Road surface damage information acquisition method and system

Abstract

The invention is suitable for the technical field of road surface damage detection, and particularly relates to a road surface damage information acquisition method and system, and the method comprises the steps: constructing a road surface damage information collection platform, determining registered vehicles, collecting the vibration index of each registered vehicle, building the corresponding relation between the vibration index and the vehicle position, and obtaining the road surface damage information. Drawing a change trend chart by taking time as a horizontal coordinate and the vibration index as a vertical coordinate, extracting abnormal fluctuation, reading the vehicle position of the registered vehicle during abnormal fluctuation through the corresponding relation, and defining the vehicle position as a suspected position; dividing a pavement damage detection area, and drawing a road distribution diagram of the detection area. According to the method, the suspected position of the pavement defect can be verified by generating the verification snapshot, so that the accuracy of pavement defect monitoring data is greatly improved, false alarms are reduced, a high-precision pavement damage database is formed, a data basis is provided for road maintenance, and the road maintenance efficiency is greatly improved.

Description

Technical Field

[0001] The present invention relates to the technical field of road damage detection, and in particular to a method and system for acquiring road damage information. Background Art

[0002] Pavement damage information refers to detailed data on various structural or functional damages that occur on the road surface, including the type, location, scope, severity of the damage and its impact on traffic safety. Common types of pavement damage include cracks (such as longitudinal cracks, transverse cracks, and mesh cracks), potholes, subsidence, rutting, bulges, peeling, and looseness. These damages may be caused by traffic loads, natural environmental factors (such as temperature changes, rain erosion), construction quality problems, or material aging.

[0003] In real life, road damage information is generally collected through manual inspections and public feedback, but it is difficult to achieve large-scale, high-frequency monitoring, and it is impossible to accurately locate the damage.

[0004] Therefore, “how to utilize passing vehicles to collect road damage information” is the technical problem that the present invention needs to solve. Summary of the invention

[0005] The purpose of the present invention is to provide a method and system for obtaining road damage information, so as to solve the problem of "how to collect road damage information by using passing vehicles" raised in the above background technology.

[0006] To achieve the above object, the present invention provides the following technical solutions: A method for obtaining road damage information, the method comprising: Constructing a road damage information collection platform, determining registered vehicles, collecting vibration indicators of each registered vehicle, establishing a corresponding relationship between the vibration indicator and the vehicle position, drawing a change trend graph with time as the horizontal coordinate and the vibration indicator as the vertical coordinate, and extracting abnormal fluctuations, reading the vehicle position of the registered vehicle at the time of abnormal fluctuations through the corresponding relationship, and defining it as a suspected position; Demarcate a detection area for road damage, draw a road distribution map of the detection area, mark the suspected position on the road distribution map, record the number of occurrences of each suspected position, and delete the suspected position whose number of occurrences is less than a threshold in the road distribution map; Receive the path plan uploaded by the registered vehicle, and when the path plan passes through a suspected position, define the corresponding registered vehicle as a target vehicle, obtain video surveillance data taken by a front-view camera of the target vehicle, capture the segments before and after the suspected position, and divide them into several video frames, define the video frame where the target vehicle is located at the suspected position as a verification snapshot, input the verification snapshot into a pre-built analysis model, output an abnormal feature, determine whether the abnormal feature is road damage, and if so, define the suspected position as a target position, integrate the target position and the corresponding verification snapshot, generate a detection report, and upload the detection report to the information collection platform.

[0007] Furthermore, the steps of constructing a road damage information collection platform, determining registered vehicles, and collecting vibration indicators of each registered vehicle include: Build an identity authentication mechanism to receive verification information uploaded by users and identify the registrants; Collect vehicle information of each registrant, where the vehicle information includes at least: vehicle type and configuration, set screening rules, and select registered vehicles.

[0008] Furthermore, the step of establishing a corresponding relationship between the vibration index and the vehicle position, taking time as the horizontal coordinate and the vibration index as the vertical coordinate, drawing a change trend graph, and extracting abnormal fluctuations includes: Collecting the driving trajectory of the registered vehicle and embedding a timestamp, and establishing a mapping between the vibration index and the abnormal fluctuation based on the timestamp and the horizontal coordinate; A historical data set of abnormal fluctuations is obtained, and fluctuation features are identified, and a source corresponding to each fluctuation feature is configured, wherein the source includes at least: road debris and road damage.

