Pavement disease determination method and device, storage medium and electronic device

Abstract

The embodiment of the invention provides a pavement disease determination method and device, a storage medium and an electronic device, and the method comprises the steps: obtaining a plurality of pavement disease parameters, the pavement disease parameters being parameters determined through target images collected by a plurality of image collection devices of a target vehicle driving on a target pavement; classifying the plurality of pavement disease parameters according to disease positions included in the pavement disease parameters to obtain pavement disease subsets; determining a target disease type existing in the pavement disease subset, and determining a first probability that the pavement disease subset has a disease corresponding to the target disease type; and when the first probability is greater than a first preset probability, sending the pavement disease subset to a target platform to indicate the target platform to determine whether a disease of a target disease type exists in the target pavement based on the target pavement disease parameter and the first probability. The problems of low road inspection efficiency and low accuracy are solved, and the effect of improving the road inspection efficiency and accuracy is achieved.

Description

Technical Field

[0001] The embodiments of the present invention relate to the field of communications, and in particular, to a method, device, storage medium and electronic device for determining road surface defects. Background Art

[0002] In related technologies, road inspection departments mainly discover damaged roads through regular maintenance vehicle patrols, photo recording, private vehicle reporting of road problems, roadside monitoring equipment inspections, and manual review of road monitoring videos. The real-time nature of the inspections and the reliability of the inspection results cannot be effectively guaranteed, and the coverage of road monitoring videos is also insufficient.

[0003] It can be seen from this that the relevant technology has technical problems of low road inspection efficiency and low accuracy.

[0004] Currently, no effective solution has been proposed for the above-mentioned problems existing in the related technologies. Summary of the invention

[0005] The embodiments of the present invention provide a method, device, storage medium and electronic device for determining road defects, so as to at least solve the technical problems of low road inspection efficiency and low accuracy existing in the related art.

[0006] According to one embodiment of the present invention, a method for determining pavement damage is provided, comprising: acquiring a plurality of pavement damage parameters, wherein the pavement damage parameters are parameters determined by target images acquired by a plurality of image acquisition devices of a target vehicle traveling on a target road, and the states of the plurality of pavement damage parameters include at least one of the following: parameters in an unmarked state, parameters marked as waiting for confirmation; classifying the plurality of pavement damage parameters according to the damage positions included in the pavement damage parameters to obtain a pavement damage subset, wherein the target damage positions of the target pavement damage parameters included in the pavement damage subset are located within the same area; determining a target damage type existing in the pavement damage subset, and determining a first probability that a damage corresponding to the target damage type exists in the pavement damage subset; and when the first probability is greater than a first preset probability, sending the pavement damage subset to a target platform to instruct the target platform to determine whether a damage of the target damage type exists in the target road based on the target pavement damage parameters and the first probability.

[0007] In an exemplary embodiment, determining the target disease type existing in the pavement disease subset includes: determining the first disease type included in each of the target pavement disease parameters included in the pavement disease subset; determining the target proportion of each of the first disease types in all disease types included in the pavement disease subset; and determining the second disease type corresponding to the maximum proportion included in the target proportion as the target disease type.

[0008] In an exemplary embodiment, determining a first probability that the pavement disease subset contains a disease corresponding to the target disease type includes: determining a first disease parameter that the disease type included in the pavement disease subset is the target disease type; determining a disease probability included in each of the first disease parameters; and determining an average value of the disease probabilities as the first probability.

[0009] In an exemplary embodiment, after sending the pavement damage subset to a target platform, the method further includes: clearing the pavement damage subset.

[0010] In an exemplary embodiment, after determining a first probability that the pavement defect subset contains a defect corresponding to the target defect type, the method further includes: when the first probability is less than or equal to a first preset probability, marking the target pavement defect parameter as a parameter in a waiting for confirmation state.

[0011] According to another embodiment of the present invention, a method for determining pavement diseases is provided, comprising: receiving a pavement disease subset sent by a roadside device, wherein the pavement disease subset is sent by the roadside device when it is determined that a first probability that a disease corresponding to a target disease type exists in the pavement disease subset is greater than a first preset probability, and the pavement disease subset is obtained by the roadside device in the following manner: acquiring a plurality of pavement disease parameters, wherein the pavement disease parameters are parameters determined by target images acquired by a plurality of image acquisition devices of a target vehicle traveling on a target road surface, and the states of the plurality of pavement disease parameters include at least one of the following: parameters in an unmarked state, parameters marked as a waiting-for-confirmation state; classifying the plurality of pavement disease parameters according to the disease locations included in the pavement disease parameters to obtain a pavement disease subset, wherein the target disease locations of the target pavement disease parameters included in the pavement disease subset are within the same area; determining whether a disease of the target disease type exists in the target road surface based on the target pavement disease parameters and the first probability.

[0012] In an exemplary embodiment, determining whether the target road surface has a disease of the target disease type based on the target road surface disease parameter and the first probability includes: determining a target basic safety message included in the target road surface disease parameter at the target moment when the target vehicle collects the target image; determining a first basic safety message of the target vehicle in a first time period before the target moment and adjacent to the target moment; determining a second basic safety message of the target vehicle in a second time period after the target moment and adjacent to the target moment; determining a second probability based on the target basic safety message, the first basic safety message, and the second basic safety message; determining a target probability based on the first probability and the second probability; and determining that the target road surface has the target disease type when the target probability is greater than a second preset probability.

[0013] In an exemplary embodiment, determining the second probability based on the target basic safety message, the first basic safety message, and the second basic safety message includes: for each target sub-message of the target type included in the target basic safety message, performing the following operations to determine the average deviation value of the target sub-message, wherein the target type is the type of any sub-message included in the target basic safety message: determining a first sub-message of the target type included in the first basic safety message, determining a second sub-message of the target type included in the second basic safety message, determining the mean square error of the target sub-message, the first sub-message, and the second sub-message, and determining the mean square error as the average deviation value; determining a first weight corresponding to each average deviation value; determining a weighted sum of the average deviation value and the first weight; and determining the objective function value of the weighted sum as the second probability.

[0014] In an exemplary embodiment, determining the target probability based on the first probability and the second probability includes: determining a second weight corresponding to the first probability, and determining a third weight corresponding to the second probability; determining a first sum of the first probability and the second weight, and determining a second sum of the second probability and the third weight; and determining the sum of the first sum and the second sum as the target probability.

[0015] In an exemplary embodiment, after determining the target probability based on the first probability and the second probability, the method further includes: when the target probability is less than or equal to the second preset probability, marking the state of the pavement defect subset as a waiting for confirmation state.

[0016] In an exemplary embodiment, before determining whether the target pavement has a disease of the target disease type in the target pavement based on the target pavement disease parameters and the first probability, the method further includes: determining a first pavement disease subset that is in a waiting for confirmation state; in the case of a second disease subset whose disease locations included in the first pavement disease subset are the target disease locations, determining supplementary pavement disease parameters of the second disease subset; and supplementing the supplementary pavement disease parameters to the target pavement disease parameters.

