WO2021012504A1 - 基于智慧交通的道路信息提示方法、装置、服务器及介质 - Google Patents

基于智慧交通的道路信息提示方法、装置、服务器及介质 Download PDF

Info

Publication number
WO2021012504A1
WO2021012504A1 PCT/CN2019/117417 CN2019117417W WO2021012504A1 WO 2021012504 A1 WO2021012504 A1 WO 2021012504A1 CN 2019117417 W CN2019117417 W CN 2019117417W WO 2021012504 A1 WO2021012504 A1 WO 2021012504A1
Authority
WO
WIPO (PCT)
Prior art keywords
accident
level
preset
place
damage
Prior art date
Application number
PCT/CN2019/117417
Other languages
English (en)
French (fr)
Inventor
曾燕玲
杨晟
Original Assignee
平安科技（深圳）有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 平安科技（深圳）有限公司 filed Critical 平安科技（深圳）有限公司
Publication of WO2021012504A1 publication Critical patent/WO2021012504A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G08—SIGNALLING
    • G08G—TRAFFIC CONTROL SYSTEMS
    • G08G1/00—Traffic control systems for road vehicles
    • G08G1/09—Arrangements for giving variable traffic instructions
    • G08G1/0962—Arrangements for giving variable traffic instructions having an indicator mounted inside the vehicle, e.g. giving voice messages
    • G08G1/0967—Systems involving transmission of highway information, e.g. weather, speed limits
    • G—PHYSICS
    • G08—SIGNALLING
    • G08G—TRAFFIC CONTROL SYSTEMS
    • G08G1/00—Traffic control systems for road vehicles
    • G08G1/16—Anti-collision systems