[0009] Furthermore, the step of marking the suspected positions in the road distribution map, recording the number of occurrences of each suspected position, and deleting the suspected positions whose number of occurrences is less than a threshold in the road distribution map includes: Inserting a label generated by the number of occurrences into the suspected position of the road distribution map; The occurrence times are divided into a plurality of levels, wherein each level corresponds to at least one handling rule.

[0010] Furthermore, the step of defining the corresponding registered vehicle as a target vehicle and obtaining the video surveillance data captured by a front-view camera of the target vehicle includes: Configuring attribute data of the video surveillance data, wherein the attribute data at least includes: the speed of the target vehicle, the type of the target vehicle and the clarity of the front-view camera; An initial frequency of video frame segmentation is configured, and the initial frequency is adjusted based on the attribute data.

[0011] Furthermore, the step of determining whether the abnormal feature is road damage, and if so, defining the suspected position as a target position comprises: Setting a risk level for each of the target locations, wherein the risk level includes at least: high, medium, and low; A deceleration point is selected from the target positions corresponding to the high risk level, and when the registered vehicle reaches the deceleration point, a deceleration reminder is issued to the registered vehicle.

[0012] Furthermore, the system comprises: A definition module is used to build an information collection platform for road damage, determine registered vehicles, collect vibration indicators of each registered vehicle, establish a corresponding relationship between the vibration indicator and the vehicle position, draw a change trend chart with time as the horizontal coordinate and the vibration indicator as the vertical coordinate, and extract abnormal fluctuations. Through the corresponding relationship, the vehicle position of the registered vehicle when the abnormal fluctuation occurs is read and defined as a suspected position; A deletion module is used to divide the detection area of ​​road damage, draw a road distribution map of the detection area, mark the suspected position in the road distribution map, record the number of occurrences of each suspected position, and delete the suspected position whose number of occurrences is less than a threshold in the road distribution map; The uploading module is used to receive the path planning uploaded by the registered vehicle. When the path planning passes through a suspected position, the corresponding registered vehicle is defined as a target vehicle, the video monitoring data captured by the front-view camera of the target vehicle is obtained, the segments before and after the suspected position are intercepted, and divided into a number of video frames, the video frame where the target vehicle is located at the suspected position is defined as a verification snapshot, the verification snapshot is input into a pre-built analysis model, and the abnormal feature is output to determine whether the abnormal feature is road damage. If so, the suspected position is defined as the target position, the target position and the corresponding verification snapshot are integrated, a detection report is generated, and the detection report is uploaded to the information collection platform.

[0013] Furthermore, the definition module includes: A construction unit is used to construct an identity authentication mechanism, receive verification information uploaded by the user, and determine the registrant; A selection unit, used to collect vehicle information of each registrant, wherein the vehicle information includes at least: vehicle type and configuration, set screening rules, and select registered vehicles; A mapping unit, used for collecting the driving trajectory of the registered vehicle and embedding a timestamp, and establishing a mapping between the vibration index and the abnormal fluctuation based on the timestamp and the horizontal coordinate; The configuration unit is used to obtain a historical data set of abnormal fluctuations, identify fluctuation features, and configure a source corresponding to each fluctuation feature, wherein the source includes at least road debris and road damage.

[0014] Furthermore, the deletion module includes: An inserting unit, used for inserting a label generated by the number of occurrences into the suspected position of the road distribution map; The division unit is used to divide the occurrence times into a plurality of levels, wherein each level corresponds to at least one handling rule.