[0017] According to another embodiment of the present invention, a device for determining pavement damage is provided, comprising: an acquisition module, for acquiring a plurality of pavement damage parameters, wherein the pavement damage parameters are parameters determined by target images acquired by a plurality of image acquisition devices of a target vehicle traveling on a target pavement, and the states of the plurality of pavement damage parameters include at least one of the following: parameters in an unmarked state, parameters marked as waiting for confirmation; a classification module, for classifying the plurality of pavement damage parameters according to the damage positions included in the pavement damage parameters to obtain a pavement damage subset, wherein the target disease positions of the target pavement damage parameters included in the pavement damage subset are located within the same area; a first determination module, for determining a target disease type existing in the pavement damage subset, and determining a first probability that the pavement damage subset has a disease corresponding to the target disease type; a sending module, for sending the pavement damage subset to a target platform when the first probability is greater than a first preset probability, to instruct the target platform to determine whether a disease of the target disease type exists in the target pavement based on the target pavement damage parameters and the first probability.

[0018] According to another embodiment of the present invention, a device for determining pavement damage is provided, comprising: a receiving module, for receiving a pavement damage subset sent by a roadside device, wherein the pavement damage subset is sent by the roadside device when it is determined that a first probability that a disease corresponding to a target disease type exists in the pavement damage subset is greater than a first preset probability, and the pavement damage subset is obtained by the roadside device in the following manner: acquiring a plurality of pavement damage parameters, wherein the pavement damage parameters are parameters determined by target images acquired by a plurality of image acquisition devices of a target vehicle traveling on a target road surface, and the states of the plurality of pavement damage parameters include at least one of the following: parameters in an unmarked state, parameters marked as a waiting-for-confirmation state; classifying the plurality of pavement damage parameters according to the disease locations included in the pavement damage parameters to obtain a pavement damage subset, wherein the target disease locations of the target pavement disease parameters included in the pavement damage subset are within the same area; a second determining module, for determining whether a disease of the target disease type exists in the target road surface based on the target pavement damage parameters and the first probability.

[0019] According to yet another embodiment of the present invention, a computer-readable storage medium is provided, in which a computer program is stored, wherein the computer program is configured to execute the steps of any one of the above method embodiments when run.

[0020] According to yet another embodiment of the present invention, there is provided an electronic device, including a memory and a processor, wherein the memory stores a computer program, and the processor is configured to run the computer program to execute the steps in any one of the above method embodiments.

[0021] According to yet another embodiment of the present invention, a computer program product is provided, including a computer program, and when the computer program is executed by a processor, the steps of the method described in each embodiment of the present application are implemented.

[0022] Through the present invention, the pavement disease parameters are determined according to the target image collected by the target vehicle during the driving process on the target road surface, and the pavement diseases are classified according to the disease positions in the pavement disease parameters, so that multiple pavement disease subsets can be obtained. By determining the target disease type existing in each pavement disease subset and the first probability of the target disease type corresponding to the disease, when the first probability is greater than the first preset probability, the pavement disease subset is sent to the target platform for confirmation, and the target platform can judge whether there is a corresponding target type of disease in the target road surface according to the target pavement disease parameters and the first probability. Therefore, the technical problems of low road inspection efficiency and low accuracy in the related technology can be solved, and the effect of improving the efficiency and accuracy of road inspection can be achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 is a hardware structure block diagram of a mobile terminal of a method for determining road surface damage according to an embodiment of the present invention;

[0024] Figure 2 The process of the method for determining pavement damage according to an embodiment of the present invention is as follows: Figure 1 ;

[0025] Figure 3 is a schematic diagram of the location of a target vehicle acquisition device according to an embodiment of the present invention;

[0026] Figure 4 The process of the method for determining pavement damage according to an embodiment of the present invention is as follows: Figure 2 ;

[0027] Figure 5 is an overall flow chart of a method for determining pavement damage according to an embodiment of the present invention;

[0028] Figure 6The structure frame of the device for determining road surface damage according to an embodiment of the present invention is Figure 1 ;

[0029] Figure 7 The structure frame of the device for determining road surface damage according to an embodiment of the present invention is Figure 2 . DETAILED DESCRIPTION

[0030] Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings and in combination with the embodiments.

[0031] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence.

[0032] The method embodiments provided in the embodiments of the present application can be executed in a mobile terminal, a computer terminal or a similar computing device. Taking running on a mobile terminal as an example, Figure 1 FIG. 1 is a hardware structure block diagram of a mobile terminal of a method for determining road surface defects according to an embodiment of the present invention. Figure 1 As shown, the mobile terminal may include one or more ( Figure 1 Only one is shown in the figure) a processor 102 (the processor 102 may include but is not limited to a processing device such as a microprocessor MCU or a programmable logic device FPGA) and a memory 104 for storing data, wherein the mobile terminal may also include a transmission device 106 and an input / output device 108 for communication functions. It can be understood by those skilled in the art that Figure 1 The structure shown is only for illustration and does not limit the structure of the mobile terminal. Figure 1 More or fewer components as shown, or with Figure 1 Different configurations shown.

[0033] The memory 104 can be used to store computer programs, for example, software programs and modules of application software, such as the computer program corresponding to the method for determining road surface diseases in the embodiment of the present invention. The processor 102 executes various functional applications and data processing by running the computer program stored in the memory 104, that is, to implement the above method. The memory 104 may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, the memory 104 may further include a memory remotely arranged relative to the processor 102, and these remote memories may be connected to the mobile terminal via a network. Examples of the above network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0034] The transmission device 106 is used to receive or send data via a network. The specific example of the above network may include a wireless network provided by a communication provider of the mobile terminal. In one example, the transmission device 106 includes a network adapter (Network Interface Controller, referred to as NIC), which can be connected to other network devices through a base station so as to communicate with the Internet. In one example, the transmission device 106 can be a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0035] In this embodiment, a method for determining road surface damage is provided. Figure 2 The process of the method for determining pavement damage according to an embodiment of the present invention is as follows: Figure 1 ,like Figure 2 As shown, the process includes the following steps:

[0036] Step S202, obtaining a plurality of pavement disease parameters, wherein the pavement disease parameters are parameters determined by target images acquired by a plurality of image acquisition devices of a target vehicle traveling on a target road, and the states of the plurality of pavement disease parameters include at least one of the following: parameters in an unmarked state, and parameters marked as a waiting-for-confirmation state;

[0037] Step S204, obtaining a plurality of pavement disease parameters, wherein the pavement disease parameters are parameters determined by target images acquired by a plurality of image acquisition devices of a target vehicle traveling on a target road, and the plurality of pavement disease parameters are at least one of the following: parameters in an unmarked state, and parameters marked as waiting for confirmation;

[0038] Step S206, determining a target disease type existing in the pavement disease subset, and determining a first probability that a disease corresponding to the target disease type exists in the pavement disease subset;

[0039] Step S208, when the first probability is greater than a first preset probability, the pavement defect subset is sent to a target platform to instruct the target platform to determine whether a defect of the target defect type exists in the target pavement based on the target pavement defect parameters and the first probability.