Definitions

  • This application relates to the field of computer technology, in particular to a method, device, server and storage medium for prompting road information based on smart traffic.
  • the function of the vehicle safety early warning system or device is to remind the driver of the possible danger, so that the driver can be more vigilant and standardize the operation to achieve the purpose of avoiding danger.
  • some navigation devices or systems can provide warning functions for dangerous road sections. For example, when the vehicle is driving to a known dangerous road section, the system sends out some kind of warning message to remind the driver to slow down, which will greatly improve driving safety and Ensure the personal safety of drivers and passengers.
  • the driver can only be reminded based on whether the current road section is an accident-prone area and whether there is a rockfall.
  • a method for prompting road information based on smart transportation includes: acquiring auto insurance report information, wherein the auto insurance report information includes at least an auto insurance image and an accident location; and according to the damage level and location of the vehicle in the auto insurance image
  • the surrounding environment information of the accident location and/or the current number of accident occurrences in the accident location evaluate the risk level of the accident location; mark the accident location according to the risk level of the accident location to obtain a high risk Information on the location of the accident and information on the location of the low-risk accident; sending warning information including the location of the high-risk accident to relevant users.
  • a road information prompting device based on smart transportation includes: an acquisition module for acquiring auto insurance report information, wherein the auto insurance report information includes at least an auto insurance image and an accident location; The damage level of the vehicle in the car insurance image, the surrounding environment information of the accident place, and/or the current number of accidents in the accident place evaluate the risk level of the accident place; the marking module is used to evaluate the risk level of the accident place according to the accident The risk level of the place of occurrence marks the place of occurrence of the accident to obtain information on the place of occurrence of high-risk accidents and information of the place of occurrence of low-risk accidents; a sending module is used to send warning information including the place of occurrence of the high-risk accident to relevant users.
  • a server includes a processor and a memory, and the processor is configured to execute at least one computer-readable instruction stored in the memory to implement the following steps: obtain car insurance report information, wherein the car insurance report information includes at least a car insurance image And the place where the accident occurred; assess the risk level of the accident place according to the damage level of the vehicle in the car insurance image, the surrounding environment information of the accident place and/or the current number of accident occurrences in the accident place; The risk level of the accident occurrence location marks the accident occurrence location to obtain high-risk accident occurrence location information and low-risk accident occurrence location information; and send warning information including the high-risk accident occurrence location to relevant users.
  • a non-volatile readable storage medium that stores at least one computer readable instruction, and when the at least one computer readable instruction is executed by a processor, the following steps are implemented: Obtain a car insurance report Information, wherein the car insurance report information includes at least a car insurance image and the location of the accident; according to the damage level of the vehicle in the car insurance image, the surrounding environment information of the accident location and/or the current accident in the accident location Evaluate the risk level of the accident location based on the number of occurrences; mark the accident location based on the risk level of the accident location to obtain information on the location of high-risk accidents and information on the location of low-risk accidents; send information including the high-risk accident The warning information of the place of occurrence to the relevant user.
  • the intelligent transportation-based road information prompt method, device, server, and storage medium provided in this application evaluate the risk level of the accident site based on the car insurance image information, and mark the risk level of the accident site Send warning messages including locations of high-risk accidents to relevant users. It can not only count accident-prone areas, and remind users of the risk levels of the accident-prone areas based on the level of vehicle damage in the accident, so that users can get enough attention, but also remind users to avoid the high-risk accident-prone areas as much as possible when traveling. , Thereby reducing the occurrence of traffic accidents.
  • FIG. 1 is a flowchart of a method for prompting road information based on smart traffic according to Embodiment 1 of the present application.
  • FIG. 2 is a flowchart of a method for improving the clarity level of a car insurance image in a road information prompt method based on smart traffic provided in Embodiment 1 of the present application.
  • FIG. 3 is a flowchart of a method for determining the risk level of an accident location according to the vehicle damage level in the car insurance image in the road information prompt method based on smart traffic provided by Embodiment 1 of the present application.
  • Fig. 4 is a diagram of functional modules in a preferred embodiment of a road information prompting device based on smart traffic of the present application provided in the second embodiment of the present application.
  • FIG. 5 is a schematic diagram of a server provided in Embodiment 3 of the present application.
  • FIG. 1 is a flowchart of a method for prompting road information based on smart traffic according to Embodiment 1 of the present application. According to different needs, the execution order in this flowchart can be changed, and some steps can be omitted.
  • Step S1 Acquire car insurance report information, where the car insurance report information includes at least a car insurance image and an accident location.
  • the server may obtain report information from the mobile terminal.
  • the mobile terminal may be a smart terminal such as a mobile phone, a tablet computer, a personal digital assistant, a wearable device (for example, a smart watch, smart glasses), or any other suitable electronic device.
  • the report information may include the car insurance image and the place where the accident occurred, and may also include the owner's information, the license plate number of the dangerous vehicle, the time of the accident, and the reason for the accident.
  • the car insurance image may be video information or image information taken by a car owner.
  • the car insurance image may be video information or image information collected on site by an operator (such as a surveyor), and the operator sends the car insurance image to a database of another system (such as an insurance company system).
  • the server may obtain the car insurance image from the other system database.
  • the car insurance image is associated with the place where the accident occurred.
  • the car insurance image may include a general term for various graphics or images, usually referring to images with visual effects, and generally may include images on paper media, negatives or photos, televisions, projectors, or computer screens.
  • the car insurance image described in this embodiment may include computer image data stored on a readable storage medium after being captured by a camera or camera device, and may include various types of computer images such as vector graphics, bitmaps, static and dynamic images.
  • the smart traffic-based road information prompt method can also improve the clarity level of the car insurance image.
  • the method for improving the clarity level of the car insurance image includes:
  • the first sharpness level of the car insurance image can be calculated through a gray-scale change function, a gradient function, or an image gray-scale entropy function.
  • the gray-scale change function, the gradient function, or the image gray-scale entropy function are existing techniques for calculating image clarity, and will not be repeated here.
  • step S11 Compare whether the first definition level is lower than a preset definition level. When the first definition level is lower than the preset definition level, step S12 is executed; when the first definition level is higher than the preset definition level, step S2 is executed.
  • S12 Enhance the sharpness level of the car insurance image to obtain a new car insurance image, and calculate a second sharpness level of the new car insurance image.
  • the method of enhancing the clarity level of the car insurance image to obtain a new car insurance image includes:
  • low-pass filtering is performed on the spatial signal of the car insurance image to obtain the low-frequency component of the car insurance image, and the spatial signal of the car insurance image is subjected to a difference calculation to obtain the high-frequency component of the car insurance image.
  • the high-frequency components of the car insurance image are identified and then classified, the noise, details, small edges, and large edges in the high-frequency components are separated, and then the noise and details in the high-frequency components , Small edges and large edges are enhanced.
  • the enhancing processing of the identified high-frequency components includes:
  • step b2 By comparing the absolute value of the high-frequency components of the points of the car insurance image with the dynamic threshold of coring noise reduction, it is judged whether the points of the car insurance image belong to noise; if the points in the car insurance image are high If the absolute value of the frequency component is less than the coring noise reduction threshold, it is confirmed that the point is noise, and step b2 is executed; if the absolute value of the high frequency component of the point in the car insurance image is greater than or equal to the coring noise reduction threshold , Confirm that the point is not noise, and go to step b3.
  • the nonlinear high-frequency enhancement curve After the nonlinear high-frequency enhancement curve is processed, the details in the high-frequency component, the different regions corresponding to the small edges and the large edges can be processed to different degrees, and the enhanced image obtained thereby has a smooth and natural transition, and The monotonicity of high-frequency components is maintained.
  • a new car insurance image is obtained, and the second definition of the new car insurance image is calculated. It is understandable that the calculation method of the second definition of the new car insurance image is consistent with the calculation method of the first definition of the car insurance image, and will not be repeated here.
  • step S13 Compare whether the second sharpness level is lower than the preset sharpness level. When the second sharpness level is lower than the preset sharpness level, return to step S12; when the second sharpness level is higher than the preset sharpness level, perform step S2.
  • the method for prompting road information based on smart traffic may further include the step of performing data preprocessing on the car insurance image, wherein the data
  • the preprocessing process includes: analog-to-digital conversion, binarization, image smoothing, transformation, enhancement, restoration, filtering, etc.
  • the clarity of the auto insurance image uploaded by the user is adjusted to obtain an auto insurance image that meets the claim settlement requirements, which can improve the work efficiency of the self-assisted compensation system. It can also avoid the trouble that the user re-uploads the car insurance image when the clarity of the car insurance image uploaded by the user does not meet the claims requirements, thereby improving the user experience.
  • the car insurance image is associated with the location of the accident and then stored in the database of the server.
  • Step S2 Evaluate the risk level of the accident place based on the damage level of the vehicle in the car insurance image, the surrounding environment information of the accident place and/or the current accident occurrence frequency of the accident place.
  • the risk level of the accident site can be evaluated according to the damage level of the vehicle in the car insurance image and/or the surrounding environment information of the accident site.
  • the damage level of the vehicle in the car insurance image is acquired according to the car insurance image, and the risk level of the accident site is evaluated according to the damage level.
  • the method of assessing the risk level of the accident site based on the damage level of the vehicle in the car insurance image and the current number of accident occurrences in the accident site includes:
  • Step S21 Acquire the damage level of the vehicle in the car insurance image.
  • the damage area recognition model generated by pre-training is called to recognize the car insurance image to obtain vehicle damage area information; the damage area of the damage area is calculated according to the vehicle damage area information; the vehicle damage area and the damage The area is input into a preset calculation model and the calculation result is obtained, wherein the preset calculation model is the product of the damage area and the weight value of the damage area plus the damage area; when the calculation result is greater than or equal to the predicted Set a value to confirm that the damage level of the vehicle in the car insurance image is high; when the calculation result is less than the preset value, it is confirmed that the damage level of the vehicle in the car insurance image is low.
  • the damage area recognition model is called to identify the damage area of the car insurance image, and then the damage area of the damage area is calculated, and a preset calculation model is used to determine the vehicle in the car insurance image according to the damage area and the corresponding damage area. The level of damage that occurred.
  • a recognition model for recognizing the damaged area in the car insurance image may be pre-trained and generated, and the recognition model may be one of multiple models related to image processing.
  • the damage area recognition model is a convolutional neural network model.
  • the vehicle damage area may include a first area, a second area, a third area, a fourth area, and a fifth area.
  • the first area is a direct collision damage area (also referred to as a primary damage area);
  • the second area is an indirect collision damage area (also referred to as a secondary damage area);
  • the third area is a mechanical damage area, namely Damaged areas such as automobile mechanical parts, power transmission system parts, accessories, etc.;
  • the fourth area is the passenger compartment area, and various damages to the car compartment, including interior parts, lights, control devices, operating devices, and decorative layers;
  • the fifth area is the exterior and paint area, that is, damage to the exterior parts of the car body and various external parts.
  • the training process of the damage area recognition model includes:
  • the parameters of the convolutional neural network model are trained with default parameters, and the parameters are continuously adjusted during the training process.
  • the verified sample image verifies the generated convolutional neural network model. If the verification pass rate is greater than or equal to the preset threshold, for example, the pass rate is greater than or equal to 98%, the training ends, and the trained convolutional neural network model is Identify the damage area recognition model of the vehicle damage area in the car insurance image; if the verification pass rate is less than a preset threshold, for example, less than 98%, increase the number of auto insurance image samples and re-execute the above steps until the verification pass rate is greater than or Equal to the preset threshold.
  • the trained convolutional neural network model is used to identify the damage area of a preset number (such as ten) of auto insurance image samples randomly selected from the auto insurance image samples in the test set, and the recognition results are compared with manual The confirmed vehicle damage level results are compared to evaluate the recognition effect of the trained convolutional neural network model.
  • the method for prompting road information based on smart traffic includes the step of calculating the size of the damaged area of the damaged area.
  • the preset calculation model may be the product of the damage area and the weight value of the damage area plus the damage area.
  • the level of damage to the vehicle in the car insurance image is confirmed according to the size of the calculation result of the preset calculation model.
  • the calculation result is greater than or equal to the preset value, it is confirmed that the damage level of the vehicle in the car insurance image is high; when the calculation result is less than the preset value, it is confirmed that the damage level of the vehicle in the car insurance image is low .
  • Step S22 Obtain historical accident data of the place where the accident occurred, where the historical accident data resulted in including the number of accidents and the damage level of the vehicle in the historical accident.
  • the database of the server stores car insurance images and accident locations in historical report information. It is understandable that the historical accident data also includes the time when the accident occurred.
  • Step S23 Determine whether the current number of accidents in the accident place is greater than a preset number, and determine whether the damage level of the vehicle in the car insurance image and the current number of accidents is higher than a first preset level.
  • the car accident at the place where the accident occurred is less than or equal to the preset number, or the damage level of the vehicle in the current number of accident occurrences and the damage level of the vehicle in the car insurance image are both lower than the first preset level, It is confirmed that the car accident at the place where the accident occurred is an accident and is not the place where a high-risk accident occurs.
  • step S24 there is no need to deliberately remind the user to execute step S24 to mark the place where the accident occurs as a low-risk accident place; when the place where the accident occurs is currently When the number of accidents is greater than the preset number, and the damage level of the vehicle in the current number of accidents and the damage level of the vehicle in the car insurance image are both higher than the first preset level, confirm that the accident site is a high-risk accident site , A serious traffic accident is prone to occur, and step S25 is executed to mark the place of the accident as a place of high-risk accident.
  • Step S24 is performed to mark the accident location as low-risk. The place where the accident occurred.
  • the surrounding environment information of the accident place may be obtained according to the car insurance image, and the risk level of the accident place may be evaluated according to the surrounding environment information of the accident place.
  • the method for assessing the risk level of the accident site according to the surrounding environment information of the accident site includes:
  • the road environment information in the car insurance image includes whether there are foreign objects (such as gravel) on the road, whether the road is rugged, and whether it is a sharp turn, and so on.
  • the road surface environment information in the car insurance image is recognized by an image recognition method.
  • the image recognition method is an existing technology and will not be repeated here.
  • step S3 when the current number of accident occurrences at the accident location is less than or equal to the preset number, or the road condition level in the current accident occurrence number is lower than the second preset level, it is confirmed that the traffic accident at the accident location belongs to Unexpected situation is not the place where the high-risk accident occurs. There is no need to deliberately remind the user, and the process goes to step S3; when the current number of accidents in the place where the accident occurs is greater than the preset number, and the road condition level in the current number of accidents is higher than the second expected When setting the level, it is confirmed that the accident site is a high-risk accident site and is prone to serious traffic accidents, and the process goes to step S3.
  • the road information prompt method based on smart traffic may be based on the current number of accidents in the accident place, the damage level of the vehicle in the car insurance image, and the surrounding environment of the accident place Information to assess the risk level of the place where the accident occurred.
  • step S3 is executed; if the current number of accident occurrences at the accident site is less than the preset number, and When the damage level of the vehicle is lower than the first preset level, and it is determined that the road surface condition level in the surrounding environment information is lower than the second preset level, step S3 is executed.
  • Step S3 marking the accident occurrence location according to the risk level of the accident occurrence location, and obtaining information about the occurrence location of a high-risk accident and a low-risk accident occurrence location.
  • the accident is confirmed
  • the car accident at the place of occurrence is an accident, not the place where a high-risk accident occurs.
  • the location of the accident is confirmed A car accident is an accident and is not a high-risk accident location. There is no need to deliberately remind the user to mark the accident location as a low-risk accident location; when the current accident occurrence number in the accident location is greater than the preset number, and the current When the road condition level in the number of accidents is higher than the second preset level, it is confirmed that the accident site is a high-risk accident site and is prone to serious traffic accidents, and the accident site is marked as a high-risk accident site.
  • the location of the accident is marked as a high-risk accident location; if the accident occurs The current number of accidents at the locality is less than the preset number, and the damage level of the vehicle is lower than the first preset level, and the road surface condition level in the determination that the surrounding environment information is lower than the second preset When setting the level, mark the place where the accident occurred as a low-risk accident place.
  • Step S4 Send a warning message including the location of the high-risk accident to the relevant user.
  • the detailed information including the place where the high-risk accident occurred can be sent to the user who signed the insurance contract with the relevant insurance company, so that the user can be prompted In future trips, try to avoid the place where the high-risk accident occurs, so as to reduce the occurrence of traffic accidents and also reduce the number of claims for insurance companies.
  • the information of the location of the high-risk accident may be sent to the navigation system. Therefore, when the user drives the vehicle to arrive at the location of the high-risk accident, the navigation system can send voice prompt information to remind the user that the vehicle is about to arrive at the location of the high-risk accident. As a result, the user's vigilance can be increased, and the user is reminded to be more careful when driving the vehicle through the place where the high-risk accident occurs, so as to avoid traffic accidents. It is also possible to provide the user with a route that avoids the place where the high-risk accident occurs when the user uses the navigation system to plan a travel route.
  • the method for prompting road information based on smart traffic can also prompt related users in combination with the time of the accident.
  • the place of occurrence of the accident is the place of occurrence of a high-risk accident
  • the time of occurrence of the accident in the historical accident data of the place of occurrence of the accident is acquired.
  • the relevant users are reminded to avoid the high-risk accident when traveling in the time period. This can reduce the occurrence of traffic accidents and reduce the number of claims for insurance companies.
  • the method for prompting road information based on smart traffic can also remind related users in combination with weather conditions when the accident occurs.
  • the weather conditions at the time of the accident in the accident location in the historical accident data are acquired.
  • the smart transportation-based road information prompt method includes acquiring auto insurance report information, where the auto insurance report information includes at least an auto insurance image and an accident location; the accident occurrence is evaluated based on the auto insurance image The risk level of the place; mark the place of the accident according to the risk level of the place of the accident; send warning information including the place of the high-risk accident to relevant users. It can count the accident-prone areas and remind users of the risk levels of the accident-prone areas according to the level of vehicle damage in the accident, so that users can get enough attention, and remind users to avoid the high-risk accident-prone areas as much as possible when traveling, thereby Can reduce the occurrence of traffic accidents.
  • FIG. 4 is a diagram of functional modules in a preferred embodiment of a road information prompting device based on smart traffic in this application.