[0015] Compared with the prior art, the present invention has the following beneficial effects: By identifying registered vehicles, it is possible to broaden the data collection carriers, increase data sources, and greatly improve the collection scope of road damage information. By identifying abnormal fluctuations, a crowdsourcing road monitoring network can be formed, so that potholes and damaged sections can be accurately located without the need to install additional complex road detection equipment. By generating verification snapshots, the suspected locations of road defects can be verified, which greatly improves the accuracy of road defect monitoring data and reduces false alarms, thereby forming a high-precision road damage database, providing a data basis for road maintenance and greatly improving road maintenance efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 A flowchart of a method for obtaining road damage information provided by an embodiment of the present invention; Figure 2 A first sub-process flowchart of the method for obtaining road damage information provided by an embodiment of the present invention; Figure 3 A second sub-flow chart of the method for obtaining road damage information provided by an embodiment of the present invention; Figure 4 A third sub-flow chart of the method for obtaining road damage information provided by an embodiment of the present invention; Figure 5 A block diagram of a road damage information acquisition system provided by an embodiment of the present invention; Figure 6 A block diagram of the components of the definition module in the road damage information acquisition system provided by an embodiment of the present invention; Figure 7 A block diagram of the composition of a deletion module in the road damage information acquisition system provided by an embodiment of the present invention; Figure 8 This is a block diagram of the composition of the upload module in the road damage information acquisition system provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0017] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0018] In Example 1, Figure 1 The implementation process of the method for obtaining road damage information provided by an embodiment of the present invention is shown and described in detail below: S100: Build an information collection platform for road damage, identify registered vehicles, collect vibration indicators of each registered vehicle, establish a correspondence between the vibration indicator and the vehicle position, use time as the horizontal coordinate and the vibration indicator as the vertical coordinate, draw a change trend graph, and extract abnormal fluctuations. Through the correspondence, read the vehicle position of the registered vehicle when the abnormal fluctuation occurs, and define it as a suspected position.

[0019] An information collection platform is constructed, wherein the information collection platform is mainly used to receive road damage information uploaded by users; the information collection platform is used to verify the identity of users, determine registrants, and define the eligible vehicles corresponding to the registrants as registered vehicles, collect vehicle information of all registered vehicles, including basic data such as vehicle model, license plate number, tire specifications, initial tire pressure and shock absorption system vibration, and upload the basic data to the information collection platform; obtain the real-time location of the vehicle through the on-board GPS or mobile network, synchronize the tire pressure data and the real-time location of the vehicle through the correspondence between tire pressure data and time, and between the real-time location of the vehicle and time; the vibration indicators include: tire pressure and shock absorption system vibration data. In the following description, the vibration indicator is tire pressure as an example.

[0020] A tire pressure change trend graph is drawn with time as the horizontal axis and the tire pressure of the registered vehicle at the corresponding time as the vertical axis. By analyzing the tire pressure changes of the registered vehicle, it can be determined whether the registered vehicle has passed through a road section with damaged road surface. The part of the tire pressure change trend graph in which the tire pressure changes is defined as an abnormal fluctuation. Furthermore, when abnormal tire pressure changes are detected at the same location for multiple registered vehicles, and the tire pressure continues to be abnormal, it is inferred that there may be road damage at that location, such as potholes, cracks or protrusions. At this time, the corresponding location is defined as a suspected location. In other words, the suspected location refers to a road section where there may be road damage.

[0021] S200: Demarcate a detection area for road damage, draw a road distribution map of the detection area, mark the suspected positions on the road distribution map, record the number of occurrences of each suspected position, and delete the suspected positions whose number of occurrences is less than a threshold in the road distribution map.

[0022] Demarcate the area where road damage detection is required, namely the detection area. The detection area can be divided based on urban administrative divisions, road grades (such as expressways, main roads, branch roads, etc.), traffic flow or historical damage records, etc., and draw a road distribution map within the detection area with reference to high-precision map data, and mark all major roads, intersections, bridges, tunnels and other key transportation infrastructure; mark the suspected locations on the road distribution map. In order to more intuitively display the road damage, the suspected locations can be classified by markings of different colors or shapes, such as potholes, cracks or protrusions corresponding to different icons, and accompanied by a damage degree assessment, which can be determined based on the degree of tire pressure change.

[0023] All suspected locations are clustered and analyzed, and similar coordinates are merged. The number of occurrences of each suspected location is recorded. If the number of occurrences is 1, it means that a registered vehicle has passed by the suspected location.