[0040] In the above embodiment, the acquisition device of the target vehicle can be a driving recorder or a vehicle-mounted camera, such as Figure 3 As shown, when the target vehicle is driving, the dashcam on the front window and the onboard cameras at various positions on the vehicle body (i.e., Figure 3The camera 1 located at the front window and the cameras 2-5 located at both sides and the rear of the vehicle as shown in the figure can collect target road surface data, wherein the data is a target image of the target road surface collected from multiple directions. By utilizing the social vehicle group to assist in road detection, there is no need to set up additional equipment, the cost is low, and the source of its image data is multi-dimensional, which is more conducive to the analysis and detection of road diseases to a certain extent.

[0041] In the above embodiment, the acquired target image data is serialized to obtain the pavement disease parameters of the target road surface, as shown in Table 1. The pavement disease parameters may include: the label of the target image, the capture time of the target image, the license plate number and vehicle type of the captured vehicle, the BSM (Blind Spot Monitoring) information of the captured vehicle, the position of the captured camera on the vehicle, the disease characteristic value of the target road surface, the disease type, the length and width of the disease, the location of the disease, and the confidence (indicating the authenticity of the data), etc. Among them, BSM can be understood as a blind spot detection system, which continuously detects the surrounding blind spots through radar sensors arranged near the rearview mirrors on both sides of the vehicle. When a vehicle or obstacle is detected entering the driver's blind spot, the BSM system will send a signal to the receiver installed on the edge of the rearview mirror to ensure driving safety. Therefore, the captured vehicle BSM in the pavement disease parameters may include: the GPS driving position information of the target vehicle, the acceleration of the three axes, the braking status, and the steering wheel angle, etc.

[0042] Table 1

[0043]

[0044] In the above embodiment, the road surface disease parameter can be a parameter confirmed by the target vehicle, or confirmed by the on-board computer of the target vehicle, or sent by the target vehicle to the car server for confirmation, but not limited thereto. After the collected video data is preprocessed, the road disease detection algorithm can be used to output information such as disease feature values, road disease types, road disease severity, disease width, and disease length. Among them, the road disease detection algorithm can be a road disease detection method based on attention mechanism and feature integration or a road disease detection method based on Transformer, but not limited thereto.

[0045] In the above embodiment, the pavement disease parameters collected by the target vehicle may be analyzed by the vehicle-road cooperative roadside equipment, and the disease position in the pavement disease parameters may be determined by longitude and latitude or based on an offset at a specific point. According to the vehicle longitude and latitude information output by the n target vehicles, the coordinates of the disease pixels in the n target road images are converted into three-dimensional space, and the longitude and latitude information of n road diseases can be obtained by combining the two. Among them, n can be 1, 2, 3..., but not limited to this. When the longitude and latitude information of the road disease and the longitude and latitude information of the roadside equipment meet the preset deviation range θ, it can be confirmed that the disease position is consistent, and the longitude and latitude deviation range can be calculated by the following formula: (X i , Y i ) - (X j , Y j ) ≤(±θ, ±θ), where θ can be 10°55' or 5°1', but is not limited thereto. The latitude and longitude information can be in degree, minute, and second format, for example, 57°55'56.6". (X i , Y i ) represents the latitude and longitude coordinates of the target vehicle, (X j , Y j ) represents the longitude and latitude coordinates of the roadside equipment, 1≤i, i represents the number of the roadside equipment. The roadside equipment receives paired data information uploaded by different target vehicles, which can constitute a large data set. The data set to be clustered can be divided into m sub-data sets (i.e., the above-mentioned road surface disease subsets) according to the location of the disease, that is, corresponding to m≤n different disaster locations. For the newly added road data, the location judgment is performed in turn and divided into the corresponding sub-data sets, where m can be 1, 2, 3..., m≤n, but not limited to this. The acceptance information obtained by the roadside equipment may include road disaster information and BSM information, etc. When the location of the road disaster is consistent, the road disease can be recorded as the same road disaster, and the target disease type of the road disaster can be judged for road images within a certain range.

[0046] In the above embodiment, the target vehicle can send the road disease parameters to the nearest vehicle-road cooperative roadside equipment. Since the longitude and latitude information of the roadside equipment and the vehicle are known, the k-nearest neighbor algorithm can be used to calculate the roadside equipment closest to the vehicle. The Euclidean distance can be used to calculate the distance. The formula is as follows: d = , where X represents longitude and Y represents latitude. By comparing the distances, the roadside equipment closest to the target vehicle can be selected for collaborative work.

[0047] In the above embodiment, since there are multiple types of diseases, the target disaster types output by each vehicle can be counted according to the road disease subset that has been divided according to the disease location, and the proportion results corresponding to different target disaster types can be calculated. The disaster type with the highest proportion can be confirmed as the target disaster type corresponding to the location of the road disease subset. When the first probability of the disease of all target vehicles with the same target disaster type as the output category is greater than the first preset probability, the results such as the disease type, disease severity, disease width and size of the road disease subset that determines the disaster type can be sent to the target platform, and the target platform can further determine whether there is a disease of the target disease type in the target road surface according to the target road disease parameters and the first probability. Among them, the first preset probability can be a floating point number between 0 and 1, and the target platform can be a cloud platform, but is not limited to this.

[0048] Through the present invention, the pavement disease parameters are determined according to the target image collected by the target vehicle during the driving process on the target road surface, and the pavement diseases are classified according to the disease positions in the pavement disease parameters, so that multiple pavement disease subsets can be obtained. By determining the target disease type existing in each pavement disease subset and the first probability of the target disease type corresponding to the disease, when the first probability is greater than the first preset probability, the pavement disease subset is sent to the target platform for confirmation, and the target platform can judge whether there is a corresponding target type of disease in the target road surface according to the target pavement disease parameters and the first probability. Therefore, the technical problems of low road inspection efficiency and low accuracy in the related technology can be solved, and the effect of improving the efficiency and accuracy of road inspection can be achieved.

[0049] Optionally, the execution entity of the above steps may be a roadside device, a roadside sensing device, a detection system including a roadside device, a service terminal including a roadside device, etc., but is not limited thereto.

[0050] In an exemplary embodiment, determining the target disease type existing in the pavement disease subset includes: determining the first disease type included in each of the target pavement disease parameters included in the pavement disease subset; determining the target proportion of each of the first disease types in all disease types included in the pavement disease subset; and determining the second disease type corresponding to the maximum proportion included in the target proportion as the target disease type.