  • the smart traffic-based road information prompting device 40 (for ease of description, hereinafter referred to as "road information prompting device 40") runs in a server.
  • the road information prompting device 40 may include multiple functional modules composed of program code segments.
  • the program code of each program segment in the road information prompting device 40 can be stored in the memory and executed by at least one processor to execute (see Figure 1 and related descriptions for details) the road information prompt function based on smart traffic .
  • the road information prompting device 40 can be divided into multiple functional modules according to the functions it performs.
  • the functional modules may include: an acquisition module 401, an evaluation module 402, a marking module 403, and a sending module 404.
  • the module referred to in this application refers to a series of computer-readable instruction segments that can be executed by at least one processor and can complete fixed functions, and are stored in a memory. In some embodiments, the functions of each module will be detailed in subsequent embodiments.
  • the acquisition module 401 is configured to acquire auto insurance report information, where the auto insurance report information includes at least a car insurance image and a place where an accident occurred.
  • the server may obtain report information from the mobile terminal.
  • the mobile terminal may be a smart terminal such as a mobile phone, a tablet computer, a personal digital assistant, a wearable device (for example, a smart watch, smart glasses), or any other suitable electronic device.
  • the report information may include the car insurance image and the place where the accident occurred, and may also include the owner's information, the license plate number of the dangerous vehicle, the time of the accident, and the reason for the accident.
  • the car insurance image may be video information or image information taken by a car owner.
  • the car insurance image may be video information or image information collected on site by an operator (such as a surveyor), and the operator sends the car insurance image to a database of another system (such as an insurance company system).
  • the server may obtain the car insurance image from the other system database.
  • the car insurance image is associated with the place where the accident occurred.
  • the car insurance image may include a general term for various graphics or images, usually referring to images with visual effects, and generally may include images on paper media, negatives or photos, televisions, projectors, or computer screens.
  • the car insurance image described in this embodiment may include computer image data stored on a readable storage medium after being captured by a camera or camera device, and may include various types of computer images such as vector graphics, bitmaps, static and dynamic images.
  • the smart transportation-based road information prompt device 40 can also improve the clarity level of the car insurance image.
  • the method for improving the clarity level of the car insurance image includes:
  • the first sharpness level of the car insurance image can be calculated by calculations such as a grayscale change function, a gradient function, or an image grayscale entropy function.
  • the gray-scale change function, the gradient function, or the image gray-scale entropy function are existing techniques for calculating image clarity, and will not be repeated here.
  • the method of enhancing the clarity level of the car insurance image to obtain a new car insurance image includes:
  • low-pass filtering is performed on the spatial signal of the car insurance image to obtain the low-frequency component of the car insurance image, and the spatial signal of the car insurance image is subjected to a difference calculation to obtain the high-frequency component of the car insurance image.
  • the high-frequency components of the car insurance image are identified and then classified, the noise, details, small edges, and large edges in the high-frequency components are separated, and then the noise and details in the high-frequency components , Small edges and large edges are enhanced.
  • the enhancing processing of the identified high-frequency components includes:
  • the value of the high frequency component corresponding to the point is set to 0.
  • the absolute value of the high-frequency component of the point in the car insurance image is greater than or equal to the cored noise reduction threshold, it is confirmed that the point is not noise, and a nonlinear high-frequency enhancement curve is applied to perform the high-frequency component of the car insurance image.
  • a nonlinear high-frequency enhancement curve is applied to perform the high-frequency component of the car insurance image.
  • the details in the high-frequency component, the different regions corresponding to the small edges and the large edges can be processed to different degrees, and the enhanced image obtained thereby has a smooth and natural transition, and The monotonicity of high-frequency components is maintained.
  • a new car insurance image is obtained, and the second definition of the new car insurance image is calculated. It is understandable that the calculation method of the second definition of the new car insurance image is consistent with the calculation method of the first definition of the car insurance image, and will not be repeated here.
  • the method for prompting road information based on smart traffic may further include the step of performing data preprocessing on the car insurance image, wherein the data
  • the preprocessing process includes: analog-to-digital conversion, binarization, image smoothing, transformation, enhancement, restoration, filtering, etc.
  • the clarity of the auto insurance image uploaded by the user is adjusted to obtain an auto insurance image that meets the claim settlement requirements, which can improve the work efficiency of the self-assisted compensation system. It can also avoid the trouble that the user re-uploads the car insurance image when the clarity of the car insurance image uploaded by the user does not meet the claims requirements, thereby improving the user experience.
  • the car insurance image is associated with the location of the accident and then stored in the database of the server.
  • the evaluation module 402 is configured to evaluate the risk level of the accident place according to the damage level of the vehicle in the car insurance image, the surrounding environment information of the accident place and/or the current number of accident occurrences in the accident place .
  • the risk level of the accident site can be evaluated according to the damage level of the vehicle in the car insurance image and/or the surrounding environment information of the accident site.
  • the damage level of the vehicle in the car insurance image is acquired according to the car insurance image, and the risk level of the accident site is evaluated according to the damage level.
  • the method for assessing the risk level of the accident site based on the damage level of the vehicle in the car insurance image and the current number of accident occurrences in the accident site includes:
  • the damage area recognition model generated by pre-training is called to recognize the car insurance image to obtain vehicle damage area information; the damage area of the damage area is calculated according to the vehicle damage area information; the vehicle damage area and the damage The area is input into a preset calculation model and the calculation result is obtained, wherein the preset calculation model is the product of the damage area and the weight value of the damage area plus the damage area; when the calculation result is greater than or equal to the predicted Set a value to confirm that the damage level of the vehicle in the car insurance image is high; when the calculation result is less than the preset value, it is confirmed that the damage level of the vehicle in the car insurance image is low.
  • the damage area recognition model is called to identify the damage area of the car insurance image, and then the damage area of the damage area is calculated, and a preset calculation model is used to determine the vehicle in the car insurance image according to the damage area and the corresponding damage area. The level of damage that occurred.
  • a recognition model for recognizing the damaged area in the car insurance image may be pre-trained and generated, and the recognition model may be one of multiple models related to image processing.
  • the damage area recognition model is a convolutional neural network model.
  • the vehicle damage area may include a first area, a second area, a third area, a fourth area, and a fifth area.
  • the first area is a direct collision damage area (also referred to as a primary damage area);
  • the second area is an indirect collision damage area (also referred to as a secondary damage area);
  • the third area is a mechanical damage area, namely Damaged areas such as automobile mechanical parts, power transmission system parts, accessories, etc.;
  • the fourth area is the passenger compartment area, and various damages to the car compartment, including interior parts, lights, control devices, operating devices, and decorative layers;
  • the fifth area is the exterior and paint area, that is, damage to the exterior parts of the car body and various external parts.
  • the training process of the damage area recognition model includes:
  • the parameters of the convolutional neural network model are trained with default parameters, and the parameters are continuously adjusted during the training process.
  • the verified sample image verifies the generated convolutional neural network model. If the verification pass rate is greater than or equal to the preset threshold, for example, the pass rate is greater than or equal to 98%, the training ends, and the trained convolutional neural network model is Identify the damage area recognition model of the vehicle damage area in the car insurance image; if the verification pass rate is less than a preset threshold, for example, less than 98%, increase the number of auto insurance image samples and re-execute the above steps until the verification pass rate is greater than or Equal to the preset threshold.
  • the trained convolutional neural network model is used to identify the damage area of a preset number (such as ten) of auto insurance image samples randomly selected from the auto insurance image samples in the test set, and the recognition results are compared with manual The confirmed vehicle damage level results are compared to evaluate the recognition effect of the trained convolutional neural network model.
  • the evaluation module 402 is further used to calculate the size of the damaged area of the damaged area.
  • the preset calculation model may be the product of the damage area and the weight value of the damage area plus the damage area.
  • the level of damage to the vehicle in the car insurance image is confirmed according to the size of the calculation result of the preset calculation model.
  • the calculation result is greater than or equal to the preset value, it is confirmed that the damage level of the vehicle in the car insurance image is high; when the calculation result is less than the preset value, the damage level of the vehicle in the car insurance image is confirmed low.
  • the database of the server stores car insurance images and accident locations in historical report information. It is understandable that the historical accident data also includes the time when the accident occurred.
  • the current number of accident occurrences in the place where the accident occurred is less than or equal to the preset number, or the damage level of the vehicle in the current number of accidents and the damage level of the vehicle in the car insurance image are both lower than the first preset level , Confirming that the car accident occurred at the place of the accident is an accident, not the place where the high-risk accident occurred, and there is no need to deliberately remind the user to mark the place of the accident as the place of low-risk accident; the current number of accidents when the accident occurred When the number of accidents is greater than the preset number, and the damage level of the vehicle in the current number of accidents and the damage level of the vehicle in the car insurance image are higher than the first preset level, it is confirmed that the accident site is a high-risk accident site and is prone to occur For a serious traffic accident, mark the place where the accident occurred as a place where a high-risk accident occurred.
  • the current number of accidents at the accident location is less than or equal to the preset number, or the damage level of the vehicle in the current number of accidents is higher than the first preset level, and the vehicle in the car insurance image
  • the damage level of is lower than the first preset level, it is confirmed that the car accident at the place of the accident is an accident and is not a high-risk accident place, and there is no need to deliberately remind the user to mark the place of the accident as a low-risk accident place.
  • the surrounding environment information of the accident place may be obtained according to the car insurance image, and the risk level of the accident place may be evaluated according to the surrounding environment information of the accident place.
  • the method for assessing the risk level of the accident site according to the surrounding environment information of the accident site includes:
  • the road environment information includes whether there are foreign objects (such as gravel) on the road surface, whether the road surface is bumpy, whether it is a sharp turn, and so on.
  • the road surface environment information in the car insurance image is recognized by an image recognition method.
  • the image recognition method is an existing technology and will not be repeated here.
  • the car accident at the place where the accident occurs is an accident If the situation is not a high-risk accident location, there is no need to deliberately remind the user to mark the accident location according to the risk level of the accident location; when the current accident occurrence number of the accident location is greater than the preset number, and the current accident occurrence
  • the road condition level is higher than the second preset level in the number of times, it is confirmed that the accident site is a high-risk accident site and is prone to serious traffic accidents, and the accident site is marked according to the risk level of the accident site.
  • the road information prompting device 40 may use the current number of accident occurrences in the accident place, the damage level of the vehicle in the car insurance image, and the surrounding environment information of the accident place. Assess the risk level of the place where the accident occurred.
  • the accident location is marked according to the risk level of the accident location;
  • the current number of accidents is less than the preset number of times, the damage level of the vehicle is lower than the first preset level, and the road surface condition level in the determining the surrounding environment information is lower than the second preset level.
  • the marking module 403 is used to mark the place of the accident according to the risk level of the place of the accident to obtain information on the place of occurrence of high-risk accidents and information of the place of occurrence of low-risk accidents.
  • the accident is confirmed
  • the car accident at the place of occurrence is an accident, not the place where a high-risk accident occurs.
  • the occurrence of the accident is confirmed A car accident in a locality is an accident and is not a high-risk accident location. There is no need to deliberately remind the user to mark the accident location as a low-risk accident location; when the current accident occurrence number in the accident location is greater than the preset number, and When the road condition level in the current number of accidents is higher than the second preset level, it is confirmed that the accident site is a high-risk accident site and is prone to serious traffic accidents, and the accident site is marked as a high-risk accident site.
  • the location of the accident is marked as a high-risk accident location; if the accident occurs The current number of accidents at the locality is less than the preset number, and the damage level of the vehicle is lower than the first preset level, and the road surface condition level in the determination that the surrounding environment information is lower than the second preset When setting the level, mark the place where the accident occurred as a low-risk accident place.
  • the sending module 404 is configured to send warning information including the location of the high-risk accident to relevant users.
  • the detailed information including the place where the high-risk accident occurred can be sent to the user who signed the insurance contract with the relevant insurance company, so that the user can be prompted In future trips, try to avoid the place where the high-risk accident occurs, so as to reduce the occurrence of traffic accidents and also reduce the number of claims for insurance companies.
  • the information of the location of the high-risk accident may be sent to the navigation system. Therefore, when the user drives the vehicle to arrive at the location of the high-risk accident, the navigation system can send voice prompt information to remind the user that the vehicle is about to arrive at the location of the high-risk accident. As a result, the user's vigilance can be increased, and the user is reminded to be more careful when driving the vehicle through the place where the high-risk accident occurs, so as to avoid traffic accidents. It is also possible to provide the user with a route that avoids the place where the high-risk accident occurs when the user uses the navigation system to plan a travel route.
  • the road information prompting device 40 can also prompt related users in combination with the time of the accident.
  • the place of occurrence of the accident is the place where a high-risk accident occurs
  • obtain the time of the accident in the historical accident data of the place where the accident occurred if the time of the accident in the historical accident data is concentrated in a certain time period, for example , From 5:00 to 9:00 in the morning, the relevant users are reminded to avoid travel where the high-risk accident occurs during the time period, thereby reducing the occurrence of traffic accidents and reducing the number of claims for insurance companies.
  • the road information prompting device 40 can also remind related users in combination with weather conditions when the accident occurs.
  • the weather conditions at the time of the accident in the accident location in the historical accident data are acquired.
  • the road information prompting device 40 provided in the present application can count accident-prone areas, and remind users of the risk levels of the accident-prone areas according to the level of vehicle damage in the accident, so that the user can get enough attention and remind the user to do everything possible when traveling. It is possible to avoid the high-risk accident-prone areas, thereby reducing the occurrence of traffic accidents.
  • the above-mentioned integrated unit implemented in the form of a software function module may be stored in a nonvolatile readable storage medium.
  • the above-mentioned software function module is stored in a storage medium, and includes several instructions to make a computer device (which can be a personal computer, a dual-screen device, or a network device, etc.) or a processor to execute the various embodiments of this application Method part.
  • FIG. 5 is a schematic diagram of a server provided in Embodiment 3 of this application.
  • the server 5 includes: a database 51, a memory 52, at least one processor 53, computer readable instructions 54 stored in the memory 52 and executable on the at least one processor 53 and at least one communication bus 55.
  • the at least one processor 53 executes the computer readable instruction 54, the steps in the above embodiment of the method for prompting road information based on smart traffic are implemented.
  • the computer-readable instructions 54 may be divided into one or more modules/units, and the one or more modules/units are stored in the memory 52 and executed by the at least one processor 53 Execute to complete this application.
  • the one or more modules/units may be a series of computer-readable instruction segments capable of completing specific functions, and the instruction segments are used to describe the execution process of the computer-readable instructions 54 in the server 5.
  • the server 5 is a device that can automatically perform numerical calculation and/or information processing in accordance with pre-set or stored instructions. Its hardware includes, but is not limited to, a microprocessor and an application specific integrated circuit (ASIC) , Field-Programmable Gate Array (FPGA), Digital Processor (Digital Signal Processor, DSP), embedded equipment, etc. Those skilled in the art can understand that the schematic diagram 5 is only an example of the server 5, and does not constitute a limitation on the server 5. It may include more or less components than those shown in the figure, or combine some components, or different components. For example, the server 5 may also include input and output devices, network access devices, buses, etc.
  • ASIC application specific integrated circuit
  • FPGA Field-Programmable Gate Array
  • DSP Digital Processor
  • embedded equipment etc.
  • the schematic diagram 5 is only an example of the server 5, and does not constitute a limitation on the server 5. It may include more or less components than those shown in the figure, or combine some components, or different components.
  • the server 5 may also include input and
  • the database (Database) 51 is a warehouse built on the server 5 to organize, store and manage data according to a data structure. Databases are usually divided into three types: hierarchical database, network database and relational database. In this embodiment, the database 51 is used to store the car insurance image information.
  • the at least one processor 53 may be a central processing unit (Central Processing Unit, CPU), or other general-purpose processors, digital signal processors (Digital Signal Processors, DSP), application specific integrated circuits (ASICs). ), ready-made programmable gate array (Field-Programmable Gate Array, FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc.
  • the processor 53 can be a microprocessor, or the processor 53 can also be any conventional processor, etc.
  • the processor 53 is the control center of the server 5, and connects each of the entire server 5 through various interfaces and lines. section.
  • the memory 52 may be used to store the computer-readable instructions 54 and/or modules/units, and the processor 53 can run or execute the computer-readable instructions and/or modules/units stored in the memory 52, and The data stored in the memory 52 is called to realize various functions of the server 5.
  • the memory 31 includes Read-Only Memory (ROM), Programmable Read-Only Memory (PROM), and Erasable Programmable Read-Only Memory (EPROM) , One-time Programmable Read-Only Memory (OTPROM), Electronically-Erasable Programmable Read-Only Memory (EEPROM), CD-ROM (Compact Disc Read- Only Memory, CD-ROM) or other optical disk storage, magnetic disk storage, tape storage, or any other non-volatile readable storage medium that can be used to carry or store data.
  • ROM Read-Only Memory
  • PROM Programmable Read-Only Memory
  • EPROM Erasable Programmable Read-Only Memory
  • OTPROM One-time Programmable Read-Only Memory
  • EEPROM Electronically-Erasable Programmable Read-Only Memory
  • CD-ROM Compact Disc Read- Only Memory
  • CD-ROM Compact Disc Read- Only Memory
  • the memory 52 stores program codes
  • the at least one processor 53 can call the program codes stored in the memory 52 to perform related functions.
  • the modules (acquisition module 401, evaluation module 402, marking module 403, and sending module 404) described in FIG. 4 are program codes stored in the memory 52 and executed by the at least one processor 53 , So as to realize the functions of the various modules to achieve the purpose of road information prompt based on smart traffic.
  • the integrated modules/units of the server 5 are implemented in the form of software functional units and sold or used as independent products, they can be stored in a non-volatile readable storage medium.
  • this application implements all or part of the processes in the above-mentioned embodiments and methods, and can also be completed by instructing relevant hardware by computer-readable instructions.
  • the computer-readable instructions can be stored in a non-volatile memory. In the read storage medium, when the computer-readable instructions are executed by the processor, the steps of the foregoing method embodiments can be implemented.
  • the computer-readable instruction includes computer-readable instruction code
  • the computer-readable instruction code may be in the form of source code, object code, executable file, or some intermediate form.
  • the non-volatile readable medium may include: any entity or device capable of carrying the computer readable instruction code, recording medium, U disk, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM, Read -Only Memory).
  • the server 5 may also include a power source (such as a battery) for supplying power to various components.
  • the power source may be logically connected to the at least one processor 53 through a power management system, thereby implementing management through the power management system. Functions such as charging, discharging, and power management.
  • the power supply may also include one or more DC or AC power supplies, recharging systems, power failure detection circuits, power converters or inverters, power supply status indicators and other arbitrary components.
  • the server 5 may also include a Bluetooth module, a Wi-Fi module, etc., which will not be repeated here.