[0024] S300: Receive the path planning uploaded by the registered vehicle. When the path planning passes through a suspected position, define the corresponding registered vehicle as a target vehicle, obtain video surveillance data taken by the front-view camera of the target vehicle, extract the segments before and after the suspected position, and divide them into several video frames. Define the video frame where the target vehicle is at the suspected position as a verification snapshot, input the verification snapshot into a pre-built analysis model, output an abnormal feature, determine whether the abnormal feature is road damage, and if so, define the suspected position as a target position, integrate the target position and the corresponding verification snapshot, generate a detection report, and upload the detection report to the information collection platform.

[0025] Receive the path plan uploaded by the registered vehicle, where the path plan is the expected route of the registered vehicle. When the path plan needs to pass through a suspected location, define the registered vehicle as a target vehicle; with the authorization of the registrant corresponding to the registered vehicle, call the front-view camera of the target vehicle to obtain video surveillance data, and from the video surveillance data, extract the clips of the target vehicle approaching, passing and leaving the suspected location; divide the clips into several video frames at a certain time interval (initial frequency); it should be noted that at least one video frame should contain image data of road damage, and the corresponding video frame should be defined as a verification snapshot.

[0026] The verification snapshot is input into the analysis model. The analysis model is built based on deep learning and computer vision technology and trained with a large amount of labeled data. It can identify various abnormal road features, such as potholes, cracks, ridges and subsidence, as well as debris on the road, such as stones, bricks and objects dropped from vehicles. Specifically, a control set is constructed using abnormal features, and the type of each abnormal feature is labeled. A convolutional neural network or a visual model is used to extract image features from the verification snapshot and match them with the abnormal features in the control set to determine whether the same image features exist in the verification snapshot. If so, the corresponding image features are defined as abnormal features. It is determined whether the abnormal features are road damage (the abnormal features may also be debris). If so, the suspected location is defined as the target location, the verification snapshot at the target location is found, and the target location is marked in the verification snapshot. A test report is generated and uploaded to the information collection platform to provide data reference for the road administration department.

[0027] In Example 2, Figure 2 The implementation process of the road damage information acquisition method provided by the embodiment of the present invention is shown. The steps of building a road damage information collection platform, determining registered vehicles, and collecting vibration indicators of each registered vehicle are described in detail as follows: S101: Build an identity authentication mechanism, receive the verification information uploaded by the user, and determine the registrant.

[0028] In the information collection platform, a registration system is set up to receive verification information uploaded by users, where the verification information includes: user mobile phone number, email address, license plate number, vehicle identification code and driver's license information, etc. Users who have completed the verification are defined as registrants.

[0029] S102: Collect vehicle information of each registrant, wherein the vehicle information includes at least vehicle type and configuration, set screening rules, and select registered vehicles.

[0030] Determine the vehicle information of each registrant, set screening rules, and define vehicles that meet the screening rules as registered vehicles; the screening rules are formulated by the administrator of the information collection platform, and the screening rules can be: only select small vehicles; for example, a user's vehicle is a heavy truck. Since heavy trucks are not very sensitive to smaller road defects (such as potholes, cracks or protrusions), which affects the accuracy of information collection, heavy trucks cannot be selected as registered vehicles.

[0031] In Example 3, Figure 2The implementation process of the road damage information acquisition method provided by an embodiment of the present invention is shown. The following is a detailed description of the steps of establishing the correspondence between the vibration index and the vehicle position, taking time as the horizontal axis and the vibration index as the vertical axis, drawing a change trend graph, and extracting abnormal fluctuations, as follows: S103: collecting the driving trajectory of the registered vehicle and embedding a timestamp, and establishing a mapping between the vibration index and the abnormal fluctuation based on the timestamp and the horizontal coordinate.

[0032] Record the driving trajectory of the registered vehicle and upload it to the information collection platform; insert a timestamp into the driving trajectory to establish the corresponding relationship between the driving trajectory and time; compare the vibration index and the driving trajectory; for example, calculate the tire pressure change on a certain horizontal axis. If the tire pressure changes, it is defined as an abnormal fluctuation, where the abnormal fluctuation is: an instantaneous drop in tire pressure, a sharp increase, or frequent fluctuations in a short period of time.

