WO2020206900A1 - 车辆定损方法、装置、计算机设备及存储介质 - Google Patents

车辆定损方法、装置、计算机设备及存储介质 Download PDF

Info

Publication number
WO2020206900A1
WO2020206900A1 PCT/CN2019/101910 CN2019101910W WO2020206900A1 WO 2020206900 A1 WO2020206900 A1 WO 2020206900A1 CN 2019101910 W CN2019101910 W CN 2019101910W WO 2020206900 A1 WO2020206900 A1 WO 2020206900A1
Authority
WO
WIPO (PCT)
Prior art keywords
vehicle
settlement
image
value
difference threshold
Prior art date
Application number
PCT/CN2019/101910
Other languages
English (en)
French (fr)
Inventor
刁春艳
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 WO2020206900A1 publication Critical patent/WO2020206900A1/zh

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00Pattern recognition
    • G06F18/20Analysing
    • G06F18/22Matching criteria, e.g. proximity measures
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06QINFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q40/00Finance; Insurance; Tax strategies; Processing of corporate or income taxes
    • G06Q40/08Insurance
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06VIMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00Scenes; Scene-specific elements
    • G06V20/40Scenes; Scene-specific elements in video content
    • G06V20/46Extracting features or characteristics from the video content, e.g. video fingerprints, representative shots or key frames
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06VIMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00Scenes; Scene-specific elements
    • G06V20/40Scenes; Scene-specific elements in video content
    • G06V20/49Segmenting video sequences, i.e. computational techniques such as parsing or cutting the sequence, low-level clustering or determining units such as shots or scenes
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06VIMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V2201/00Indexing scheme relating to image or video recognition or understanding
    • G06V2201/08Detecting or categorising vehicles