Landscapes

  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Atmospheric Sciences (AREA)
  • Financial Or Insurance-Related Operations Such As Payment And Settlement (AREA)
  • Traffic Control Systems (AREA)

Abstract

一种基于智慧交通的道路信息提示方法、装置、服务器即存储介质，所述方法包括：获取车险报案信息，其中，所述车险报案信息至少包括车险图像和事故发生地（S1）；根据所述车险图像评估所述事故发生地的风险级别（S2）；根据所述事故发生地的风险级别标记所述事故发生地，得到高风险事故发生地信息和低风险事故发生地信息（S3）；发送包括高风险事故发生地的警示信息至相关用户（S4）。通过上述方法可以统计事故多发地，并采用机器学习判定事故中车辆损伤级别，根据所述车辆损伤级别提醒用户所述事故多发地的危险级别，使用户得到足够重视，还可以提醒用户出行时尽可能避开所述危险级别高的事故多发地，从而减少交通事故的发生。

Description

基于智慧交通的道路信息提示方法、装置、服务器及介质
本申请要求于2019年07月25日提交中国专利局，申请号为201910678463.6申请名称为“危险路段识别方法、装置、服务器及存储介质”的中国专利申请的优先权，其全部内容通过引用结合在本申请中。
技术领域
本申请涉及计算机技术领域，具体涉及一种基于智慧交通的道路信息提示方法、装置、服务器及存储介质。
背景技术
车辆安全预警***或装置的作用就是为了提醒驾驶员可能存在的危险，以使驾驶员提高警惕，规范操作来达到规避危险的目的。目前，一些导航装置或***可以提供对于危险路段的提示功能，例如当车辆行驶到已知的危险路段时，***发出某种预警信息，提示驾驶员减速慢行，将很大程度提升行车安全及保障行车人员和乘客的人身安全。然而，现有技术中仅能根据当前路段是否为事故多发地、是否有落石来提醒驾驶员。
发明内容
鉴于以上内容，有必要提出一种基于智慧交通的道路信息提示方法、装置、服务器及存储介质，能够根据车险报案信息评估事故发生地的风险级别，以提示用户避开高风险道路。
一种基于智慧交通的道路信息提示方法，所述方法包括：获取车险报案信息，其中，所述车险报案信息至少包括车险图像和事故发生地；根据所述车险图像中的车辆的损伤级别、所述事故发生地的周围环境信息和/或所述事故发生地的当前事故发生次数评估所述事故发生地的风险级别；根据所述事故发生地的风险级别标记所述事故发生地，得到高风险事故发生地信息和低风险事故发生地信息；发送包括所述高风险事故发生地的警示信息至相关用户。
一种基于智慧交通的道路信息提示装置，所述装置包括：获取模块，用于获取车险报案信息，其中，所述车险报案信息至少包括车险图像和事故发生地；评估模块，用于根据所述车险图像中的车辆的损伤级别、所述事故发生地的周围环境信息和/或所述事故发生地的当前事故发生次数评估所述事故发生地的风险级别；标记模块，用于根据所述事故发生地的风险级别标记所述事故发生地，得到高风险事故发生地信息和低风险事故发生地信息；发送模块，用于发送包括所述高风险事故发生地的警示信息至相关用户。
一种服务器，所述服务器包括处理器和存储器，所述处理器用于执行存储器中存储的至少一个计算机可读指令时实现以下步骤：获取车险报案信息，其中，所述车险报案信息至少包括车险图像和事故发生地；根据所述车险图像中的车辆的损伤级别、所述事故发生地的周围环境信息和/或所述事故发生地的当前事故发生次数评估所述事故发生地的风险级别；根据所述事故发生地的风险级别标记所述事故发生地，得到高风险事故发生地信息和低风险事故发生地信息；发送包括所述高风险事故发生地的警示信息至相关用户。
一种非易失性可读存储介质，所述非易失性可读存储介质存储有至少一个计算机可读指令，所述至少一个计算机可读指令被处理器执行时实现以下步骤：获取车险报案信息，其中，所述车险报案信息至少包括车险图像和事故发生地；根据所述车险图像中的车辆的损伤级别、所述事故发生地的周围环境信息和/或所述事故发生地的当前事故发生次数评估所述事故发 生地的风险级别；根据所述事故发生地的风险级别标记所述事故发生地，得到高风险事故发生地信息和低风险事故发生地信息；发送包括所述高风险事故发生地的警示信息至相关用户。
由以上技术方案可知，本申请提供的基于智慧交通的道路信息提示方法、装置、服务器及存储介质，根据车险图像信息评估事故发生地的风险级别，并在标记所述事故发生地的风险级别后发送包括高风险事故发生地的警示信息至相关用户。不仅可以统计事故多发地，并根据事故中车辆损伤级别提醒用户所述事故多发地的危险级别，使用户得到足够重视，还可以提醒用户出行时尽可能避开所述危险级别高的事故多发地，从而减少交通事故的发生。
附图说明
图1是本申请实施例一提供的基于智慧交通的道路信息提示方法的流程图。
图2是本申请实施例一提供的基于智慧交通的道路信息提示方法中提高车险图像的清晰度等级的方法的流程图。
图3是本申请实施例一提供的基于智慧交通的道路信息提示方法中根据车险图像中车辆损伤级别确定事故发生地的风险级别的方法的流程图。
图4是本申请实施例二提供的本申请基于智慧交通的道路信息提示装置较佳实施例中的功能模块图。
图5是本申请实施例三提供的服务器的示意图。
具体实施方式
实施例一
图1是本申请实施例一提供的基于智慧交通的道路信息提示方法的流程图。根据不同的需求，该流程图中的执行顺序可以改变，某些步骤可以省略。
步骤S1，获取车险报案信息，其中，所述车险报案信息至少包括车险图像和事故发生地。
本实施例中，所述服务器可以从移动终端获取报案信息。所述移动终端可以是手机、平板电脑、个人数字助理、可穿戴设备(例如、智能手表、智能眼镜)等智能终端或其他任意适用的电子设备。所述报案信息可以包括车险图像和事故发生地，还可以包括车主的信息、出险车辆的车牌号、出险时间以及出险原因等。所述车险图像可以是车主拍摄的视频信息或图像信息。
在其他实施方式中，所述车险图像可以是作业人员(如查勘员)现场采集的视频信息或图像信息，所述作业人员将所述车险图像发送至其他***(如保险公司***)数据库中。所述服务器可以从所述其他***数据库中获取所述车险图像。所述车险图像与所述事故发生地相关联。
所述车险图像可以包括各种图形或影像的总称，通常指具有视觉效果的画面，一般可以包括纸介质上的、底片或照片上的、电视、投影仪或计算机屏幕等上的画面。本实施例中所述的车险图像可以包括通过照相或摄像设备拍摄后存储在可读存储介质上的计算机图像数据，可以包括矢量图、位图，静态、动态图像等多种类型的计算机图像。
优选地，如图2所示，在获取车险报案信息后，所述基于智慧交通的道路信息提示方法还可以提高车险图像的清晰度等级。所述提高车险图像的清晰度等级的方法包括：
S10，计算所述车险图像的第一清晰度等级。
在本实施方式中，可以通过灰度变化函数、梯度函数、或者图像灰度熵函数等来计算得到所述车险图像的第一清晰度等级。灰度变化函数、梯度函数、或者图像灰度熵函数为计算图像清晰度的现有技术，在此不再赘述。
S11，比对所述第一清晰度等级是否低于预设清晰度等级。当所述第一清晰度等级低于所述预设清晰度等级时，执行步骤S12；当所述第一清晰度等级高于所述预设清晰度 等级时，执行步骤S2。
S12，增强所述车险图像的清晰度等级以得到新的车险图像，并计算所述新的车险图像的第二清晰度等级。
在本实施方式中，增强所述车险图像的清晰度等级以得到新的车险图像的方法包括：
a：计算所述车险图像中的高频分量和低频分量。
具体地，通过对所述车险图像的空域信号进行低通滤波后得到所述车险图像的低频分量，对所述车险图像的空域信号进行差值运算后得到所述车险图像的高频分量。
b：对所述车险图像中的高频分量进行识别，并对识别后的高频分量进行增强处理。
具体地，通过对所述车险图像的高频分量识别后进行分类，分出所述高频分量中的噪声、细节、小边缘和大边缘，再分别对所述高频分量中的噪声、细节、小边缘和大边缘进行增强处理。
所述对识别后的高频分量进行增强处理包括：
b1：计算核化降噪动态阈值，判断所述车险图像中的点是否属于噪声；
通过将所述车险图像的点的高频分量的绝对值与所述核化降噪动态阈值进行比对，来判断所述车险图像的点是否属于噪声；若所述车险图像中的点的高频分量的绝对值小于所述核化降噪阈值，则确认所述点为噪声，执行步骤b2；若所述车险图像中的点的高频分量的绝对值大于等于所述核化降噪阈值，则确认所述点不是噪声，执行步骤b3。
b2：将所述点对应的高频分量的值设置为0。
通过将所述点对应的高频分量的值设置为0以抑制掉小幅高频噪声，达到核化降噪的目的。
b3：应用非线性高频增强曲线对所述车险图像的高频分量进行增强。
通过所述非线性高频增强曲线处理后可以对所述高频分量中的细节、小边缘和大边缘对应的不同区域段进行不同程度的处理，由此得到的增强图像，过渡平滑自然，且保持了高频分量的单调性。
c：将增强后的高频分量与所述低频分量叠加得到新的车险图像。
在增强所述车险图像的清晰度后得到新的车险图像，并计算所述新的车险图像的第二清晰度。可以理解的是，所述新的车险图像的第二清晰度的计算方法与所述车险图像的第一清晰度的计算方法一致，不再赘述。
S13，比对所述第二清晰度等级是否低于所述预设清晰度等级。当所述第二清晰度等级低于所述预设清晰度等级时，返回步骤S12；当所述第二清晰度等级高于所述预设清晰度等级时，执行步骤S2。
可以理解的是，在计算所述车险图像的第一清晰度等级之前，所述基于智慧交通的道路信息提示方法还可以包括：对所述车险图像进行数据预处理的步骤，其中，所述数据预处理过程包括：模数转换、二值化、图像的平滑、变换、增强、恢复、滤波等。
在本实施方式中，通过对用户上传的车险图像的清晰度进行调整，从而得到满足理赔要求的车险图像，可以提高自助理赔***的工作效率。还可以避免由于用户上传的车险图像的清晰度不符合理赔要求时，麻烦用户重新上传车险图像的情况出现，提高了用户体验。
可以理解的是，在获取车险报案信息后，将所述车险图像和事故发生地关联后存储至服务器的数据库中。
步骤S2，根据所述车险图像中的车辆的损伤级别、所述事故发生地的周围环境信息和/或所述事故发生地的当前事故发生次数评估所述事故发生地的风险级别。
在本实施方式中，可以根据车险图像中车辆的损伤级别，以及/或者事故发生地的周围环境信息来评估所述事故发生地的风险级别。
在第一实施例中，根据所述车险图像获取所述车险图像中车辆的损伤级别，并根据所述损伤级别评估所述事故发生地的风险级别。
具体地，如图3所示，根据所述车险图像中的车辆的损伤级别和所述事故发生地的当前事故发生次数评估所述事故发生地的风险级别的方法包括：
步骤S21，获取所述车险图像中车辆的损伤级别。
本实施例中，调用预先训练生成的损伤区域识别模型识别所述车险图像，得到车辆损伤区域信息；根据所述车辆损伤区域信息计算所述损伤区域的损伤面积；将所述车辆损伤区域和损伤面积输入预设计算模型并得到计算结果，其中，所述预设计算模型为所述损伤区域与所述损伤区域的权重值的乘积加上所述损伤面积；当所述计算结果大于或等于预设值，确认所述车险图像中车辆的损伤级别高；当所述计算结果小于所述预设值，确认所述车险图像中车辆的损伤级别低。
具体地，调用损伤区域识别模型对车险图像进行损伤区域识别，再计算所述损伤区域的损伤面积大小，根据所述损伤区域及对应的损伤面积采用预设计算模型来确定所述车险图像中车辆出现损伤的级别。
本实施例中，可以预先训练生成用于识别车险图像中的损伤区域的识别模型，所述识别模型可以是图像处理相关的多种模型中的一种。
优选地，所述损伤区域识别模型为卷积神经网络模型。
一般来说，所述车辆损伤区域可以包括第一区域、第二区域、第三区域、第四区域及第五区域。所述第一区域为直接碰撞损伤区(又称为一次损伤区)；所述第二区域为间接碰撞损伤区(又称为二次损伤区)；所述第三区域为机械损伤区，即汽车机械零件、动力传动***零件、附件等损伤区；所述第四区域为乘员舱区，及车厢的各种损坏，包括内饰件、灯、控制装置、操纵装置和饰层等；所述第五区域为外饰和漆面区，即车身外饰件及外部各种零部件的损伤。
优选地，所述损伤区域识别模型的训练过程包括：
1)获取预设数量的车险图像样本；
2)从所述车险图像样本中提取出预设比例的车险图像作为待训练的样本图片，并将所述预设数量的车险图像样本中剩余的车险图像样本作为待验证的样本图片；
3)利用各待训练的样本图片进行模型训练，以生成所述卷积神经网络模型，并利用各待验证的样本图片对所生成的卷积神经网络模型进行验证；
4)若验证通过率大于等于预设阈值，则训练完成，否则增加所述车险图像样本的数量，以重新进行训练及验证。
示例性的，假设获取10万张理赔车险图像样本图片。提取预设比例的理赔保单样本图片作为训练集，并将预设数量的车险图像样本图片中剩余的车险图像样本图片作为测试集，训练集中的车险图像样本图片的数量大于测试集中的车险图像样本图片的数量，例如将车险图像样本图片中的80％的车险图像样本图片作为训练集，将剩余的20％的车险图像样本图片作为测试集。
在第一次训练卷积神经网络模型时，所述卷积神经网络模型的参数采用默认的参数进行训练，在训练过程不断调整参数，在训练生成所述卷积神经网络模型后，利用各待验证的样本图片对所生成的卷积神经网络模型进行验证，如果验证通过率大于等于预设阈值，例如通过率大于等于98％，则训练结束，以所述训练得到的卷积神经网络模型为识别所述车险图像中车辆损伤区域的损伤区域识别模型；如果验证通过率小于预设阈值，例如小于98％，则增加车险图像样本的数量，并重新执行上述的步骤，直至验证通过率大于或者等于预设阈值。
在测试时，使用训练得到的卷积神经网络模型对从所述测试集中的车险图像样本中 随机选取的预设个数(如十张)车险图像样本进行损伤区域识别，并将识别结果与人工确认的车辆损伤级别结果进行对比，以评估所训练的卷积神经网络模型的识别效果。
在识别所述车险图像中车辆的损伤区域后，所述基于智慧交通的道路信息提示方法包括计算所述损伤区域的损伤面积大小的步骤。
可以理解的是，在计算所述损伤区域的损伤面积大小时需要先计算所述车险图像中车辆的损伤区域，再按照一定比例计算得到所述损伤区域的实际面积大小。根据所述损伤区域、对应所述损伤区域的权重值及对应的损伤面积采用预设计算模型来确定所述车险图像中车辆出现损伤的级别。
示例性的，所述预设计算模型可以是所述损伤区域与所述损伤区域的权重值的乘积加上所述损伤面积。所述车险图像中车辆出现损伤的级别根据所述预设计算模型的计算结果大小来进行确认。当所述计算结果大于等于预设值时，确认所述车险图像中车辆出现损伤的级别高；当所述计算结果小于所述预设值时，确认所述车险图像中车辆出现损伤的级别低。
步骤S22，获取所述事故发生地的历史事故数据，其中，所述历史事故数据致使包括事故发生次数及所述历史事故中车辆的损伤级别。
本实施例中，所述服务器的数据库中存储了历史报案信息中的车险图像及事故发生地。可以理解的是，所述历史事故数据还包括事故发生时间。
步骤S23，判断所述事故发生地的当前事故发生次数是否大于预设次数，及判断所述车险图像中车辆的损伤级别和当前事故发生次数中车辆的损伤级别是否高于第一预设级别。
本实施例中，为了解决现有技术中仅提示用户某地段为事故多发地，并不能引起用户足够重视而在所述事故多发地依然出现交通事故的问题。本方案不仅统计所述事故发生地发生事故的次数，还可以确定所述事故中车辆的损伤级别，将所述事故发生次数与车辆损伤级别进行关联来评定所述地段是否为高风险事故发生地。