[0033] S104: Acquire a historical data set of abnormal fluctuations, identify fluctuation features, and configure a source corresponding to each fluctuation feature, wherein the source at least includes: road debris and road damage.

[0034] By monitoring the tire pressure data of registered vehicles over a long period of time, a historical data set of abnormal fluctuations is constructed. When the tire pressure fluctuates abnormally, key features such as fluctuation amplitude, duration, fluctuation frequency and change rate are extracted. Cluster analysis is performed on the historical data of tire pressure to summarize different types of fluctuation features and establish a feature library, in which each source corresponds to a fluctuation feature. For example, in the tire pressure change trend graph, by finding a single peak, the corresponding source is small debris such as road stones and wood blocks. In real life, when a vehicle runs over stones, it will cause a single fluctuation in tire pressure in a short period of time, and a single peak will appear in the tire pressure change trend graph. However, this is different from the peak caused by road damage. When a vehicle passes through a damaged road section, multiple continuous peaks and troughs will generally be generated.

[0035] In Example 4, Figure 3 The implementation process of the road damage information acquisition method provided by an embodiment of the present invention is shown. The steps of marking the suspected position in the road distribution map, recording the number of occurrences of each suspected position, and deleting the suspected position whose number of occurrences is less than a threshold in the road distribution map are described in detail as follows: S201: inserting a label generated by the number of occurrences into the suspected position of the road distribution map.

[0036] Using the number of occurrences, labels are generated and inserted into the suspected positions.

[0037] S202: Divide the occurrence times into several levels, wherein each level corresponds to at least one handling rule.

[0038] The number of occurrences of each suspected location is determined, and the number of occurrences is divided into several levels, where each level corresponds to a handling rule. The higher the level, the more times the corresponding suspected location has been run over. The handling rule may be: sending a repair reminder to the road administration department.

[0039] For example, the first level is divided into: 0-5 corresponding occurrences, the second level, the corresponding occurrences are: 6-15, the third level, the corresponding occurrences are above 16, and the corresponding handling rules for the third level are: immediately report to the road administration department for handling.

[0040] In Example 5, Figure 4 The implementation process of the road damage information acquisition method provided by the embodiment of the present invention is shown. The steps of defining the corresponding registered vehicle as a target vehicle and acquiring the video monitoring data captured by the front-view camera of the target vehicle are described in detail as follows: S301: Configuring attribute data of the video surveillance data, wherein the attribute data at least includes: the speed of the target vehicle, the type of the target vehicle, and the clarity of the front-view camera.

[0041] Determine the registered vehicle corresponding to each video surveillance data and read out the attribute data, including: the speed of the target vehicle at the suspected location, the type of the target vehicle (whether it is a private car, different types of vehicle tires, different perception levels of road damage of different sizes) and the clarity of the target vehicle's front-view camera, etc.

[0042] S302: configuring an initial frequency of video frame segmentation, and adjusting the initial frequency according to the attribute data.

[0043] The initial frequency of the video frame is adjusted according to the attribute data; for example, assuming the initial frequency is 1 second per frame, if the target vehicle is traveling at a slower speed, the initial frequency can be increased to 2 seconds per frame. The specific correspondence should be pre-determined by the management personnel of the information collection platform.

[0044] In Example 6, Figure 4 The implementation process of the road damage information acquisition method provided by the embodiment of the present invention is shown. The following is a detailed description of the steps of determining whether the abnormal feature is road damage and, if so, defining the suspected position as the target position, as follows: S303: Setting a risk level for each target location, wherein the risk level includes at least: high, medium and low.

[0045] The verification snapshot corresponding to the abnormal feature is input into the analysis model again, and the risk level is output. The risk level can be determined based on the pixel range corresponding to the abnormal feature in the verification snapshot. According to the size of this pixel range, the risk level is divided into high, medium and low. The higher the risk level, the larger the area of ​​road damage.

[0046] S304: Select a deceleration point from the target position corresponding to the high risk level, and send a deceleration reminder to the registered vehicle when the registered vehicle reaches the deceleration point.

