose acquisition time is outside the corresponding target time period.

Optionally, the filtering module is also used to:

Determine the corresponding weather state according to the collection time and collection address of the image data;

Filter out image data whose weather state is other than the target weather state, where the target weather state includes a weather state with visibility greater than or equal to a third threshold.

Optionally, the filtering module is also used to:

According to the collection time and collection address of the image data, determine the straight line formed by the sun and the vehicle, and the driving direction of the vehicle;

Determine the angle between the straight line and the traveling direction as the solar azimuth angle;

Image data whose sun azimuth angle is outside the preset angle range is filtered out.

The above device embodiment corresponds to the foregoing method embodiment. For specific description, please refer to the description in the foregoing method embodiment. The embodiments of the present application will not be described again here.

Figure 10 is a structural block diagram of an electronic device provided by an embodiment of the present application. The electronic device 600 includes a memory 602 and at least one processor 601 .

Among them, memory 602 stores computer execution instructions.

At least one processor 601 executes computer execution instructions stored in the memory 602, so that the electronic device 600 implements the aforementioned method in FIG. 2 .

In addition, the electronic device 600 may also include a receiver 603 and a transmitter 604, the receiver 603 is used to receive information from other devices or devices and forward it to the processor 601, and the transmitter 604 is used to send the information to other devices or devices. equipment.

In an exemplary embodiment, a non-transitory computer-readable storage medium is also provided. Computer-executable instructions are stored in the computer-readable storage medium. When executed by the processor, the computer-executable instructions are used to implement the above-mentioned road network. Update method.

In an exemplary embodiment, a computer program product is also provided for implementing the aforementioned road network update method.

Other embodiments of the present application will be readily apparent to those skilled in the art from consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of this application that follow the general principles of this application and include common knowledge or customary technical means in the technical field that are not disclosed in this application. . It is intended that the specification and examples be considered as exemplary only, with a true scope and spirit of the application being indicated by the following claims.

It is to be understood that the present application is not limited to the precise structures described above and illustrated in the accompanying drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the application is limited only by the appended claims.

Claims (14)

1. A road network update method, characterized in that the method includes: Obtain image data collected while the vehicle is driving; Generate driving characteristics of the vehicle based on the image data; Match the driving characteristics with the road characteristics of the road network roads in the road network database; The corresponding attribute information of the successfully matched road network road in the road network database is updated through the image data. 2. The method of claim 1, wherein the road characteristics include at least one of: fenced area, road direction, and road height. 3. The method according to claim 2, characterized in that generating the driving characteristics of the vehicle according to the image data includes: The driving characteristics of the vehicle are generated according to the collection information corresponding to the image data. The collection information includes at least one of the following: collection address, collection time, height information and camera posture when collecting images. 4. The method according to claim 3, wherein the road direction includes at least one of the following: the pointing direction from the first position to the second position on the road network, the extension of the road network The three-dimensional vector corresponding to the direction; The driving characteristics include at least one of the following: driving trajectory, driving direction, driving height and the camera attitude; Matching the driving characteristics with the road characteristics of the road network roads in the road network database includes: Perform at least one of the following matching processes: matching between the driving trajectory and the fence area, matching between the driving direction and the pointing direction, matching between the driving height and the road height, and the matching between the camera pose and the three-dimensional vector; If at least one of the matching processes is successful, it is determined that the driving characteristics and the road characteristics are successfully matched. 5. The method according to claim 4, characterized in that generating the driving characteristics of the vehicle according to the collection information corresponding to the image data includes: Generate the driving trajectory of the vehicle according to the collection addresses corresponding to the plurality of image data corresponding to the same vehicle; Performing matching between the driving trajectory and the fence area includes: From the driving trajectory, determine the number of collection addresses falling into the fence area; If the number of collected addresses is greater than or equal to the first threshold, it is determined that the driving trajectory matches the fence area successfully. 6. The method according to claim 4, characterized in that generating the driving characteristics of the vehicle according to the collection information corresponding to the image data includes: Determine the first time for the vehicle to arrive at the first position and the second time for the vehicle to arrive at the second position based on the collection address and collection time corresponding to at least two frames of the image data corresponding to the same vehicle; If the first time is less than the second time, determine the traveling direction to be the direction from the first position to the second position; If the second time is less than the first time, the traveling direction is determined to be the direction from the second position to the first position. 7. The method according to claim 4, characterized in that generating the driving characteristics of the vehicle according to the collection information corresponding to the image data includes: According to the collection information of the image data and the picture information of the image data, determine the shape of the driving route of the vehicle, the driving speed and whether there are obstructions above the vehicle; If the shape of the driving route is a straight line, the driving speed is constant, and there is no obstruction above the vehicle, then the driving height is determined to be the elevated road height; Otherwise, the driving height is determined to be an ordinary road height. 8. The method according to any one of claims 1 to 7, characterized in that the road network road successfully matched by the image data update is preceded by the corresponding attribute information in the road network database. include: Filter out at least one of the following image data: image data outside the target time period, image data outside the target area, image data collected outside the target weather state, image data with an occlusion area on the driving road greater than or equal to the second threshold, and Image data whose sun azimuth angle is outside the preset angle range. The sun azimuth angle includes the angle between the straight line between the sun and the vehicle and the driving direction of the vehicle. 9. The method of claim 8, wherein filtering out at least one of the following image data includes: Determine a sunshine time period based on the collection time and collection address of the image data, and the sunshine time period is associated with the sunrise time and sunset time corresponding to the collection address at the collection time; Generate the target time period according to the sunshine time period and/or a preset time period before the current time; Filter out image data whose acquisition time is outside the corresponding target time period. 10. The method according to claim 8, characterized in that filtering out at least one of the following image data includes: Determine the corresponding weather state according to the collection time and collection address of the image data; Filter out image data whose weather state is other than the target weather state, where the target weather state includes a weather state with visibility greater than or equal to a third threshold. 11. The method of claim 8, wherein filtering out at least one of the following image data includes: According to the collection time and collection address of the image data, determine the straight line formed by the sun and the vehicle, and the driving direction of the vehicle; Determine the angle between the straight line and the traveling direction as the solar azimuth angle; Image data whose sun azimuth angle is outside the preset angle range is filtered out. 12. A road network updating device, characterized in that it includes: The image data acquisition module is used to acquire image data collected while the vehicle is driving; A driving characteristic generation module, configured to generate the driving characteristics of the vehicle according to the image data; A matching module for matching the driving characteristics with the road characteristics of the road network roads in the road network database; An update module, configured to update the corresponding attribute information of the successfully matched road network road in the road network database through the image data. 13. An electronic device including a memory and at least one processor; Wherein, the memory stores computer execution instructions; The at least one processor executes the computer execution instructions stored in the memory, so that the electronic device implements the road network update method described in any one of claims 1 to 11. 14. A computer-readable storage medium, characterized in that computer-executable instructions are stored in the computer-readable storage medium, and when executed by a processor, the computer-executable instructions are used to implement any one of claims 1 to 11 the method described.