具体地，当所述事故发生地的当前事故发生次数小于等于预设次数，或者当前事故发生次数中车辆的损伤级别和所述车险图像中车辆的损伤级别都低于第一预设级别时，确认所述事故发生地发生车祸属于意外情况，并非高风险事故发生地，无需刻意提醒用户，执行步骤S24，将所述事故发生地标记为低风险事故发生地；当所述事故发生地的当前事故发生次数大于预设次数，且当前事故发生次数中车辆的损伤级别和所述车险图像中车辆的损伤级别都高于第一预设级别时，确认所述事故发生地属于高风险事故发生地，容易出现严重交通事故，执行步骤S25，将所述事故发生地标记为高风险事故发生地。
需要说明的是，若当所述事故发生地的当前事故发生次数小于等于预设次数，或者当前事故发生次数中车辆的损伤级别高于所述第一预设级别，而所述车险图像中车辆的损伤级别低于第一预设级别时，确认所述事故发生地发生车祸属于意外情况，并非高风险事故发生地，无需刻意提醒用户，执行步骤S24，将所述事故发生地标记为低风险事故发生地。
在第二实施例中，可以根据所述车险图像获取所述事故发生地的周围环境信息，并根据所述事故发生地的周围环境信息评估所述事故发生地的风险级别。
具体地，所述根据事故发生地的周围环境信息评估所述事故发生地的风险级别的方法包括：
(1)识别所述车险图像中路面环境信息；所述路面环境信息包括路面是否有异物(如碎石)、路面是否崎岖、及是否为急转弯等。在本实施例中，通过图像识别方法来识别所述车险图像中路面环境信息，所述图像识别方法为现有技术，在此不赘述。
(2)根据所述路面环境信息判断所述路面状况级别；
当所述路面有异物或者所述路面崎岖或者当前道路为急转弯时，确定所述路面状况糟糕，易发生交通事故；当路面没有异物且所述路面平坦且当前道路不是急转弯时，确定所述路面状况良好，不易发生交通事故。
(3)获取所述事故发生地的历史事故数据，其中，所述历史事故数据致使包括事故发生次数及所述历史事故中路面状况级别；
(4)判断所述事故发生地的当前事故发生次数是否大于所述预设次数，及判断当前事故发生次数中路面状况级别是否高于第二预设级别；
具体地，当所述事故发生地的当前事故发生次数小于或等于所述预设次数，或者当前事故发生次数中路面状况级别低于第二预设级别时，确认所述事故发生地发生车祸属于意外情况，并非高风险事故发生地，无需刻意提醒用户，流程进入步骤S3；当所述事故发生地的当前事故发生次数大于预设次数，且当前事故发生次数中路面状况级别高于第二预设级别时，确认所述事故发生地属于高风险事故发生地，容易出现严重交通事故，流程进入步骤S3。
在第三实施例中，所述基于智慧交通的道路信息提示方法可以根据所述事故发生地的当前事故发生次数，及所述车险图像中车辆的损伤级别，及所述事故发生地的周围环境信息，来评估所述事故发生地的风险级别。
具体地，若所述事故发生地的当前事故发生次数大于所述预设次数，且所述车辆的损伤级别高于所述第一预设级别，或所述事故发生地的当前事故发生次数大于所述预设次数且确定所述周围环境信息中路面状况级别高于所述第二预设级别时，执行步骤S3；若所述事故发生地的当前事故发生次数小于所述预设次数，且所述车辆的损伤级别低于所述第一预设级别，且所述确定所述周围环境信息中路面状况级别低于所述第二预设级别时，执行步骤S3。
步骤S3，根据所述事故发生地的风险级别标记所述事故发生地，得到高风险事故发生地信息和低风险事故发生地信息。
具体地，在第一实施例中，当所述事故发生地的当前事故发生次数小于等于预设次数，或者当前事故发生次数中车辆的损伤级别低于第一预设级别时，确认所述事故发生地发生车祸属于意外情况，并非高风险事故发生地，无需刻意提醒用户，将所述事故发生地标记为低风险事故发生地；当所述事故发生地的当前事故发生次数大于预设次数，且当前事故发生次数中车辆的损伤级别高于第一预设级别时，确认所述事故发生地属于高风险事故发生地，容易出现严重交通事故，将所述事故发生地标记为高风险事故发生地。
在第二实施例中，当所述事故发生地的当前事故发生次数小于等于所述预设次数，或者当前事故发生次数中路面状况级别低于第二预设级别时，确认所述事故发生地发生车祸属于意外情况，并非高风险事故发生地，无需刻意提醒用户，将所述事故发生地标记为低风险事故发生地；当所述事故发生地的当前事故发生次数大于预设次数，且当前事故发生次数中路面状况级别高于第二预设级别时，确认所述事故发生地属于高风险事故发生地，容易出现严重交通事故，将所述事故发生地标记为高风险事故发生地。
在第三实施例中，若所述事故发生地的当前事故发生次数大于所述预设次数，且所述车辆的损伤级别高于所述第一预设级别，或所述事故发生地的当前事故发生次数大于所述预设次数且确定所述周围环境信息中路面状况级别高于所述第二预设级别时，将所述事故发生地标记为高风险事故发生地；若所述事故发生地的当前事故发生次数小于所述预设次数，且所述车辆的损伤级别低于所述第一预设级别，且所述确定所述周围环境信息中路面状况级别低于所述第二预设级别时，将所述事故发生地标记为低风险事故发生地。
步骤S4，发送包括所述高风险事故发生地的警示信息至相关用户。
本实施例中，在确定所述事故发生地为高风险事故发生地时，可以发送包括所述高风险事故发生地的详细信息至与相关保险公司签订保险合同的用户，从而可以提示所述用户在以后的出行中尽量避开所述高风险事故发生地，从而可以减少交通事故的发生，也可以为保险公司减少理赔次数。
当然，也可以发送包括低风险事故发生地的详细信息至与相关保险公司签订保险合同的用户，从而可以提示所述用户在以后的出行中优先选择所述低风险事故发生地，从而可以减少交通事故的发生。
优选地，在将所述事故发生地标记为高风险事故发生地之后，可以发送所述高风险事故发生地信息至导航***。从而可以在用户驾驶车辆即将到达所述高风险事故发生地时，通过导航***发送语音提示信息提醒用户车辆即将到达高风险事故发生地。由此可以增加用户警惕性，提醒用户在驾驶车辆经过所述高风险事故发生地时倍加小心，避免发生交通事故。还可以在用户使用所述导航***规划出行路线时，给用户提供一条避开所述高风险事故发生地的路线。
优选地，所述基于智慧交通的道路信息提示方法还可以结合事故发生时间来提示相关用户。
当确定所述事故发生地为高风险事故发生地时，获取所述事故发生地的历史事故数据中的事故发生时间。
若所述历史事故数据中的事故发生时间集中在某时间段内，例如，早上5:00-9:00，则在所述时间段内提醒相关用户避开所述高风险事故发生地出行，从而可以减少交通事故的发生，也可以为保险公司减少理赔次数。
优选地，所述基于智慧交通的道路信息提示方法还可以结合事故发生时的天气状况来提醒相关用户。
当确定所述事故发生地为高风险事故发生地时，获取所述历史事故数据中事故发生地发生事故时的天气情况。
若所述历史事故数据中事故发生地发生事故时的天气恶劣，如出现大雾，或下大雨，或下大雪时，则提醒用户在恶劣天气应避开所述高风险事故发生地出行，从而可以减少交通事故的发生，也可以为保险公司减少理赔次数。
综上所述，本申请提供的基于智慧交通的道路信息提示方法，包括获取车险报案信息，其中，所述车险报案信息至少包括车险图像和事故发生地；根据所述车险图像评估所述事故发生地的风险级别；根据所述事故发生地的风险级别标记所述事故发生地；发送包括高风险事故发生地的警示信息至相关用户。可以统计事故多发地，并根据事故中车辆损伤级别提醒用户所述事故多发地的危险级别，使用户得到足够重视，并提醒用户出行时尽可能避开所述危险级别高的事故多发地，从而可以减少交通事故的发生。
以上所述，仅是本申请的具体实施方式，但本申请的保护范围并不局限于此，对于本领域的普通技术人员来说，在不脱离本申请创造构思的前提下，还可以做出改进，但这些均属于本申请的保护范围。
下面结合图4和图5，分别对实现上述基于智慧交通的道路信息提示方法的服务器的功能模块及硬件结构进行介绍。
实施例二
图4为本申请基于智慧交通的道路信息提示装置较佳实施例中的功能模块图。
在一些实施例中，所述基于智慧交通的道路信息提示装置40(为了便于描述，下文简称“道路信息提示装置40”)运行于服务器中。所述道路信息提示装置40可以包括多个由程序代码段所组成的功能模块。所述道路信息提示装置40中的各个程序段的程序代 码可以存储于存储器中，并由至少一个处理器所执行，以执行(详见图1及其相关描述)基于智慧交通的道路信息提示功能。
本实施例中，所述道路信息提示装置40根据其所执行的功能，可以被划分为多个功能模块。所述功能模块可以包括：获取模块401、评估模块402、标记模块403及发送模块404。本申请所称的模块是指一种能够被至少一个处理器所执行并且能够完成固定功能的一系列计算机可读指令段，其存储在存储器中。在一些实施例中，关于各模块的功能将在后续的实施例中详述。
所述获取模块401用于获取车险报案信息，其中，所述车险报案信息至少包括车险图像和事故发生地。
本实施例中，所述服务器可以从移动终端获取报案信息。所述移动终端可以是手机、平板电脑、个人数字助理、可穿戴设备(例如、智能手表、智能眼镜)等智能终端或其他任意适用的电子设备。所述报案信息可以包括车险图像和事故发生地，还可以包括车主的信息、出险车辆的车牌号、出险时间以及出险原因等。所述车险图像可以是车主拍摄的视频信息或图像信息。
在其他实施方式中，所述车险图像可以是作业人员(如查勘员)现场采集的视频信息或图像信息，所述作业人员将所述车险图像发送至其他***(如保险公司***)数据库中。所述服务器可以从所述其他***数据库中获取所述车险图像。所述车险图像与所述事故发生地相关联。
所述车险图像可以包括各种图形或影像的总称，通常指具有视觉效果的画面,一般可以包括纸介质上的、底片或照片上的、电视、投影仪或计算机屏幕等上的画面。本实施例中所述的车险图像可以包括通过照相或摄像设备拍摄后存储在可读存储介质上的计算机图像数据，可以包括矢量图、位图，静态、动态图像等多种类型的计算机图像。
优选地，在获取车险报案信息后，在获取车险报案信息后，所述基于智慧交通的道路信息提示装置40还可以提高车险图像的清晰度等级。所述提高车险图像的清晰度等级的方法包括：
(1)计算所述车险图像的第一清晰度等级；
在本实施方式中，可以通过灰度变化函数、梯度函数、或者图像灰度熵函数等计算来计算所述车险图像的第一清晰度等级。灰度变化函数、梯度函数、或者图像灰度熵函数为计算图像清晰度的现有技术，在此不再赘述。
(2)将所述第一清晰度等级与预设清晰度等级进行比对；当所述第一清晰度等级低于所述预设清晰度等级时，增强所述车险图像的清晰度等级以得到新的车险图像，并计算所述新的车险图像的第二清晰度等级；当所述第一清晰度等级高于所述预设清晰度等级时，根据所述车险图像中的车辆的损伤级别、所述事故发生地的周围环境信息和/或所述事故发生地的当前事故发生次数评估所述事故发生地的风险级别。
在本实施方式中，增强所述车险图像的清晰度等级以得到新的车险图像的方法包括：
a：计算所述车险图像中的高频分量和低频分量。
具体地，通过对所述车险图像的空域信号进行低通滤波后得到所述车险图像的低频分量，对所述车险图像的空域信号进行差值运算后得到所述车险图像的高频分量。
b：对所述车险图像中的高频分量进行识别，并对识别后的高频分量进行增强处理。
具体地，通过对所述车险图像的高频分量识别后进行分类，分出所述高频分量中的噪声、细节、小边缘和大边缘，再分别对所述高频分量中的噪声、细节、小边缘和大边缘进行增强处理。
所述对识别后的高频分量进行增强处理包括：
计算核化降噪动态阈值，判断所述车险图像中的点是否属于噪声。
通过将所述车险图像的点的高频分量的绝对值与所述核化降噪动态阈值进行比对，来判断所述车险图像的点是否属于噪声。
若所述车险图像中的点的高频分量的绝对值小于所述核化降噪阈值，则确认所述点为噪声，将所述点对应的高频分量的值设置为0。通过将所述点对应的高频分量的值设置为0以抑制掉小幅高频噪声，达到核化降噪的目的。
若所述车险图像中的点的高频分量的绝对值大于等于所述核化降噪阈值，则确认所述点不是噪声，应用非线性高频增强曲线对所述车险图像的高频分量进行增强。通过所述非线性高频增强曲线处理后可以对所述高频分量中的细节、小边缘和大边缘对应的不同区域段进行不同程度的处理，由此得到的增强图像，过渡平滑自然，且保持了高频分量的单调性。
c：将增强后的高频分量与所述低频分量叠加得到新的车险图像。
在增强所述车险图像的清晰度后得到新的车险图像，并计算所述新的车险图像的第二清晰度。可以理解的是，所述新的车险图像的第二清晰度的计算方法与所述车险图像的第一清晰度的计算方法一致，不再赘述。
(3)将所述第二清晰度等级与所述预设清晰度等级进行比对；当所述第二清晰度等级低于所述预设清晰度等级时，增强所述车险图像的清晰度等级以得到新的车险图像，并计算所述新的车险图像的第二清晰度等级；当所述第二清晰度等级高于所述预设清晰度等级时，根据所述车险图像评估所述事故发生地的风险级别。
可以理解的是，在计算所述车险图像的第一清晰度等级之前，所述基于智慧交通的道路信息提示方法还可以包括：对所述车险图像进行数据预处理的步骤，其中，所述数据预处理过程包括：模数转换、二值化、图像的平滑、变换、增强、恢复、滤波等。
在本实施方式中，通过对用户上传的车险图像的清晰度进行调整，从而得到满足理赔要求的车险图像，可以提高自助理赔***的工作效率。还可以避免由于用户上传的车险图像的清晰度不符合理赔要求时，麻烦用户重新上传车险图像的情况出现，提高了用户体验。
可以理解的是，在获取车险报案信息后，将所述车险图像和事故发生地关联后存储至服务器的数据库中。
所述评估模块402用于根据所述车险图像中的车辆的损伤级别、所述事故发生地的周围环境信息和/或所述事故发生地的当前事故发生次数评估所述事故发生地的风险级别。
在本实施方式中，可以根据车险图像中车辆的损伤级别，以及/或者事故发生地的周围环境信息来评估所述事故发生地的风险级别。
在第一实施例中，根据所述车险图像获取所述车险图像中车辆的损伤级别，并根据所述损伤级别评估所述事故发生地的风险级别。
具体地，根据所述车险图像中的车辆的损伤级别和所述事故发生地的当前事故发生次数评估所述事故发生地的风险级别的方法包括：
(a)获取所述车险图像中车辆的损伤级别。
本实施例中，调用预先训练生成的损伤区域识别模型识别所述车险图像，得到车辆损伤区域信息；根据所述车辆损伤区域信息计算所述损伤区域的损伤面积；将所述车辆损伤区域和损伤面积输入预设计算模型并得到计算结果，其中，所述预设计算模型为所述损伤区域与所述损伤区域的权重值的乘积加上所述损伤面积；当所述计算结果大于或等于预设值，确认所述车险图像中车辆的损伤级别高；当所述计算结果小于所述预设值，确认所述车险图像中车辆的损伤级别低。
具体地，调用损伤区域识别模型对车险图像进行损伤区域识别，再计算所述损伤区 域的损伤面积大小，根据所述损伤区域及对应的损伤面积采用预设计算模型来确定所述车险图像中车辆出现损伤的级别。
本实施例中，可以预先训练生成用于识别车险图像中的损伤区域的识别模型，所述识别模型可以是图像处理相关的多种模型中的一种。
优选地，所述损伤区域识别模型为卷积神经网络模型。
一般来说，所述车辆损伤区域可以包括第一区域、第二区域、第三区域、第四区域及第五区域。所述第一区域为直接碰撞损伤区(又称为一次损伤区)；所述第二区域为间接碰撞损伤区(又称为二次损伤区)；所述第三区域为机械损伤区，即汽车机械零件、动力传动***零件、附件等损伤区；所述第四区域为乘员舱区，及车厢的各种损坏，包括内饰件、灯、控制装置、操纵装置和饰层等；所述第五区域为外饰和漆面区，即车身外饰件及外部各种零部件的损伤。
优选地，所述损伤区域识别模型的训练过程包括：
1)获取预设数量的车险图像样本；
2)从所述车险图像样本中提取出预设比例的车险图像作为待训练的样本图片，并将所述预设数量的车险图像样本中剩余的车险图像样本作为待验证的样本图片；
3)利用各待训练的样本图片进行模型训练，以生成所述卷积神经网络模型，并利用各待验证的样本图片对所生成的卷积神经网络模型进行验证；