[0047] With the high-risk target location as the center, deceleration points are set at a preset distance before and after the center. When a registered vehicle arrives at the deceleration point, a deceleration reminder is sent to the registered vehicle. The deceleration reminder can be pushed to the driver of the registered vehicle through the on-board terminal (such as smart car machine, navigation system), mobile device (such as mobile phone APP) or HUD head-up display.

[0048] In Example 7, different from Example 1, in this embodiment of the present invention, the method further includes: Based on the target location, a manual review task is generated and published to the information collection platform; From the registrants, the task recipient is identified and a dialogue channel is established. When the task recipient arrives at the target location, the dialogue channel is activated, feedback information is generated, and the feedback information is integrated into the detection report.

[0049] Using the target location, manual review tasks are generated and published to the information collection platform.

[0050] For example, a manual review task is: "Task number: RD-20240101-001, target location: longitude and latitude (31.2304, 121.4737) (a main road in Shanghai), number of occurrences: 10 times, review task content: whether there are obvious potholes on this section of road."

[0051] If a registrant accepts the task, the registrant is defined as the task recipient, and multiple channels such as mobile applications and vehicle-mounted terminals are opened to the task recipient so that the task recipient can upload feedback information to the information collection platform; the feedback information may be: there are indeed obvious potholes at the location.

[0052] In order to ensure the driving safety of the task recipient, the feedback information should be uploaded to the information collection platform in the form of voice, and the feedback information should be integrated into the test report to further improve the accuracy of the test report.

[0053] Figure 5The following is a structural block diagram of a road surface damage information acquisition system according to an embodiment of the present invention. The road surface damage information acquisition system 1 includes: A definition module 11 is used to build an information collection platform for road damage, determine registered vehicles, collect vibration indicators of each registered vehicle, establish a corresponding relationship between the vibration indicator and the vehicle position, draw a change trend graph with time as the horizontal coordinate and the vibration indicator as the vertical coordinate, and extract abnormal fluctuations. Through the corresponding relationship, the vehicle position of the registered vehicle when the abnormal fluctuation occurs is read and defined as a suspected position; A deletion module 12 is used to divide the detection area of ​​road damage, draw a road distribution map of the detection area, mark the suspected position in the road distribution map, record the number of occurrences of each suspected position, and delete the suspected position whose number of occurrences is less than a threshold in the road distribution map; The uploading module 13 is used to receive the path planning uploaded by the registered vehicle. When the path planning passes through a suspected position, the corresponding registered vehicle is defined as a target vehicle, the video monitoring data taken by the front-view camera of the target vehicle is obtained, the segments before and after the suspected position are intercepted, and divided into a number of video frames, the video frame where the target vehicle is located at the suspected position is defined as a verification snapshot, the verification snapshot is input into a pre-built analysis model, and the abnormal feature is output to determine whether the abnormal feature is road damage. If so, the suspected position is defined as the target position, the target position and the corresponding verification snapshot are integrated, a detection report is generated, and the detection report is uploaded to the information collection platform.

[0054] Figure 6 The structure block diagram of the road damage information acquisition system provided by the embodiment of the present invention is shown, and the definition module 11 includes: The construction unit 111 is used to construct an identity authentication mechanism, receive the verification information uploaded by the user, and determine the registrant; The selection unit 112 is used to collect vehicle information of each registrant, wherein the vehicle information at least includes: vehicle type and configuration, set screening rules, and select registered vehicles; A mapping unit 113 is used to collect the driving trajectory of the registered vehicle and embed a timestamp, and establish a mapping between the vibration index and the abnormal fluctuation based on the timestamp and the horizontal coordinate; The configuration unit 114 is used to obtain a historical data set of abnormal fluctuations, identify fluctuation features, and configure a source corresponding to each fluctuation feature, wherein the source at least includes: road debris and road damage.

[0055] Figure 7 The structure block diagram of the road damage information acquisition system provided by the embodiment of the present invention is shown, and the deletion module 12 includes: An inserting unit 121, configured to insert a label generated by the number of occurrences into the suspected position of the road distribution map; The division unit 122 is used to divide the occurrence times into several levels, wherein each level corresponds to at least one handling rule.