[0051] In the above embodiment, each target vehicle in the road damage subset may output a different disaster type (i.e., the first damage type mentioned above), which can be represented by class1(data), class2(data), ..., class i (data) ,…,class m`(data) is used to represent the proportion of disaster types output by each vehicle in the road disease subset (i.e., the above target proportion). The proportion result can be calculated by the following formula: , that is, the proportion of disease type k = the number of disease types determined as k / the total number of different disaster types, and the disaster type class with the highest proportion k (i.e., the second disease type mentioned above) is determined as the target disease type corresponding to the road surface disease subset. For example, there are 50 target vehicles that output the disaster type of the target road surface, of which 45 target vehicles output the disaster type of potholes, and 5 target vehicles output the disaster type of cracks. If the proportion of potholes is greater than that of cracks, the target disaster type of this road surface disease subset can be determined as potholes. Among them, the results of 45 vehicles are marked as valid data, and the results of 5 vehicles are marked as invalid data. Invalid data can also include incorrect evaluation results in historical evaluation results.

[0052] In an exemplary embodiment, determining a first probability that the pavement disease subset contains a disease corresponding to the target disease type includes: determining a first disease parameter that the disease type included in the pavement disease subset is the target disease type; determining a disease probability included in each of the first disease parameters; and determining an average value of the disease probabilities as the first probability.

[0053] In the above embodiment, the defect probabilities of all target vehicles whose output categories are the target defect types can be determined to form a vector representation, such as the following formula: P(data,class)=[p1(data,class k ),p2(data,class k ),…,p i (data,class k ),…,p l (data,class k )], where p i (data,class k ) represents the probability that the road data is a road disease after model inference, l≤m≤n. The disease probability is normalized and can be expressed as , the normalized average value s(data,class k ) is determined as the first probability.

[0054] In an exemplary embodiment, after sending the pavement damage subset to a target platform, the method further includes: clearing the pavement damage subset.

[0055] In the above embodiment, after combining the consistency analysis of the disease location and the consistency analysis of the disease type, k≤m road disease results can be obtained, which are written into the list of disasters on the target road section, and the paired valid data are uploaded to the cloud platform for intelligent analysis again. In order to avoid the roadside equipment end from bearing the data pressure that is difficult to load, the uploaded data information also includes the disease feature values ​​extracted from each vehicle. After uploading to the cloud platform, the roadside equipment end clears the recorded paired valid data (that is, the pavement disease subset for which the target disease type has been determined).

[0056] In an exemplary embodiment, after determining a first probability that the pavement defect subset contains a defect corresponding to the target defect type, the method further includes: when the first probability is less than or equal to a first preset probability, marking the target pavement defect parameter as a parameter in a waiting for confirmation state.

[0057] In the above embodiment, when the first probability s(data, class k ) does not satisfy (i.e., is less than or equal to) the first preset probability, the target road surface disease parameter can be marked as a waiting for confirmation state. The parameters in the waiting for confirmation state will continue to change during the process of continuously receiving data through the roadside equipment. When the continuously changing first probability satisfies the first preset probability, the waiting for confirmation state can be changed to valid data for uploading to the cloud.

[0058] In this embodiment, a method for determining road surface damage is also provided. Figure 4 The process of the method for determining pavement damage according to an embodiment of the present invention is as follows: Figure 2 ,like Figure 4 As shown, the process includes the following steps:

[0059] Step S402, receiving a pavement disease subset sent by a roadside device, wherein the pavement disease subset is sent by the roadside device when it is determined that a first probability that a disease corresponding to a target disease type exists in the pavement disease subset is greater than a first preset probability, and the pavement disease subset is obtained by the roadside device in the following manner: acquiring a plurality of pavement disease parameters, wherein the pavement disease parameters are parameters determined by target images acquired by a plurality of image acquisition devices of a target vehicle traveling on a target road surface, and the states of the plurality of pavement disease parameters include at least one of the following: parameters in an unmarked state, parameters marked as waiting for confirmation; classifying the plurality of pavement disease parameters according to the disease locations included in the pavement disease parameters to obtain a pavement disease subset, wherein the target disease locations of the target pavement disease parameters included in the pavement disease subset are located within the same area;

[0060] Step S404: determining whether a disease of the target disease type exists in the target road surface based on the target road surface disease parameter and the first probability.

[0061] In the above embodiment, the acquisition device of the target vehicle can be a driving recorder or a vehicle-mounted camera, such as Figure 3 As shown, when the target vehicle is driving, the dashcam on the front window and the onboard cameras at various positions on the vehicle body (i.e., Figure 3 The camera 1 located at the front window and the cameras 2-5 located at both sides and the rear of the vehicle shown in the figure can collect target road surface data, wherein the data is a target image of the target road surface collected from multiple directions. The acquired target image data is serialized to obtain road surface disease parameters of the target road surface, as shown in Table 1, and the road surface disease parameters may include: the label of the target image, the capture time of the target image, the license plate number and vehicle type of the captured vehicle, the BSM (Blind Spot Monitoring) information of the captured vehicle, the position of the captured camera on the vehicle, the disease characteristic value of the target road surface, the disease type, the length and width of the disease, the location of the disease, and the confidence (indicating the authenticity of the data), etc. Among them, BSM can be understood as a blind spot detection system, which continuously detects the surrounding blind spots through radar sensors arranged near the rearview mirrors on both sides of the vehicle. When a vehicle or obstacle is detected to enter the driver's blind spot, the BSM system will send a signal to the receiver installed on the edge of the rearview mirror to ensure driving safety. Therefore, the captured vehicle BSM in the pavement disease parameters may include: GPS location information of the target vehicle, acceleration of three axes, braking status, steering wheel angle, etc.

[0062] In the above embodiment, the road surface disease parameter can be a parameter confirmed by the target vehicle, or confirmed by the on-board computer of the target vehicle, or sent by the target vehicle to the car server for confirmation, but not limited thereto. After the collected video data is preprocessed, the road disease detection algorithm can be used to output information such as disease feature values, road disease types, road disease severity, disease width, and disease length. Among them, the road disease detection algorithm can be a road disease detection method based on attention mechanism and feature integration or a road disease detection method based on Transformer, but not limited thereto.