4)若验证通过率大于等于预设阈值，则训练完成，否则增加所述车险图像样本的数量，以重新进行训练及验证。
示例性的，假设获取10万张理赔车险图像样本图片。提取预设比例的理赔保单样本图片作为训练集，并将预设数量的车险图像样本图片中剩余的车险图像样本图片作为测试集，训练集中的车险图像样本图片的数量大于测试集中的车险图像样本图片的数量，例如将车险图像样本图片中的80％的车险图像样本图片作为训练集，将剩余的20％的车险图像样本图片作为测试集。
在第一次训练卷积神经网络模型时，所述卷积神经网络模型的参数采用默认的参数进行训练，在训练过程不断调整参数，在训练生成所述卷积神经网络模型后，利用各待验证的样本图片对所生成的卷积神经网络模型进行验证，如果验证通过率大于等于预设阈值，例如通过率大于等于98％，则训练结束，以所述训练得到的卷积神经网络模型为识别所述车险图像中车辆损伤区域的损伤区域识别模型；如果验证通过率小于预设阈值，例如小于98％，则增加车险图像样本的数量，并重新执行上述的步骤，直至验证通过率大于或者等于预设阈值。
在测试时，使用训练得到的卷积神经网络模型对从所述测试集中的车险图像样本中随机选取的预设个数(如十张)车险图像样本进行损伤区域识别，并将识别结果与人工确认的车辆损伤级别结果进行对比，以评估所训练的卷积神经网络模型的识别效果。
在识别所述车险图像中车辆的损伤区域后，所述评估模块402还用于计算所述损伤区域的损伤面积大小。
可以理解的是，在计算所述损伤区域的损伤面积大小时需要先计算所述车险图像中车辆的损伤区域，再按照一定比例计算得到所述损伤区域的实际面积大小。根据所述损伤区域、对应所述损伤区域的权重值及对应的损伤面积采用预设计算模型来确定所述车险图像中车辆出现损伤的级别。
示例性的，所述预设计算模型可以是所述损伤区域与所述损伤区域的权重值的乘积加上所述损伤面积。所述车险图像中车辆出现损伤的级别根据所述预设计算模型的计算结果大小来进行确认。当所述计算结果大于或等于预设值时，确认所述车险图像中车辆出现损伤的级别高；当所述计算结果小于所述预设值时，确认所述车险图像中车辆出现 损伤的级别低。
(b)获取所述事故发生地的历史事故数据，其中，所述历史事故数据致使包括事故发生次数及所述历史事故中车辆的损伤级别。
本实施例中，所述服务器的数据库中存储了历史报案信息中的车险图像及事故发生地。可以理解的是，所述历史事故数据还包括事故发生时间。
(c)判断所述事故发生地的当前事故发生次数是否大于预设次数，及判断所述车险图像中车辆的损伤级别和当前事故发生次数中车辆的损伤级别是否高于第一预设级别。
本实施例中，为了解决现有技术中仅提示用户某地段为事故多发地，并不能引起用户足够重视而在所述事故多发地依然出现交通事故的问题。本方案不仅统计所述事故发生地发生事故的次数，还可以确定所述事故中车辆的损伤级别，将所述事故发生次数与车辆损伤级别进行关联来评定所述地段是否为高风险事故发生地。
具体地，当所述事故发生地的当前事故发生次数小于或等于预设次数，或者当前事故发生次数中车辆的损伤级别和所述车险图像中车辆的损伤级别都低于第一预设级别时，确认所述事故发生地发生车祸属于意外情况，并非高风险事故发生地，无需刻意提醒用户，将所述事故发生地标记为低风险事故发生地；当所述事故发生地的当前事故发生次数大于预设次数，且当前事故发生次数中车辆的损伤级别和所述车险图像中车辆的损伤级别都高于第一预设级别时，确认所述事故发生地属于高风险事故发生地，容易出现严重交通事故，将所述事故发生地标记为高风险事故发生地。
需要说明的是，若当所述事故发生地的当前事故发生次数小于等于预设次数，或者当前事故发生次数中车辆的损伤级别高于所述第一预设级别，而所述车险图像中车辆的损伤级别低于第一预设级别时，确认所述事故发生地发生车祸属于意外情况，并非高风险事故发生地，无需刻意提醒用户，将所述事故发生地标记为低风险事故发生地。
在第二实施例中，可以根据所述车险图像获取所述事故发生地的周围环境信息，并根据所述事故发生地的周围环境信息评估所述事故发生地的风险级别。
具体地，所述根据事故发生地的周围环境信息评估所述事故发生地的风险级别的方法包括：
(1)识别所述车险图像中路面环境信息。所述路面环境信息包括路面是否有异物(如碎石)、路面是否崎岖、及是否为急转弯等。在本实施例中，通过图像识别方法来识别所述车险图像中路面环境信息，所述图像识别方法为现有技术，在此不赘述。
(2)根据所述路面环境信息判断所述路面状况级别。
当所述路面有异物或者所述路面崎岖或者当前道路为急转弯时，确定所述路面状况糟糕，易发生交通事故；当路面没有异物且所述路面平坦且当前道路不是急转弯时，确定所述路面状况良好，不易发生交通事故。
(3)获取所述事故发生地的历史事故数据，其中，所述历史事故数据致使包括事故发生次数及所述历史事故中路面状况级别。
(4)判断所述事故发生地的当前事故发生次数是否大于所述预设次数，及判断当前事故发生次数中路面状况级别是否高于第二预设级别。
具体地，当所述事故发生地的当前事故发生次数小于等于所述预设次数，或者当前事故发生次数中路面状况级别低于第二预设级别时，确认所述事故发生地发生车祸属于意外情况，并非高风险事故发生地，无需刻意提醒用户，根据所述事故发生地的风险级别标记所述事故发生地；当所述事故发生地的当前事故发生次数大于预设次数，且当前事故发生次数中路面状况级别高于第二预设级别时，确认所述事故发生地属于高风险事故发生地，容易出现严重交通事故，根据所述事故发生地的风险级别标记所述事故发生地。
在第三实施例中，所述道路信息提示装置40可以根据所述事故发生地的当前事故发生次数，及所述车险图像中车辆的损伤级别，及所述事故发生地的周围环境信息，来评估所述事故发生地的风险级别。
具体地，若所述事故发生地的当前事故发生次数大于所述预设次数，且所述车辆的损伤级别高于所述第一预设级别，或所述事故发生地的当前事故发生次数大于所述预设次数且确定所述周围环境信息中路面状况级别高于所述第二预设级别时，根据所述事故发生地的风险级别标记所述事故发生地；若所述事故发生地的当前事故发生次数小于所述预设次数，且所述车辆的损伤级别低于所述第一预设级别，且所述确定所述周围环境信息中路面状况级别低于所述第二预设级别时，根据所述事故发生地的风险级别标记所述事故发生地。
所述标记模块403用于根据所述事故发生地的风险级别标记所述事故发生地，得到高风险事故发生地信息和低风险事故发生地信息。
具体地，在第一实施例中，当所述事故发生地的当前事故发生次数小于等于预设次数，或者当前事故发生次数中车辆的损伤级别低于第一预设级别时，确认所述事故发生地发生车祸属于意外情况，并非高风险事故发生地，无需刻意提醒用户，将所述事故发生地标记为低风险事故发生地；当所述事故发生地的当前事故发生次数大于预设次数，且当前事故发生次数中车辆的损伤级别高于第一预设级别时，确认所述事故发生地属于高风险事故发生地，容易出现严重交通事故，将所述事故发生地标记为高风险事故发生地。
在第二实施例中，当所述事故发生地的当前事故发生次数小于或等于所述预设次数，或者当前事故发生次数中路面状况级别低于第二预设级别时，确认所述事故发生地发生车祸属于意外情况，并非高风险事故发生地，无需刻意提醒用户，将所述事故发生地标记为低风险事故发生地；当所述事故发生地的当前事故发生次数大于预设次数，且当前事故发生次数中路面状况级别高于第二预设级别时，确认所述事故发生地属于高风险事故发生地，容易出现严重交通事故，将所述事故发生地标记为高风险事故发生地。
在第三实施例中，若所述事故发生地的当前事故发生次数大于所述预设次数，且所述车辆的损伤级别高于所述第一预设级别，或所述事故发生地的当前事故发生次数大于所述预设次数且确定所述周围环境信息中路面状况级别高于所述第二预设级别时，将所述事故发生地标记为高风险事故发生地；若所述事故发生地的当前事故发生次数小于所述预设次数，且所述车辆的损伤级别低于所述第一预设级别，且所述确定所述周围环境信息中路面状况级别低于所述第二预设级别时，将所述事故发生地标记为低风险事故发生地。
所述发送模块404用于发送包括所述高风险事故发生地的警示信息至相关用户。
本实施例中，在确定所述事故发生地为高风险事故发生地时，可以发送包括所述高风险事故发生地的详细信息至与相关保险公司签订保险合同的用户，从而可以提示所述用户在以后的出行中尽量避开所述高风险事故发生地，从而可以减少交通事故的发生，也可以为保险公司减少理赔次数。
优选地，在将所述事故发生地标记为高风险事故发生地之后，可以发送所述高风险事故发生地信息至导航***。从而可以在用户驾驶车辆即将到达所述高风险事故发生地时，通过导航***发送语音提示信息提醒用户车辆即将到达高风险事故发生地。由此可以增加用户警惕性，提醒用户在驾驶车辆经过所述高风险事故发生地时倍加小心，避免发生交通事故。还可以在用户使用所述导航***规划出行路线时，给用户提供一条避开所述高风险事故发生地的路线。
优选地，所述道路信息提示装置40还可以结合事故发生时间来提示相关用户。
当确定所述事故发生地为高风险事故发生地时，获取所述事故发生地的历史事故数据中的事故发生时间；若所述历史事故数据中的事故发生时间集中在某时间段内，例如，早上5:00-9:00，则在所述时间段内提醒相关用户避开所述高风险事故发生地出行，从而可以减少交通事故的发生，也可以为保险公司减少理赔次数。
优选地，所述道路信息提示装置40还可以结合事故发生时的天气状况来提醒相关用户。
当确定所述事故发生地为高风险事故发生地时，获取所述历史事故数据中事故发生地发生事故时的天气情况。
若所述历史事故数据中事故发生地发生事故时的天气恶劣，如出现大雾，或下大雨，或下大雪时，则提醒用户在恶劣天气应避开所述高风险事故发生地出行，从而可以减少交通事故的发生，也可以为保险公司减少理赔次数。
综上所述，本申请提供的道路信息提示装置40可以统计事故多发地，并根据事故中车辆损伤级别提醒用户所述事故多发地的危险级别，使用户得到足够重视，并提醒用户出行时尽可能避开所述危险级别高的事故多发地，从而可以减少交通事故的发生。
上述以软件功能模块的形式实现的集成的单元，可以存储在一个非易失性可读取存储介质中。上述软件功能模块存储在一个存储介质中，包括若干指令用以使得一台计算机设备(可以是个人计算机，双屏设备，或者网络设备等)或处理器(processor)执行本申请各个实施例所述方法的部分。
实施例三
图5为本申请实施例三提供的服务器的示意图。
所述服务器5包括：数据库51、存储器52、至少一个处理器53、存储在所述存储器52中并可在所述至少一个处理器53上运行的计算机可读指令54及至少一条通讯总线55。
所述至少一个处理器53执行所述计算机可读指令54时实现上述基于智慧交通的道路信息提示方法实施例中的步骤。
示例性的，所述计算机可读指令54可以被分割成一个或多个模块/单元，所述一个或者多个模块/单元被存储在所述存储器52中，并由所述至少一个处理器53执行，以完成本申请。所述一个或多个模块/单元可以是能够完成特定功能的一系列计算机可读指令段，该指令段用于描述所述计算机可读指令54在所述服务器5中的执行过程。
所述服务器5是一种能够按照事先设定或存储的指令，自动进行数值计算和/或信息处理的设备，其硬件包括但不限于微处理器、专用集成电路(Application Specific Integrated Circuit，ASIC)、可编程门阵列(Field－Programmable Gate Array，FPGA)、数字处理器(Digital Signal Processor，DSP)、嵌入式设备等。本领域技术人员可以理解，所述示意图5仅仅是服务器5的示例，并不构成对服务器5的限定，可以包括比图示更多或更少的部件，或者组合某些部件，或者不同的部件，例如所述服务器5还可以包括输入输出设备、网络接入设备、总线等。
所述数据库(Database)51是按照数据结构来组织、存储和管理数据的建立在所述服务器5上的仓库。数据库通常分为层次式数据库、网络式数据库和关系式数据库三种。在本实施方式中，所述数据库51用于存储所述车险图像信息。
所述至少一个处理器53可以是中央处理单元(Central Processing Unit，CPU)，还可以是其他通用处理器、数字信号处理器(Digital Signal Processor，DSP)、专用集成电路(Application Specific Integrated Circuit，ASIC)、现成可编程门阵列(Field-Programmable Gate Array，FPGA)或者其他可编程逻辑器件、分立门或者晶体管逻辑器件、分立硬件组件等。该处理器53可以是微处理器或者该处理器53也可以是任何常规的处理器等， 所述处理器53是所述服务器5的控制中心，利用各种接口和线路连接整个服务器5的各个部分。
所述存储器52可用于存储所述计算机可读指令54和/或模块/单元，所述处理器53通过运行或执行存储在所述存储器52内的计算机可读指令和/或模块/单元，以及调用存储在存储器52内的数据，实现所述服务器5的各种功能。所述存储器31包括只读存储器(Read-Only Memory，ROM)、可编程只读存储器(Programmable Read-Only Memory，PROM)、可擦除可编程只读存储器(Erasable Programmable Read-Only Memory，EPROM)、一次可编程只读存储器(One-time Programmable Read-Only Memory，OTPROM)、电子擦除式可复写只读存储器(Electrically-Erasable Programmable Read-Only Memory，EEPROM)、只读光盘(Compact Disc Read-Only Memory，CD-ROM)或其他光盘存储器、磁盘存储器、磁带存储器、或者任何其他能够用于携带或存储数据的非易失性可读的存储介质。
所述存储器52中存储有程序代码，且所述至少一个处理器53可调用所述存储器52中存储的程序代码以执行相关的功能。例如，图4中所述的各个模块(获取模块401、评估模块402、标记模块403及发送模块404)是存储在所述存储器52中的程序代码，并由所述至少一个处理器53所执行，从而实现所述各个模块的功能以达到基于智慧交通的道路信息提示目的。
所述服务器5集成的模块/单元如果以软件功能单元的形式实现并作为独立的产品销售或使用时，可以存储在一个非易失性可读取存储介质中。基于这样的理解，本申请实现上述实施例方法中的全部或部分流程，也可以通过计算机可读指令来指令相关的硬件来完成，所述的计算机可读指令可存储于一非易失性可读存储介质中，该计算机可读指令在被处理器执行时，可实现上述各个方法实施例的步骤。其中，所述计算机可读指令包括计算机可读指令代码，所述计算机可读指令代码可以为源代码形式、对象代码形式、可执行文件或某些中间形式等。所述非易失性可读介质可以包括：能够携带所述计算机可读指令代码的任何实体或装置、记录介质、U盘、移动硬盘、磁盘、光盘、计算机存储器、只读存储器(ROM，Read-Only Memory)。
尽管未示出，所述服务器5还可以包括给各个部件供电的电源(比如电池)，优选的，电源可以通过电源管理***与所述至少一个处理器53逻辑相连，从而通过电源管理***实现管理充电、放电、以及功耗管理等功能。电源还可以包括一个或一个以上的直流或交流电源、再充电***、电源故障检测电路、电源转换器或者逆变器、电源状态指示器等任意组件。所述服务器5还可以包括蓝牙模块、Wi-Fi模块等，在此不再赘述。
最后应说明的是，以上实施例仅用以说明本申请的技术方案而非限制，尽管参照较佳实施例对本申请进行了详细说明，本领域的普通技术人员应当理解，可以对本申请的技术方案进行修改或等同替换，而不脱离本申请技术方案的精神范围。