[0056] Figure 8 The structure diagram of the road damage information acquisition system provided by the embodiment of the present invention is shown, and the uploading module 13 includes: The attribute unit 131 is used to configure the attribute data of the video surveillance data, wherein the attribute data at least includes: the speed of the target vehicle, the type of the target vehicle and the clarity of the front view camera; An adjustment unit 132, configured to configure an initial frequency of video frame segmentation, and adjust the initial frequency according to the attribute data; A setting unit 133, configured to set a risk level for each of the target locations, wherein the risk level includes at least: high, medium and low; The sending unit 134 is used to select a deceleration point in the target position corresponding to the high risk level, and send a deceleration reminder to the registered vehicle when the registered vehicle reaches the deceleration point.

[0057] The definition module 11 is mainly used to complete step S100, the deletion module 12 is mainly used to complete step S200, and the upload module 13 is mainly used to complete step S300; The construction unit 111 is mainly used to complete step S101, the selection unit 112 is mainly used to complete step S102, the mapping unit 113 is mainly used to complete step S103, and the configuration unit 114 is mainly used to complete step S104; The inserting unit 121 is mainly used to complete step S201, and the dividing unit 122 is mainly used to complete step S202; The attribute unit 131 is mainly used to complete step S301, the adjustment unit 132 is mainly used to complete step S302, the setting unit 133 is mainly used to complete step S303, and the issuing unit 134 is mainly used to complete step S304.

[0058] The technical features of the above-described embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above-described embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0059] The above-mentioned embodiments only express several implementation methods of the present invention, and the description thereof is relatively specific and detailed, but it cannot be understood as limiting the scope of the patent of the present invention. It should be pointed out that, for ordinary technicians in this field, several variations and improvements can be made without departing from the concept of the present invention, which all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be subject to the attached claims.

[0060] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for obtaining road damage information, characterized in that: The method comprises: Constructing a road damage information collection platform, determining registered vehicles, collecting vibration indicators of each registered vehicle, establishing a corresponding relationship between the vibration indicator and the vehicle position, drawing a change trend graph with time as the horizontal coordinate and the vibration indicator as the vertical coordinate, and extracting abnormal fluctuations, reading the vehicle position of the registered vehicle at the time of abnormal fluctuations through the corresponding relationship, and defining it as a suspected position; Demarcate a detection area for road damage, draw a road distribution map of the detection area, mark the suspected position on the road distribution map, record the number of occurrences of each suspected position, and delete the suspected position whose number of occurrences is less than a threshold in the road distribution map; Receive the path plan uploaded by the registered vehicle, and when the path plan passes through a suspected position, define the corresponding registered vehicle as a target vehicle, obtain video surveillance data taken by a front-view camera of the target vehicle, capture the segments before and after the suspected position, and divide them into several video frames, define the video frame where the target vehicle is located at the suspected position as a verification snapshot, input the verification snapshot into a pre-built analysis model, output an abnormal feature, determine whether the abnormal feature is road damage, and if so, define the suspected position as a target position, integrate the target position and the corresponding verification snapshot, generate a detection report, and upload the detection report to the information collection platform.

2. The method for obtaining road damage information according to claim 1, characterized in that: The steps of constructing a road damage information collection platform, determining registered vehicles, and collecting vibration indicators of each registered vehicle include: Build an identity authentication mechanism to receive verification information uploaded by users and identify the registrants; Collect vehicle information of each registrant, where the vehicle information includes at least: vehicle type and configuration, set screening rules, and select registered vehicles.

3. The method for obtaining road damage information according to claim 2, characterized in that: The step of establishing a corresponding relationship between the vibration index and the vehicle position, taking time as the horizontal coordinate and the vibration index as the vertical coordinate, drawing a change trend graph, and extracting abnormal fluctuations comprises: Collecting the driving trajectory of the registered vehicle and embedding a timestamp, and establishing a mapping between the vibration index and the abnormal fluctuation based on the timestamp and the horizontal coordinate; A historical data set of abnormal fluctuations is obtained, and fluctuation features are identified, and a source corresponding to each fluctuation feature is configured, wherein the source includes at least: road debris and road damage.