[0063] In the above embodiment, the pavement disease parameters collected by the target vehicle may be analyzed by the vehicle-road cooperative roadside equipment, and the disease position in the pavement disease parameters may be determined by longitude and latitude or based on an offset at a specific point. According to the vehicle longitude and latitude information output by the n target vehicles, the coordinates of the disease pixels in the n target road images are converted into three-dimensional space, and the longitude and latitude information of n road diseases can be obtained by combining the two. Among them, n can be 1, 2, 3..., but not limited to this. When the longitude and latitude information of the road disease and the longitude and latitude information of the roadside equipment meet the preset deviation range θ, it can be confirmed that the disease position is consistent, and the longitude and latitude deviation range can be calculated by the following formula: (X i , Y i ) - (X j , Y j ) ≤(±θ, ±θ), where θ can be 10°55' or 5°1', but is not limited thereto. The latitude and longitude information can be in degree, minute, and second format, for example, 57°55'56.6". (X i , Y i ) represents the latitude and longitude coordinates of the target vehicle, (X j , Y j ) represents the longitude and latitude coordinates of the roadside equipment, 1≤i, i represents the number of the roadside equipment. The roadside equipment receives paired data information uploaded by different target vehicles, which can constitute a large data set. The data set to be clustered can be divided into m sub-data sets (i.e., the above-mentioned road surface disease subsets) according to the location of the disease, that is, corresponding to m≤n different disaster locations. For the newly added road data, the location judgment is performed in turn and divided into the corresponding sub-data sets, where m can be 1, 2, 3..., m≤n, but not limited to this. The information obtained by the roadside equipment may include road disaster information and BSM information. When the location of the road disaster is consistent, the road disease can be recorded as the same road disaster, and the target disease type of the road disaster can be judged for road images within a certain range.

[0064] In the above embodiment, since there are multiple types of diseases, the target disaster types output by each vehicle can be counted according to the road disease subsets that have been divided according to the disease positions, and the proportions corresponding to different target disaster types can be calculated. The disaster type with the highest proportion can be confirmed as the target disaster type corresponding to the position of the road disease subset. When the first probability of the disease of all target vehicles with the same target disaster type as the output category is greater than the first preset probability, the results such as the disease type, disease severity, disease width and size of the road disease subset of the determined disaster type can be sent to the target platform, and the target platform can further determine whether there is a disease of the target disease type in the target road surface according to the target road disease parameters and the first probability.

[0065] Through the present invention, a road surface disease subset sent by a roadside device is received, and it can be determined whether a target disease type exists in the target road surface according to the target road surface disease parameter and the first probability. The road surface disease subset is obtained by classifying the disease position in the disease parameter determined by the roadside device from the target image collected by the target vehicle, and is sent by the roadside device when it is determined that the first probability of the disease corresponding to the target disease type existing in the road surface disease subset is greater than the first preset probability. Therefore, the problem of low road inspection efficiency and low accuracy can be solved, and the effect of improving road inspection efficiency and accuracy is achieved.

[0066] Optionally, the execution entity of the above steps may be a cloud platform, a traffic management system terminal, a cloud storage detection system, etc., but is not limited thereto.

[0067] In an exemplary embodiment, determining whether the target road surface has a disease of the target disease type based on the target road surface disease parameter and the first probability includes: determining a target basic safety message included in the target road surface disease parameter at the target moment when the target vehicle collects the target image; determining a first basic safety message of the target vehicle in a first time period before the target moment and adjacent to the target moment; determining a second basic safety message of the target vehicle in a second time period after the target moment and adjacent to the target moment; determining a second probability based on the target basic safety message, the first basic safety message, and the second basic safety message; determining a target probability based on the first probability and the second probability; and determining that the target road surface has the target disease type when the target probability is greater than a second preset probability.

[0068] In the above embodiment, the BSM information at the time t when the defect exists (i.e., the above target basic safety information) and the BSM information in the time period (t-nq, … t-2q, tq , t, t+q, t+2q …, t+nq) (i.e., the above first basic safety message and the second basic safety message) are determined respectively. The second probability that the BSM information is abnormal data can be determined based on the target basic safety message, the first basic safety message, and the second basic safety message. The probability of the final road defect type (i.e., the above target probability) can be determined by weighting the first probability and the second probability. When the target probability is greater than the second preset probability, it can be determined that the target road surface has the target defect type.

[0069] In an exemplary embodiment, determining the second probability based on the target basic safety message, the first basic safety message, and the second basic safety message includes: for each target sub-message of the target type included in the target basic safety message, performing the following operations to determine the average deviation value of the target sub-message, wherein the target type is the type of any sub-message included in the target basic safety message: determining a first sub-message of the target type included in the first basic safety message, determining a second sub-message of the target type included in the second basic safety message, determining the mean square error of the target sub-message, the first sub-message, and the second sub-message, and determining the mean square error as the average deviation value; determining a first weight corresponding to each average deviation value; determining a weighted sum of the average deviation value and the first weight; and determining the objective function value of the weighted sum as the second probability.

[0070] In the above embodiment, the deviation values ​​in the target basic safety message, the first basic safety message and the second basic safety message (i.e. the above target sub-message, the first sub-message and the second sub-message) are determined respectively, and the average deviation value of the BSM information is calculated. The average deviation value can be calculated by the following formula: , where q / s,n is the value obtained through training, n can represent the number of segmented BSM data, x represents the influencing factors, which can include three-axis acceleration, brake status, steering wheel angle, vehicle speed, etc., and δ represents the average deviation value. The cloud platform uses the historically collected BSM data sets, including BSM data sets when the disease exists and when the disease does not exist, to divide them into training sets and validation sets for deep model training and validation, and can obtain all model parameters including n,q. The algorithm can be integrated in the cloud.

[0071] In the above embodiment, the cloud platform analyzes the uploaded k≤m pavement disease subsets respectively, and the analysis process can mainly analyze the BSM information within each pavement disease subset. In order to balance the impact of different vehicles and obtain better prediction results, each BSM information is analyzed, and finally the BSM information of the pavement disease subset can be confirmed by weighted method. The BSM information can refer to Table 2, as shown in the following formula: , where the weighting factors of three-axis acceleration, braking status, steering wheel angle, vehicle speed, etc. are , the deviation value vector can be expressed as .

[0072] Table 2

[0073]

[0074] In the above embodiment, the probability that the BSM is abnormal data (ie, the above second probability) can be calculated using the sigmoid nonlinear function: , where the sigmoid function value is a value between [0,1]. The closer the function value is to 1, the higher the probability that the BSM data of the pavement disease subset is abnormal, that is, the higher the probability that the road state is abnormal.

[0075] In an exemplary embodiment, determining the target probability based on the first probability and the second probability includes: determining a second weight corresponding to the first probability, and determining a third weight corresponding to the second probability; determining a first sum of the first probability theory and the second weight, and determining a second sum of the second probability and the third weight; and determining the sum of the first sum and the second sum as the target probability.

[0076] In the above embodiment, the target vehicle BSM information in each road surface disease subset is fused with the road disease detection result, which can be calculated by the following formula: P=λs+(1-λ)p, where λ represents the second weight, which is between 0 and 1; (1-λ) represents the third weight. The weighted P value is determined as the probability of the final road disease type (i.e., the above target probability). The condition of the disease can also be determined based on the P value, which can include dimensions such as road disease type, road disease length and width, repair urgency, repair difficulty, and road disaster risk level.

[0077] In an exemplary embodiment, after determining the target probability based on the first probability and the second probability, the method further includes: when the target probability is less than or equal to the second preset probability, marking the state of the pavement defect subset as a waiting for confirmation state.