Claims (20)

  1. 一种基于智慧交通的道路信息提示方法，其特征在于，所述方法包括：
    获取车险报案信息，其中，所述车险报案信息至少包括车险图像和事故发生地；
    根据所述车险图像中的车辆的损伤级别、所述事故发生地的周围环境信息和/或所述事故发生地的当前事故发生次数评估所述事故发生地的风险级别；
    根据所述事故发生地的风险级别标记所述事故发生地，得到高风险事故发生地信息和低风险事故发生地信息；
    发送包括所述高风险事故发生地的警示信息至相关用户。
  2. 如权利要求1所述的基于智慧交通的道路信息提示方法，其特征在于，在所述获取车险报案信息后，所述方法还包括：
    计算所述车险图像的第一清晰度等级；
    将所述第一清晰度等级与预设清晰度等级进行比对；
    当所述第一清晰度等级低于所述预设清晰度等级时，增强所述车险图像的清晰度等级以得到新的车险图像，并计算所述新的车险图像的第二清晰度等级；
    将所述第二清晰度等级与所述预设清晰度等级进行比对；
    当所述第二清晰度等级高于所述预设清晰度等级时，执行所述根据所述车险图像中的车辆的损伤级别、所述事故发生地的周围环境信息和/或所述事故发生地的当前事故发生次数评估所述事故发生地的风险级别的步骤。
  3. 如权利要求1所述的基于智慧交通的道路信息提示方法，其特征在于，根据所述车险图像中的车辆的损伤级别和所述事故发生地的当前事故发生次数评估所述事故发生地的风险级别包括：
    获取所述车险图像中车辆的损伤级别；
    获取所述事故发生地的历史事故数据，其中，所述历史事故数据包括事故发生次数及所述历史事故中车辆的损伤级别；
    判断所述事故发生地的当前事故发生次数是否大于预设次数，及判断所述车险图像中车辆的损伤级别和当前事故发生次数中车辆的损伤级别是否高于第一预设级别；
    当所述事故发生地的当前事故发生次数小于或等于所述预设次数，或者所述车险图像中车辆的损伤级别和当前事故发生次数中车辆的损伤级别都低于第一预设级别时，将所述事故发生地标记为低风险事故发生地；
    当所述事故发生地的当前事故发生次数大于预设次数，且所述车险图像中车辆的损伤级别和当前事故发生次数中车辆的损伤级别都高于第一预设级别时，将所述事故发生地标记为高风险事故发生地。
  4. 如权利要求3所述的基于智慧交通的道路信息提示方法，其特征在于，所述获取所述车险图像中车辆的损伤级别的步骤包括：
    调用预先训练生成的损伤区域识别模型识别所述车险图像，得到车辆损伤区域信息；
    根据所述车辆损伤区域信息计算所述损伤区域的损伤面积；
    将所述车辆损伤区域和损伤面积输入预设计算模型并得到计算结果，其中，所述预设计算模型为所述损伤区域与所述损伤区域的权重值的乘积加上所述损伤面积；
    判断所述计算结果是否大于或等于预设值；
    当所述计算结果大于或等于所述预设值，确认所述车险图像中车辆的损伤级别高；
    当所述计算结果小于所述预设值，确认所述车险图像中车辆的损伤级别低。
  5. 如权利要求3所述的基于智慧交通的道路信息提示方法，其特征在于，根据所述事故 发生地的周围环境信息和所述事故发生地的当前事故发生次数评估所述事故发生地的风险级别包括：
    识别所述车险图像中路面环境信息，其中，所述路面环境信息包括路面是否有异物、路面是否崎岖及是否为急转弯；
    根据所述路面环境信息判断所述路面状况级别；
    获取所述事故发生地的历史事故数据，其中，所述历史事故数据至少包括事故发生次数及所述历史事故中路面状况级别；
    判断所述事故发生地的当前事故发生次数是否大于所述预设次数，及判断当前事故发生次数中路面状况级别是否高于第二预设级别；
    当所述事故发生地的当前事故发生次数小于或等于所述预设次数，或者当前事故发生次数中路面状况级别低于第二预设级别时，将所述事故发生地标记为低风险事故发生地；
    当所述事故发生地的当前事故发生次数大于所述预设次数，且当前事故发生次数中路面状况级别高于所述第二预设级别时，将所述事故发生地标记为高风险事故发生地。
  6. 如权利要求5所述的基于智慧交通的道路信息提示方法，其特征在于：
    若所述事故发生地的当前事故发生次数大于所述预设次数，且所述车辆的损伤级别高于所述第一预设级别，或所述事故发生地的当前事故发生次数大于所述预设次数且确定所述周围环境信息中路面状况级别高于所述第二预设级别时，将所述事故发生地标记为高风险事故发生地；
    若所述事故发生地的当前事故发生次数小于或等于所述预设次数，且所述车辆的损伤级别低于所述第一预设级别，且所述确定所述周围环境信息中路面状况级别低于所述第二预设级别时，将所述事故发生地标记为低风险事故发生地。
  7. 如权利要求5所述的基于智慧交通的道路信息提示方法，其特征在于，所述根据所述路面环境信息判断所述路面状况级别包括：
    当所述路面有异物或者所述路面崎岖或者当前道路为急转弯时，确定所述路面状况级别高；
    当路面没有异物且所述路面平坦且当前道路不是急转弯时，确定所述路面状况级别低。
  8. 一种基于智慧交通的道路信息提示装置，其特征在于，所述装置包括：
    获取模块，用于获取车险报案信息，其中，所述车险报案信息至少包括车险图像和事故发生地；
    评估模块，用于根据所述车险图像中的车辆的损伤级别、所述事故发生地的周围环境信息和/或所述事故发生地的当前事故发生次数评估所述事故发生地的风险级别；
    标记模块，用于根据所述事故发生地的风险级别标记所述事故发生地，得到高风险事故发生地信息和低风险事故发生地信息；
    发送模块，用于发送包括所述高风险事故发生地的警示信息至相关用户。
  9. 一种服务器，其特征在于，所述服务器包括处理器和存储器，所述处理器用于执行存储器中存储的至少一个计算机可读指令时实现以下步骤：
    获取车险报案信息，其中，所述车险报案信息至少包括车险图像和事故发生地；
    根据所述车险图像中的车辆的损伤级别、所述事故发生地的周围环境信息和/或所述事故发生地的当前事故发生次数评估所述事故发生地的风险级别；
    根据所述事故发生地的风险级别标记所述事故发生地，得到高风险事故发生地信息和低风险事故发生地信息；
    发送包括所述高风险事故发生地的警示信息至相关用户。
  10. 如权利要求9所述的服务器，其特征在于，在所述获取车险报案信息后，所述处理器执行所述至少一个计算机可读指令时还用以实现以下步骤：
    计算所述车险图像的第一清晰度等级；
    将所述第一清晰度等级与预设清晰度等级进行比对；
    当所述第一清晰度等级低于所述预设清晰度等级时，增强所述车险图像的清晰度等级以得到新的车险图像，并计算所述新的车险图像的第二清晰度等级；
    将所述第二清晰度等级与所述预设清晰度等级进行比对；
    当所述第二清晰度等级高于所述预设清晰度等级时，执行所述根据所述车险图像中的车辆的损伤级别、所述事故发生地的周围环境信息和/或所述事故发生地的当前事故发生次数评估所述事故发生地的风险级别。
  11. 如权利要求9所述的服务器，其特征在于，所述根据所述车险图像中的车辆的损伤级别和所述事故发生地的当前事故发生次数评估所述事故发生地的风险级别时，所述处理器执行所述至少一个计算机可读指令以实现以下步骤：
    获取所述车险图像中车辆的损伤级别；
    获取所述事故发生地的历史事故数据，其中，所述历史事故数据包括事故发生次数及所述历史事故中车辆的损伤级别；
    判断所述事故发生地的当前事故发生次数是否大于预设次数，及判断所述车险图像中车辆的损伤级别和当前事故发生次数中车辆的损伤级别是否高于第一预设级别；
    当所述事故发生地的当前事故发生次数小于或等于所述预设次数，或者所述车险图像中车辆的损伤级别和当前事故发生次数中车辆的损伤级别都低于第一预设级别时，将所述事故发生地标记为低风险事故发生地；
    当所述事故发生地的当前事故发生次数大于预设次数，且所述车险图像中车辆的损伤级别和当前事故发生次数中车辆的损伤级别都高于第一预设级别时，将所述事故发生地标记为高风险事故发生地。
  12. 如权利要求11所述的服务器，其特征在于，所述获取所述车险图像中车辆的损伤级别时，所述处理器执行所述至少一个计算机可读指令以实现以下步骤：
    调用预先训练生成的损伤区域识别模型识别所述车险图像，得到车辆损伤区域信息；
    根据所述车辆损伤区域信息计算所述损伤区域的损伤面积；
    将所述车辆损伤区域和损伤面积输入预设计算模型并得到计算结果，其中，所述预设计算模型为所述损伤区域与所述损伤区域的权重值的乘积加上所述损伤面积；
    判断所述计算结果是否大于或等于预设值；
    当所述计算结果大于或等于所述预设值，确认所述车险图像中车辆的损伤级别高；
    当所述计算结果小于所述预设值，确认所述车险图像中车辆的损伤级别低。
  13. 如权利要求11所述的服务器，其特征在于，所述根据所述事故发生地的周围环境信息和所述事故发生地的当前事故发生次数评估所述事故发生地的风险级别时，所述处理器执行所述至少一个计算机可读指令以实现以下步骤：
    识别所述车险图像中路面环境信息，其中，所述路面环境信息包括路面是否有异物、路面是否崎岖及是否为急转弯；
    根据所述路面环境信息判断所述路面状况级别；
    获取所述事故发生地的历史事故数据，其中，所述历史事故数据至少包括事故发生次数及所述历史事故中路面状况级别；
    判断所述事故发生地的当前事故发生次数是否大于所述预设次数，及判断当前事故发生次数中路面状况级别是否高于第二预设级别；
    当所述事故发生地的当前事故发生次数小于或等于所述预设次数，或者当前事故发生次数中路面状况级别低于第二预设级别时，将所述事故发生地标记为低风险事故发生地；
    当所述事故发生地的当前事故发生次数大于所述预设次数，且当前事故发生次数中路面状况级别高于所述第二预设级别时，将所述事故发生地标记为高风险事故发生地。
  14. 如权利要求13所述的服务器，其特征在于，所述处理器执行所述至少一个计算机可读指令还用以实现以下步骤：
    若所述事故发生地的当前事故发生次数大于所述预设次数，且所述车辆的损伤级别高于所述第一预设级别，或所述事故发生地的当前事故发生次数大于所述预设次数且确定所述周围环境信息中路面状况级别高于所述第二预设级别时，将所述事故发生地标记为高风险事故发生地；
    若所述事故发生地的当前事故发生次数小于或等于所述预设次数，且所述车辆的损伤级别低于所述第一预设级别，且所述确定所述周围环境信息中路面状况级别低于所述第二预设级别时，将所述事故发生地标记为低风险事故发生地。
  15. 一种非易失性可读存储介质，其特征在于，所述非易失性可读存储介质存储有至少一个计算机可读指令，所述至少一个计算机可读指令被处理器执行时实现以下步骤：
    获取车险报案信息，其中，所述车险报案信息至少包括车险图像和事故发生地；
    根据所述车险图像中的车辆的损伤级别、所述事故发生地的周围环境信息和/或所述事故发生地的当前事故发生次数评估所述事故发生地的风险级别；
    根据所述事故发生地的风险级别标记所述事故发生地，得到高风险事故发生地信息和低风险事故发生地信息；
    发送包括所述高风险事故发生地的警示信息至相关用户。
  16. 如权利要求15所述的存储介质，其特征在于，在所述获取车险报案信息后，所述至少一个计算机可读指令被所述处理器执行还用以实现以下步骤：
    计算所述车险图像的第一清晰度等级；
    将所述第一清晰度等级与预设清晰度等级进行比对；
    当所述第一清晰度等级低于所述预设清晰度等级时，增强所述车险图像的清晰度等级以得到新的车险图像，并计算所述新的车险图像的第二清晰度等级；
    将所述第二清晰度等级与所述预设清晰度等级进行比对；
    当所述第二清晰度等级高于所述预设清晰度等级时，执行所述根据所述车险图像中的车辆的损伤级别、所述事故发生地的周围环境信息和/或所述事故发生地的当前事故发生次数评估所述事故发生地的风险级别。
  17. 如权利要求15所述的存储介质，其特征在于，所述根据所述车险图像中的车辆的损伤级别和所述事故发生地的当前事故发生次数评估所述事故发生地的风险级别时，所述至少一个计算机可读指令被所述处理器执行以实现以下步骤：
    获取所述车险图像中车辆的损伤级别；
    获取所述事故发生地的历史事故数据，其中，所述历史事故数据包括事故发生次数及所述历史事故中车辆的损伤级别；
    判断所述事故发生地的当前事故发生次数是否大于预设次数，及判断所述车险图像中车辆的损伤级别和当前事故发生次数中车辆的损伤级别是否高于第一预设级别；
    当所述事故发生地的当前事故发生次数小于或等于所述预设次数，或者所述车险图像中车辆的损伤级别和当前事故发生次数中车辆的损伤级别都低于第一预设级别时，将所述事故发生地标记为低风险事故发生地；
    当所述事故发生地的当前事故发生次数大于预设次数，且所述车险图像中车辆的损伤级别和当前事故发生次数中车辆的损伤级别都高于第一预设级别时，将所述事故发生地标记为高风险事故发生地。
  18. 如权利要求17所述的存储介质，其特征在于，所述获取所述车险图像中车辆的损伤 级别时，所述至少一个计算机可读指令被所述处理器执行以实现以下步骤：
    调用预先训练生成的损伤区域识别模型识别所述车险图像，得到车辆损伤区域信息；
    根据所述车辆损伤区域信息计算所述损伤区域的损伤面积；
    将所述车辆损伤区域和损伤面积输入预设计算模型并得到计算结果，其中，所述预设计算模型为所述损伤区域与所述损伤区域的权重值的乘积加上所述损伤面积；
    判断所述计算结果是否大于或等于预设值；
    当所述计算结果大于或等于所述预设值，确认所述车险图像中车辆的损伤级别高；
    当所述计算结果小于所述预设值，确认所述车险图像中车辆的损伤级别低。
  19. 如权利要求17所述的存储介质，其特征在于，所述根据所述事故发生地的周围环境信息和所述事故发生地的当前事故发生次数评估所述事故发生地的风险级别时，所述至少一个计算机可读指令被所述处理器执行以实现以下步骤：
    识别所述车险图像中路面环境信息，其中，所述路面环境信息包括路面是否有异物、路面是否崎岖及是否为急转弯；
    根据所述路面环境信息判断所述路面状况级别；
    获取所述事故发生地的历史事故数据，其中，所述历史事故数据至少包括事故发生次数及所述历史事故中路面状况级别；
    判断所述事故发生地的当前事故发生次数是否大于所述预设次数，及判断当前事故发生次数中路面状况级别是否高于第二预设级别；
    当所述事故发生地的当前事故发生次数小于或等于所述预设次数，或者当前事故发生次数中路面状况级别低于第二预设级别时，将所述事故发生地标记为低风险事故发生地；
    当所述事故发生地的当前事故发生次数大于所述预设次数，且当前事故发生次数中路面状况级别高于所述第二预设级别时，将所述事故发生地标记为高风险事故发生地。
  20. 如权利要求19所述的存储介质，其特征在于，所述至少一个计算机可读指令被所述处理器执行还用以实现以下步骤：
    若所述事故发生地的当前事故发生次数大于所述预设次数，且所述车辆的损伤级别高于所述第一预设级别，或所述事故发生地的当前事故发生次数大于所述预设次数且确定所述周围环境信息中路面状况级别高于所述第二预设级别时，将所述事故发生地标记为高风险事故发生地；
    若所述事故发生地的当前事故发生次数小于或等于所述预设次数，且所述车辆的损伤级别低于所述第一预设级别，且所述确定所述周围环境信息中路面状况级别低于所述第二预设级别时，将所述事故发生地标记为低风险事故发生地。
PCT/CN2019/117417 2019-07-25 2019-11-12 基于智慧交通的道路信息提示方法、装置、服务器及介质 WO2021012504A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201910678463.6 2019-07-25
CN201910678463.6A CN110473418B (zh) 2019-07-25 2019-07-25 危险路段识别方法、装置、服务器及存储介质