4. The method for obtaining road damage information according to claim 1, characterized in that: The step of marking the suspected positions in the road distribution map, recording the number of occurrences of each suspected position, and deleting the suspected positions whose number of occurrences is less than a threshold in the road distribution map comprises: Inserting a label generated by the number of occurrences into the suspected position of the road distribution map; The occurrence times are divided into a plurality of levels, wherein each level corresponds to at least one handling rule.

5. The method for obtaining road damage information according to claim 1, characterized in that: The step of defining the corresponding registered vehicle as a target vehicle and obtaining the video surveillance data captured by a front-view camera of the target vehicle comprises: Configuring attribute data of the video surveillance data, wherein the attribute data at least includes: the speed of the target vehicle, the type of the target vehicle and the clarity of the front-view camera; An initial frequency of video frame segmentation is configured, and the initial frequency is adjusted based on the attribute data.

6. The method for obtaining road damage information according to claim 2, characterized in that: The step of determining whether the abnormal feature is road damage and, if so, defining the suspected location as a target location comprises: Setting a risk level for each of the target locations, wherein the risk level includes at least: high, medium, and low; A deceleration point is selected from the target positions corresponding to the high risk level, and when the registered vehicle reaches the deceleration point, a deceleration reminder is issued to the registered vehicle.

7. The method for obtaining road damage information according to claim 6, characterized in that: The method further comprises: Based on the target location, a manual review task is generated and published to the information collection platform; From the registrants, the task recipient is identified and a dialogue channel is established. When the task recipient arrives at the target location, the dialogue channel is activated, feedback information is generated, and the feedback information is integrated into the detection report.

8. A road damage information acquisition system, characterized in that: The system comprises: A definition module is used to build an information collection platform for road damage, determine registered vehicles, collect vibration indicators of each registered vehicle, establish a corresponding relationship between the vibration indicator and the vehicle position, draw a change trend chart with time as the horizontal coordinate and the vibration indicator as the vertical coordinate, and extract abnormal fluctuations. Through the corresponding relationship, the vehicle position of the registered vehicle when the abnormal fluctuation occurs is read and defined as a suspected position; A deletion module is used to divide the detection area of ​​road damage, draw a road distribution map of the detection area, mark the suspected position in the road distribution map, record the number of occurrences of each suspected position, and delete the suspected position whose number of occurrences is less than a threshold in the road distribution map; The uploading module is used to receive the path planning uploaded by the registered vehicle. When the path planning passes through a suspected position, the corresponding registered vehicle is defined as a target vehicle, the video monitoring data captured by the front-view camera of the target vehicle is obtained, the segments before and after the suspected position are intercepted, and divided into a number of video frames, the video frame where the target vehicle is located at the suspected position is defined as a verification snapshot, the verification snapshot is input into a pre-built analysis model, and the abnormal feature is output to determine whether the abnormal feature is road damage. If so, the suspected position is defined as the target position, the target position and the corresponding verification snapshot are integrated, a detection report is generated, and the detection report is uploaded to the information collection platform.

9. The road damage information acquisition system according to claim 8, characterized in that: The definition module includes: A construction unit is used to construct an identity authentication mechanism, receive verification information uploaded by the user, and determine the registrant; A selection unit, used to collect vehicle information of each registrant, wherein the vehicle information includes at least: vehicle type and configuration, set screening rules, and select registered vehicles; A mapping unit, used to collect the driving trajectory of the registered vehicle and embed a timestamp, and establish a mapping between the vibration data and the abnormal fluctuation based on the timestamp and the horizontal coordinate; The configuration unit is used to obtain a historical data set of abnormal fluctuations, identify fluctuation features, and configure a source corresponding to each fluctuation feature, wherein the source includes at least road debris and road damage.

10. The road damage information acquisition system according to claim 8, characterized in that: The deletion module comprises: An inserting unit, used for inserting a label generated by the number of occurrences into the suspected position of the road distribution map; The division unit is used to divide the occurrence times into a plurality of levels, wherein each level corresponds to at least one handling rule.

Images