[0078] In the above embodiment, when the target probability P does not meet (that is, is less than or equal to) the preset threshold, the status of this road disease subset is marked as waiting for confirmation. For example, the probability that the fused road disease type is determined to be a pothole is greater than 0.85 (preset threshold), and combined with other BSM information, it is comprehensively determined that the disaster risk level of the road is high and the urgency of repair is high. At this time, the road disease management and control platform can send it to the person in charge of road maintenance for final confirmation. After the maintenance personnel determine the type of road disease and whether the road disease exists, the "abnormal" label of the BSM data can be obtained. When the probability of the BSM abnormal data is 1, it indicates that the road disease exists; when the probability of the BSM abnormal data is 0, it indicates that the road disease does not exist.

[0079] In an exemplary embodiment, before determining whether the target pavement has a disease of the target disease type in the target pavement based on the target pavement disease parameters and the first probability, the method further includes: determining a first pavement disease subset that is in a waiting for confirmation state; in the case of a second disease subset whose disease locations included in the first pavement disease subset are the target disease locations, determining supplementary pavement disease parameters of the second disease subset; and supplementing the supplementary pavement disease parameters to the target pavement disease parameters.

[0080] In the above embodiment, when the defect position in a certain road defect subset (i.e., the first road defect subset) marked as waiting for confirmation is a certain defect subset of the target defect position, the supplementary road defect parameters of the second defect subset can be added to the target road defect parameters, that is, the first road defect subset marked as waiting for confirmation will be updated when it continuously receives new data through the roadside data, and when it is determined to belong to a certain target defect position, the data will be added to the target defect parameters. Therefore, in the case of continuous data update, some road defect subsets in the waiting for confirmation state can also be updated to valid data, that is, it is confirmed that the road defect exists.

[0081] In an exemplary embodiment, the overall process of the method for determining pavement damage can be referred to Figure 5 ,like Figure 5 As shown, the process includes:

[0082] Step S502, target vehicle road image acquisition;

[0083] Step S504, the target vehicle performs road video analysis;

[0084] Step S506, the target vehicle determines whether there is a road disease, if yes, executes step S508, if no, executes step S520;

[0085] Step S508, the target vehicle uploads the video to the roadside equipment;

[0086] Step S510, road damage consistency analysis of roadside equipment;

[0087] Step S512, determining whether the road damage data is in the "waiting for confirmation" state, if yes, executing step S506, if no, executing step S514;

[0088] Step S514, the roadside equipment uploads the road surface image and vehicle BMS information to the cloud platform for intelligent analysis;

[0089] Step S516, the cloud platform determines whether the road damage data is in the "waiting for confirmation" state, if yes, execute step S506, if no, execute step S518;

[0090] Step S518, the cloud platform outputs the disease results;

[0091] Step S520, end.

[0092] In the above embodiment, road graphics are obtained through social vehicles, and preliminary disease detection can be completed using the driving recorder configured in the vehicle or the road disease monitoring system deployed in the vehicle-mounted camera, so as to effectively detect road diseases and realize the effective use of social resources. When the existence of road diseases is detected, the disease data is transmitted to the roadside equipment. When it is determined that the road disease does not exist, it is directly terminated. It is only uploaded when it is determined that the disease exists, so as to avoid large-scale packaging and uploading of videos that have not been "filtered" by feature extraction and accumulation in the cloud, causing excessive delays in data transmission. The roadside equipment obtains and stores the road disease information collected by each vehicle in the current range, analyzes the consistency of the disease location and disease type, and obtains detailed disease conditions within the control range of the roadside equipment. The information obtained may include: disease type, disease severity, disease size, disease width, etc. The data set judged as awaiting confirmation will be retained, and the data set judged as indeed having road damage will be sent to the cloud platform. The cloud platform will conduct a secondary analysis based on the vehicle's BMS information and the uploaded road damage data set to generate a probability of an "abnormal" state, which is used to determine whether there is an abnormality in the road state. The data set judged as awaiting confirmation will be retained, and the data set judged as indeed having road damage will be analyzed. This will allow the high-accuracy road damage situation to be obtained, including the location and severity of the damage, to be automatically detected and recorded, wherein the weight of each factor can be obtained through deep learning training.

[0093] In the above embodiment, the use of roadside equipment to assist in information analysis and transmission can solve the problem of data transmission delay or loss caused by uploading a large amount of data to the cloud platform at the same time, and reduce the pressure on back-end data storage and analysis. In addition, based on machine vision analysis and combined with the vehicle's driving status information for analysis, the relevant parts can be synchronized in time after the road disease analysis results are determined, which greatly improves the timeliness and accuracy of detection.

[0094] Through the description of the above implementation methods, those skilled in the art can clearly understand that the method according to the above embodiment can be implemented by means of software plus a necessary general hardware platform, and of course can also be implemented by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes a number of instructions for enabling a terminal device (which can be a mobile phone, computer, server, or network device, etc.) to execute the methods described in each embodiment of the present invention.

[0095] In this embodiment, a device for determining road surface defects is also provided, which is used to implement the above-mentioned embodiments and preferred implementation modes, and the descriptions that have been made will not be repeated. As used below, the term "module" can be a combination of software and / or hardware that implements a predetermined function. Although the device described in the following embodiments is preferably implemented in software, the implementation of hardware, or a combination of software and hardware, is also possible and conceivable.

[0096] Figure 6 The structure frame of the device for determining road surface damage according to an embodiment of the present invention is Figure 1 ,like Figure 6 As shown, the device comprises:

[0097] An acquisition module 62 is used to acquire a plurality of pavement disease parameters, wherein the pavement disease parameters are parameters determined by target images acquired by a plurality of image acquisition devices of a target vehicle traveling on a target road, and the states of the plurality of pavement disease parameters include at least one of the following: parameters in an unmarked state, and parameters marked as a waiting-for-confirmation state;

[0098] A classification module 64 is used to classify the plurality of pavement disease parameters according to the disease locations included in the pavement disease parameters to obtain a pavement disease subset, wherein the target disease locations of the target pavement disease parameters included in the pavement disease subset are located within the same area;

[0099] A first determination module 66 is used to determine a target disease type existing in the pavement disease subset, and to determine a first probability that a disease corresponding to the target disease type exists in the pavement disease subset;

[0100] The sending module 68 sends the pavement disease subset to the target platform when the first probability is greater than the first preset probability, so as to instruct the target platform to determine whether the target disease type exists in the target pavement based on the target pavement disease parameters and the first probability.

[0101] In an exemplary embodiment, the first determination module 66 includes: a first determination unit for determining the first disease type included in each of the target pavement disease parameters included in the pavement disease subset; a second determination unit for determining the target proportion of each of the first disease types in all disease types included in the pavement disease subset; and a third determination unit for determining the second disease type corresponding to the maximum proportion included in the target proportion as the target disease type.