Publications (1)

Publication Number Publication Date
WO2021012504A1 true WO2021012504A1 (zh) 2021-01-28

Family

ID=68509564

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2019/117417 WO2021012504A1 (zh) 2019-07-25 2019-11-12 基于智慧交通的道路信息提示方法、装置、服务器及介质

Country Status (2)

Country Link
CN (1) CN110473418B (zh)
WO (1) WO2021012504A1 (zh)

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20210312566A1 (en) * 2019-12-27 2021-10-07 Capital One Services, Llc Systems and methods for predictive model generation
CN114814888A (zh) * 2022-06-30 2022-07-29 南京源兴智达信息科技有限公司 一种基于北斗三号的手持终端设备
CN115240408A (zh) * 2022-06-15 2022-10-25 中国第一汽车股份有限公司 一种基于区块链技术的车辆事故预警方法、***、电子设备和存储介质
CN116935308A (zh) * 2023-07-10 2023-10-24 南京易自助网络科技有限公司 基于车辆情景ai智能识别的洗车机安全监控***及方法
CN117745124A (zh) * 2023-11-29 2024-03-22 交通运输部规划研究院 智慧高速公路评价方法及装置

Families Citing this family (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110992718A (zh) * 2019-12-28 2020-04-10 浙江永尚光电有限公司 一种运用于路灯的道路智能控制端及方法
CN111815986B (zh) * 2020-09-02 2021-01-01 深圳市城市交通规划设计研究中心股份有限公司 一种交通事故预警方法、装置、终端设备及存储介质
CN112270836B (zh) * 2020-10-15 2023-11-21 腾讯科技（深圳）有限公司 车流量控制方法、装置、设备及存储介质
CN112633713A (zh) * 2020-12-28 2021-04-09 广州羊城通有限公司 基于乘车记录的风险地区提醒方法及装置
CN114582132B (zh) * 2022-05-05 2022-08-09 四川九通智路科技有限公司 一种基于机器视觉的车辆碰撞检测预警***及方法
CN117057606A (zh) * 2023-08-15 2023-11-14 广州地铁设计研究院股份有限公司 风险预测模型训练方法、风险预测方法及相关设备

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20140375446A1 (en) * 2013-06-20 2014-12-25 Denso Corporation Accident report system
CN108337477A (zh) * 2017-01-18 2018-07-27 通用汽车环球科技运作有限责任公司 车辆环境成像***和方法
CN108492556A (zh) * 2018-03-26 2018-09-04 京东方科技集团股份有限公司 一种交通事故规避信息的发布方法及云服务器
CN108513674A (zh) * 2018-03-26 2018-09-07 深圳市锐明技术股份有限公司 一种车前积雪与结冰的检测报警方法、存储介质和服务器
CN109933062A (zh) * 2017-12-15 2019-06-25 百度（美国）有限责任公司 自动驾驶车辆的报警***
CN110033011A (zh) * 2018-12-14 2019-07-19 阿里巴巴集团控股有限公司 车祸事故处理方法和装置、电子设备

Family Cites Families (15)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US10304137B1 (en) * 2012-12-27 2019-05-28 Allstate Insurance Company Automated damage assessment and claims processing
CN105975669B (zh) * 2016-04-29 2019-08-27 大连楼兰科技股份有限公司 基于cae碰撞仿真的汽车零部件损伤评定方法及装置
CN106296027A (zh) * 2016-08-18 2017-01-04 深圳市永兴元科技有限公司 车险理赔预警方法及装置
CN106251667B (zh) * 2016-08-18 2019-04-02 深圳市永兴元科技股份有限公司 车辆驾驶提示方法及装置
CN106652562B (zh) * 2017-02-09 2021-09-10 吉林大学 一种高速公路道路交通安全预警方法
CN112435215B (zh) * 2017-04-11 2024-02-13 创新先进技术有限公司 一种基于图像的车辆定损方法、移动终端、服务器
CN107194323B (zh) * 2017-04-28 2020-07-03 阿里巴巴集团控股有限公司 车辆定损图像获取方法、装置、服务器和终端设备
KR102414456B1 (ko) * 2017-10-23 2022-06-30 현대자동차주식회사 대화 시스템, 이를 포함하는 차량 및 유고 정보 처리 방법
CN108090838B (zh) * 2017-11-21 2020-09-29 阿里巴巴集团控股有限公司 识别车辆受损部件的方法、装置、服务器、客户端及***
CN108734702A (zh) * 2018-04-26 2018-11-02 平安科技（深圳）有限公司 车损判定方法、服务器及存储介质
CN108769624B (zh) * 2018-07-09 2020-09-22 北京精友世纪软件技术有限公司 一种智能车险移动视频查勘***
CN109523389A (zh) * 2018-09-19 2019-03-26 平安科技（深圳）有限公司 基于图像识别的车损处理方法、装置、设备及介质
CN109492143A (zh) * 2018-09-21 2019-03-19 平安科技（深圳）有限公司 图像数据处理方法、装置、计算机设备及存储介质
CN109767597B (zh) * 2019-01-19 2021-05-11 跨越速运集团有限公司 一种车辆事故预警方法及***
CN109872554A (zh) * 2019-03-22 2019-06-11 山西省交通科学研究院有限公司 一种高速公路雾区提升行车安全的车路预警***

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20140375446A1 (en) * 2013-06-20 2014-12-25 Denso Corporation Accident report system
CN108337477A (zh) * 2017-01-18 2018-07-27 通用汽车环球科技运作有限责任公司 车辆环境成像***和方法
CN109933062A (zh) * 2017-12-15 2019-06-25 百度（美国）有限责任公司 自动驾驶车辆的报警***
CN108492556A (zh) * 2018-03-26 2018-09-04 京东方科技集团股份有限公司 一种交通事故规避信息的发布方法及云服务器
CN108513674A (zh) * 2018-03-26 2018-09-07 深圳市锐明技术股份有限公司 一种车前积雪与结冰的检测报警方法、存储介质和服务器
CN110033011A (zh) * 2018-12-14 2019-07-19 阿里巴巴集团控股有限公司 车祸事故处理方法和装置、电子设备

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20210312566A1 (en) * 2019-12-27 2021-10-07 Capital One Services, Llc Systems and methods for predictive model generation
CN115240408A (zh) * 2022-06-15 2022-10-25 中国第一汽车股份有限公司 一种基于区块链技术的车辆事故预警方法、***、电子设备和存储介质
CN114814888A (zh) * 2022-06-30 2022-07-29 南京源兴智达信息科技有限公司 一种基于北斗三号的手持终端设备
CN116935308A (zh) * 2023-07-10 2023-10-24 南京易自助网络科技有限公司 基于车辆情景ai智能识别的洗车机安全监控***及方法
CN116935308B (zh) * 2023-07-10 2024-04-09 南京易自助网络科技有限公司 基于车辆情景ai智能识别的洗车机安全监控***及方法
CN117745124A (zh) * 2023-11-29 2024-03-22 交通运输部规划研究院 智慧高速公路评价方法及装置

Also Published As

Publication number Publication date
CN110473418A (zh) 2019-11-19
CN110473418B (zh) 2022-05-20

Similar Documents

Publication Publication Date Title
WO2021012504A1 (zh) 基于智慧交通的道路信息提示方法、装置、服务器及介质
US11526167B1 (en) Autonomous vehicle component maintenance and repair
US11727495B1 (en) Collision risk-based engagement and disengagement of autonomous control of a vehicle
US11592308B2 (en) Data processing system communicating with a map data processing system to generate a display of one or more segments of one or more vehicle routes
US11537146B2 (en) Fleet vehicle feature activation
US11441916B1 (en) Autonomous vehicle trip routing
WO2022257201A1 (zh) 基于人机混合增强智能的城市交通安全预警方法及***
CN109617942B (zh) 产品数据推送方法、装置、计算机设备及存储介质
CN114407827B (zh) 车门控制方法、装置、设备、存储介质及自动驾驶车辆
CN112861567B (zh) 一种车型分类方法及装置
CN114677848B (zh) 感知预警***、方法、装置及计算机程序产品
CN105225490A (zh) 位于交通路口的违章电子识别平台
CN114911813B (zh) 车载感知模型的更新方法、装置、电子设备及存储介质
Huq et al. Manifesting the Policy Level Adaptation of Autonomous Vehicle Technologies in Developing Countries: The Context of Bangladesh
CN114822015A (zh) 车辆违章行为的判别方法、装置、存储介质及电子设备
CN116993156A (zh) 驾驶行为安全等级识别方法、装置及电子设备
KR20230174559A (ko) 딥러닝 기반 불법 개조 차량 탐지 장치 및 방법
CN118333768A (zh) 用于车险理赔的端云一体化***、方法及存储介质
CN117496735A (zh) 一种指示牌变化提示方法、装置、设备及存储介质

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 19938861

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 19938861

Country of ref document: EP

Kind code of ref document: A1