[0102] In an exemplary embodiment, the first determination module 66 includes: a fourth determination unit, used to determine that the disease type included in the pavement disease subset is a first disease parameter of the target disease type; a fifth determination unit, used to determine the disease probability included in each of the first disease parameters; and a sixth determination unit, used to determine the average value of the disease probabilities as the first probability.

[0103] In an exemplary embodiment, the apparatus further includes: a clearing module for clearing the pavement defect subset after sending the pavement defect subset to a target platform.

[0104] In an exemplary embodiment, the device also includes: a first marking module, which is used to mark the target pavement disease parameter as a parameter in a waiting for confirmation state after determining a first probability that the pavement disease subset contains a disease corresponding to the target disease type, if the first probability is less than or equal to a first preset probability.

[0105] Figure 7 The structure frame of the device for determining road surface damage according to an embodiment of the present invention is Figure 2 ,like Figure 7 As shown, the device comprises:

[0106] The receiving module 72 is used to receive a pavement disease subset sent by a roadside device, wherein the pavement disease subset is sent by the roadside device when it is determined that the first probability that a disease corresponding to a target disease type exists in the pavement disease subset is greater than a first preset probability, and the pavement disease subset is obtained by the roadside device in the following manner: acquiring a plurality of pavement disease parameters, wherein the pavement disease parameters are parameters determined by target images acquired by a plurality of image acquisition devices of a target vehicle traveling on a target road surface, and the states of the plurality of pavement disease parameters include at least one of the following: parameters in an unmarked state, parameters marked as waiting for confirmation; classifying the plurality of pavement disease parameters according to the disease locations included in the pavement disease parameters to obtain a pavement disease subset, wherein the target disease locations of the target pavement disease parameters included in the pavement disease subset are located within the same area;

[0107] The second determination module 74 is configured to determine whether a disease of the target disease type exists in the target road surface based on the target road surface disease parameter and the first probability.

[0108] In an exemplary embodiment, the second determination module 74 includes: a seventh determination unit, which is used to determine the target basic safety message of the target vehicle at the target moment when the target image is collected, which is included in the target road surface disease parameter; an eighth determination unit, which is used to determine the first basic safety message of the target vehicle in a first time period before the target moment and adjacent to the target moment; a ninth determination unit, which is used to determine the second basic safety message of the target vehicle in a second time period after the target moment and adjacent to the target moment; a tenth determination unit, which is used to determine a second probability based on the target basic safety message, the first basic safety message and the second basic safety message; an eleventh determination unit, which is used to determine a target probability based on the first probability and the second probability; and a twelfth determination unit, which is used to determine that the target road surface has the target disease type when the target probability is greater than a second preset probability.

[0109] In an exemplary embodiment, the tenth determination unit includes: a first determination subunit, used to perform the following operations for each target sub-message of the target type included in the target basic safety message to determine the average deviation value of the target sub-message, wherein the target type is the type of any sub-message included in the target basic safety message: determine the first sub-message of the target type included in the first basic safety message, determine the second sub-message of the target type included in the second basic safety message, determine the mean square error of the target sub-message, the first sub-message and the second sub-message, and determine the mean square error as the average deviation value; a second determination subunit, used to determine the first weight corresponding to each of the average deviation values; a third determination subunit, used to determine the weighted sum of the average deviation value and the first weight; and a fourth determination subunit, used to determine the objective function value of the weighted sum value as the second probability.

[0110] In an exemplary embodiment, the eleventh determination unit includes: a fifth determination subunit, used to determine the second weight corresponding to the first probability, and to determine the third weight corresponding to the second probability; a sixth determination subunit, used to determine the first sum of the first probability theory and the second weight, and to determine the second sum of the second probability and the third weight; and a seventh determination subunit, used to determine the sum of the first sum and the second sum as the target probability.

[0111] In an exemplary embodiment, the device also includes: a second marking module, which is used to mark the status of the pavement disease subset as a waiting for confirmation status after determining the target probability based on the first probability and the second probability, if the target probability is less than or equal to the second preset probability.

[0112] In an exemplary embodiment, the device also includes: a third determination module, used to determine a first pavement disease subset that is in a waiting for confirmation state before determining whether the target disease type of disease exists in the target pavement based on the target pavement disease parameters and the first probability; a fourth determination module, used to determine the supplementary pavement disease parameters of the second disease subset when the disease location included in the first pavement disease subset is a second disease subset of the target disease location; and a supplement module, used to supplement the supplementary pavement disease parameters to the target pavement disease parameters.

[0113] It should be noted that the above modules can be implemented by software or hardware. For the latter, it can be implemented in the following ways, but not limited to: the above modules are all located in the same processor; or the above modules are located in different processors in any combination.

[0114] An embodiment of the present invention further provides a computer-readable storage medium, in which a computer program is stored, wherein the computer program is configured to execute the steps of any of the above method embodiments when running.

[0115] In an exemplary embodiment, the computer-readable storage medium may include, but is not limited to, various media that can store computer programs, such as a USB flash drive, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk or an optical disk.

[0116] An embodiment of the present invention further provides an electronic device, including a memory and a processor, wherein a computer program is stored in the memory, and the processor is configured to run the computer program to execute the steps in any one of the above method embodiments.

[0117] In an exemplary embodiment, the electronic device may further include a transmission device and an input / output device, wherein the transmission device is connected to the processor, and the input / output device is connected to the processor.

[0118] An embodiment of the present invention further provides a computer program product, including a computer program, which implements the steps of the method in each embodiment of the present application when the computer program is executed by a processor.

[0119] For specific examples in this embodiment, reference may be made to the examples described in the above embodiments and exemplary implementation modes, and this embodiment will not be described in detail herein.

[0120] Obviously, those skilled in the art should understand that the above modules or steps of the present invention can be implemented by a general computing device, they can be concentrated on a single computing device, or distributed on a network composed of multiple computing devices, they can be implemented by a program code executable by a computing device, so that they can be stored in a storage device and executed by the computing device, and in some cases, the steps shown or described can be executed in a different order than here, or they can be made into individual integrated circuit modules, or multiple modules or steps therein can be made into a single integrated circuit module for implementation. Thus, the present invention is not limited to any specific combination of hardware and software.

[0121] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for determining pavement damage, characterized in that: include: Acquire a plurality of pavement disease parameters, wherein the pavement disease parameters are parameters determined by target images acquired by a plurality of image acquisition devices of a target vehicle traveling on a target road, and the states of the plurality of pavement disease parameters include at least one of the following: parameters in an unmarked state, and parameters marked as a waiting-for-confirmation state; Classifying a plurality of the pavement disease parameters according to the disease locations included in the pavement disease parameters to obtain a pavement disease subset, wherein the target disease locations of the target pavement disease parameters included in the pavement disease subset are located within the same area; Determining a target disease type existing in the pavement disease subset, and determining a first probability that a disease corresponding to the target disease type exists in the pavement disease subset; When the first probability is greater than a first preset probability, the pavement defect subset is sent to a target platform to instruct the target platform to determine whether there is a defect of the target defect type in the target pavement based on the target pavement defect parameters and the first probability.

2. The method according to claim 1, characterized in that: The target disease types for determining the presence of the pavement disease subset include: Determining a first disease type included in each of the target pavement disease parameters included in the pavement disease subset; Determine a target proportion of each of the first damage types among all the damage types included in the pavement damage subset; The second disease type corresponding to the maximum proportion included in the target proportion is determined as the target disease type.

3. The method according to claim 1, characterized in that Determining the first probability that the pavement defect subset has a defect corresponding to the target defect type includes: Determining that the disease type included in the pavement disease subset is a first disease parameter of the target disease type; determining a disease probability included in each of the first disease parameters; An average value of the disease probabilities is determined as the first probability.

4. The method according to claim 1, characterized in that: After sending the pavement disease subset to the target platform, the method further includes: The subset of pavement defects is removed.

5. The method according to claim 1, characterized in that After determining the first probability that the pavement defect subset has a defect corresponding to the target defect type, the method further includes: When the first probability is less than or equal to a first preset probability, the target pavement damage parameter is marked as a parameter in a waiting for confirmation state.

6. A method for determining pavement damage, characterized in that: include: Receive a pavement disease subset sent by a roadside device, wherein the pavement disease subset is sent by the roadside device when it is determined that a first probability that a disease corresponding to a target disease type exists in the pavement disease subset is greater than a first preset probability, and the pavement disease subset is obtained by the roadside device in the following manner: acquiring a plurality of pavement disease parameters, wherein the pavement disease parameters are parameters determined by target images acquired by a plurality of image acquisition devices of a target vehicle traveling on a target road surface, and the states of the plurality of pavement disease parameters include at least one of the following: parameters in an unmarked state, parameters marked as a waiting-for-confirmation state; classifying the plurality of pavement disease parameters according to the disease locations included in the pavement disease parameters to obtain a pavement disease subset, wherein the target disease locations of the target pavement disease parameters included in the pavement disease subset are located within the same area; Based on the target road surface disease parameter and the first probability, it is determined whether there is a disease of the target disease type in the target road surface.

7. The method according to claim 6, characterized in that Determining whether there is a disease of the target disease type in the target road surface based on the target road surface disease parameter and the first probability includes: Determine the target basic safety information included in the target road surface disease parameter at the target time when the target vehicle acquires the target image; Determining a first basic safety message of the target vehicle within a first time period before the target time and adjacent to the target time; Determining a second basic safety message of the target vehicle within a second time period after the target time and adjacent to the target time; determining a second probability based on the target basic safety message, the first basic safety message, and the second basic safety message; determining a target probability based on the first probability and the second probability; When the target probability is greater than a second preset probability, it is determined that the target road surface has the target disease type.

8. The method according to claim 7, characterized in that Determining a second probability based on the target basic safety message, the first basic safety message, and the second basic safety message includes: For each target sub-message of a target type included in the target basic safety message, the following operations are performed to determine an average deviation value of the target sub-message, wherein the target type is the type of any sub-message included in the target basic safety message: determining a first sub-message of the target type included in the first basic safety message, determining a second sub-message of the target type included in the second basic safety message, determining a mean square deviation of the target sub-message, the first sub-message, and the second sub-message, and determining the mean square deviation as the average deviation value; Determine a first weight corresponding to each of the average deviation values; determining a weighted sum of the average deviation value and the first weight; An objective function value of the weighted sum value is determined as the second probability.

9. The method according to claim 7, characterized in that: Determining the target probability based on the first probability and the second probability includes: Determine a second weight corresponding to the first probability, and determine a third weight corresponding to the second probability; determining a first sum of the first probability and the second weight, and determining a second sum of the second probability and the third weight; The sum of the first sum and the second sum is determined as the target probability.

10. The method according to claim 7, characterized in that After determining the target probability based on the first probability and the second probability, the method further includes: When the target probability is less than or equal to the second preset probability, the state of the pavement defect subset is marked as a waiting for confirmation state.

11. The method according to claim 6, characterized in that Before determining whether the target road surface has a disease of the target disease type based on the target road surface disease parameter and the first probability, the method further includes: Determine the first pavement disease subset whose status is waiting for confirmation; In a case where the damage location included in the first pavement damage subset is a second damage subset of the target damage location, determining a supplementary pavement damage parameter of the second damage subset; The supplementary pavement damage parameters are added to the target pavement damage parameters.

12. A device for determining road surface defects, characterized in that: include: An acquisition module is used to acquire a plurality of pavement disease parameters, wherein the pavement disease parameters are parameters determined by target images acquired by a plurality of image acquisition devices of a target vehicle traveling on a target road, and the states of the plurality of pavement disease parameters include at least one of the following: parameters in an unmarked state, and parameters marked as a waiting-for-confirmation state; A classification module, used for classifying a plurality of the pavement disease parameters according to the disease locations included in the pavement disease parameters to obtain a pavement disease subset, wherein the target disease locations of the target pavement disease parameters included in the pavement disease subset are located within the same area; A first determination module is used to determine a target disease type existing in the pavement disease subset, and to determine a first probability that a disease corresponding to the target disease type exists in the pavement disease subset; A sending module is configured to send the pavement disease subset to a target platform when the first probability is greater than a first preset probability, so as to instruct the target platform to determine whether a disease of the target disease type exists in the target pavement based on the target pavement disease parameters and the first probability.

13. A device for determining road surface defects, characterized in that: include: A receiving module, for receiving a pavement disease subset sent by a roadside device, wherein the pavement disease subset is sent by the roadside device when it is determined that a first probability that a disease corresponding to a target disease type exists in the pavement disease subset is greater than a first preset probability, and the pavement disease subset is obtained by the roadside device in the following manner: acquiring a plurality of pavement disease parameters, wherein the pavement disease parameters are parameters determined by target images acquired by a plurality of image acquisition devices of a target vehicle traveling on a target road surface, and the states of the plurality of pavement disease parameters include at least one of the following: parameters in an unmarked state, parameters marked as a waiting-for-confirmation state; classifying the plurality of pavement disease parameters according to the disease locations included in the pavement disease parameters to obtain a pavement disease subset, wherein the target disease locations of the target pavement disease parameters included in the pavement disease subset are within the same area; The second determination module is used to determine whether the target road surface has a disease of the target disease type based on the target road surface disease parameter and the first probability.

14. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, wherein the computer program is configured to execute the steps of the method described in any one of claims 1 to 5, or execute the steps of the method described in any one of claims 6 to 11 when running.

15. An electronic device comprising a memory and a processor, characterized in that: A computer program is stored in the memory, and the processor is configured to run the computer program to execute the steps of the method described in any one of claims 1 to 5, or to execute the steps of the method described in any one of claims 6 to 11.

16. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the computer program implements the steps of the method described in any one of claims 1 to 5, or executes the steps of the method described in any one of claims 6 to 11.

Images