Definitions

  • Vehicle damage assessment method Vehicle damage assessment method, device, computer equipment and storage medium
  • This application relates to the field of image detection, and in particular, to a method, device, computer equipment, and storage medium for determining vehicle damage.
  • the present application provides a vehicle damage assessment method, device, computer equipment, and storage medium, which are used to accurately collect a claim image, and quickly determine the damage based on the claim image, thereby improving the efficiency of claim settlement and enhancing user experience.
  • a method for determining the damage of a vehicle including:
  • the characteristic data includes multiple vehicle parameters of the claim vehicle, and multiple image data corresponding to each of the vehicle parameters before the claim vehicle is damaged
  • the characteristic data includes multiple vehicle parameters of the claim vehicle, and multiple image data corresponding to each of the vehicle parameters before the claim vehicle is damaged
  • [0012] grabbing an initial frame from the fixed-loss video according to each of the vehicle parameters, and associating the initial frame with the vehicle parameters;
  • a vehicle damage assessment device including:
  • the claim settlement module is configured to receive a claim settlement instruction including a unique identification of a claim settlement vehicle sent by a user, and obtain, according to the unique identification, a damage assessment video of the claim settlement vehicle shot according to preset shooting requirements;
  • a data acquisition module for acquiring characteristic data of the claim settlement vehicle includes multiple vehicle parameters of the claims settlement vehicle, and corresponding vehicle parameters before the claim settlement vehicle is damaged Multiple image data;
  • An image capture module configured to capture an initial frame from the fixed-loss video according to each of the vehicle parameters, and associate the initial frame with the vehicle parameters
  • An image comparison module configured to retrieve the image data corresponding to the vehicle parameter according to the vehicle parameter associated with the initial frame, and compare the retrieved image data with the initial frame Compare and obtain a comparison result, and use the initial frame in which the comparison result meets a preset claim settlement requirement as a claims image;
  • the value evaluation module is used to evaluate the damage value of the claim settlement vehicle according to the claim settlement image and preset claim settlement rules.
  • a computer device including a memory, a processor, and computer-readable instructions stored in the memory and capable of running on the processor, and the processor implements the following steps when the processor executes the computer-readable instructions :
  • the characteristic data includes multiple vehicle parameters of the claim vehicle, and multiple image data corresponding to each of the vehicle parameters before the claim vehicle is damaged
  • [0025] retrieve the image data corresponding to the vehicle parameter according to the vehicle parameter associated with the initial frame, compare the retrieved image data with the initial frame, and obtain a comparison result, Using the initial frame in which the comparison result meets the preset claim settlement requirement as a claim settlement image;
  • the characteristic data includes multiple vehicle parameters of the claim vehicle, and multiple image data corresponding to each of the vehicle parameters before the claim vehicle is damaged
  • [0030] retrieve the image data corresponding to the vehicle parameter according to the vehicle parameter associated with the initial frame, compare the retrieved image data with the initial frame, and obtain a comparison result, Using the initial frame in which the comparison result meets the preset claim settlement requirement as a claim settlement image;
  • FIG. 1 is a schematic diagram of an application environment of the vehicle damage assessment method in an embodiment of the present application.
  • FIG. 2 is a flowchart of a vehicle damage assessment method in an embodiment of the present application
  • FIG. 3 is a flowchart of step S30 of a vehicle damage assessment method in an embodiment of the present application.
  • FIG. 4 is a flowchart of step S40 of the vehicle damage assessment method in an embodiment of the present application.
  • FIG. 5 is a flowchart of step S50 of a vehicle damage assessment method in an embodiment of the present application.
  • FIG. 6 is a flowchart of step S50 of a vehicle damage assessment method in another embodiment of the present application.
  • FIG. 7 is a schematic diagram of a vehicle damage assessment device in an embodiment of the present application.
  • FIG. 8 is a schematic diagram of an image capture module of a vehicle damage assessment device in an embodiment of the present application.
  • FIG. 9 is a schematic diagram of an image comparison module of a vehicle damage assessment device in an embodiment of the present application.
  • FIG. 10 is a schematic diagram of a computer device in an embodiment of the present application.
  • the vehicle damage assessment method provided by the present application can be applied in the application environment as shown in FIG. 1, where a client (computer equipment/terminal device) communicates with a server (server) through a network. Perform framing processing on the fixed-loss video, and obtain a plurality of initial frames that may meet the claims requirements through preliminary comparison with vehicle parameters, and further obtain a claim image that meets the requirements from the initial frames according to the characteristic data, and Evaluate the damage value of the claim vehicle through the claim image.
  • the client includes, but is not limited to, various personal computers, laptops, smart phones, tablets, and portable wearable devices.
  • the server can be implemented as an independent server or a server cluster composed of multiple servers.
  • a method for determining vehicle damage is provided. Taking the method applied to the server in FIG. 1 as an example for description, the method includes the following steps:
  • S10 Receive a claim settlement instruction containing a unique identification of a claim settlement vehicle sent by a user, and obtain a damage assessment video of the claim settlement vehicle that is shot according to preset shooting requirements according to the unique identification.
  • the shooting requirement means that the damage fixing video must include the whole vehicle image and the partial detail image corresponding to the damaged part of the claim settlement vehicle required for settlement.
  • the claim settlement instruction may be sent to the server by the user by clicking a preset button.
  • the claim vehicle refers to a vehicle that has been insured in advance and has suffered an insured accident recorded on the insurance policy and caused value damage.
  • this application can be applied to the case of determining the damage of the claim vehicle, and can also be applied to In the case of determining the damage of other claim items other than the claim settlement vehicle, for example, it can also be applied to determining the damage of a house after a fire (insured accident), and a production machine that has been damaged due to a production accident.
  • the damage assessment video is a video recording the vehicle condition after the insurance accident of the claim settlement vehicle and the scene scene.
  • the unique identifier is a unique identifier that distinguishes different claim settlement vehicles, such as a license plate number.
  • the user can initiate a claim settlement request for settlement.
  • the claim settlement request instruction sent by the user Collect the damage assessment video of the claim settlement vehicle according to the unique identifier.
  • the loss assessment video may be shot by the user and sent to the server, or may be shot by the surveyor and sent to the server.
  • the server obtains the damage assessment video of the claim settlement vehicle for claim settlement in a subsequent step.
  • the partial detail image refers to a detail image containing the damaged part of the claim vehicle
  • the entire vehicle image refers to the entire vehicle image containing the partial detail image.
  • S20 Acquire characteristic data of the claims vehicle; the characteristic data includes multiple vehicle parameters of the claims vehicle, and image data corresponding to each of the vehicle parameters before the claims vehicle is damaged.
  • the feature data is data of various features and components of the claim settlement vehicle, and the vehicle parameters refer to the parts or accessories of the vehicle, and its position in the vehicle.
  • the front windshield It is a vehicle parameter, for example, the front left tire.
  • the image data is images of various angles of the claim settlement vehicle, or detailed images of various components, etc., and the images in the image data are all the original images of the claims settlement vehicle, that is, the image when it is not damaged;
  • Each of the image data corresponds to one of the vehicle parameters, for example, the front windshield corresponds to a front windshield image.
  • the characteristic data also includes the maintenance reference price and the replacement reference price corresponding to each of the vehicle data, so as to be used in the subsequent steps to evaluate the damage value of the claim vehicle.
  • the characteristic data may be pre-stored in the database of the server, and the characteristic data only needs to be obtained from It can be retrieved from the database of the server; it can also be retrieved from a third-party server to obtain the feature data for use in the subsequent steps to capture the frame of the loss-specific video through the feature data, and through the feature Data, assess the damage value of the claim vehicle.
  • S30 Grab an initial frame from the fixed-loss video according to each of the vehicle parameters, and associate the initial frame with the vehicle parameters.
  • the initial frame is a frame captured from the fixed-loss video according to vehicle parameters, and each initial frame represents an image.
  • the initial frame containing the vehicle parameters is captured from the damage assessment video. For example, when each of the vehicle parameters includes the vehicle body, tires, glass, etc., In the damaged video, grab the initial frame containing the body, glass, tires, etc.
  • a claim settlement requirement refers to the requirement that the image used for claim settlement set by each insurance company must meet
  • the claim settlement requirement refers to the requirement that the image used for claim settlement set by each insurance company must meet
  • the claim settlement image it must have a front body view and a side body view of the claim vehicle (Whole vehicle image), and the partial detail map of the damaged part
  • the fixed damage video taken according to the shooting requirements may not contain the images corresponding to all vehicle parameters, but only the whole vehicle with the damaged part The image and the local detail map are sufficient. Therefore, when performing the loss determination, only the image corresponding to the vehicle parameters contained in the loss determination video needs to be captured.
  • the vehicle parameters in the feature data include the Various parts (body, glass, tires, etc.), and various parts in the car (seats, steering wheel, engine), and the damage assessment video does not contain the corresponding images of each part of the car, but only contains the parts of the body surface At this time, you only need to capture the images of each part of the vehicle body surface in the fixed loss video, and the captured image is the initial frame, and then the captured initial frame is corresponding to it Are associated with the vehicle parameters for obtaining claims images in subsequent steps.
  • the claim settlement image is an image used for claim settlement
  • the claim settlement requirement refers to a requirement that must be met by each insurance company’s image used for claim settlement.
  • the claim settlement requirement is an The initial frame corresponding to the vehicle parameter (the image of the part corresponding to the vehicle parameter after the damage) is compared with the image data (the image of the part corresponding to the vehicle parameter before the damage) to determine the severity of the damage. When the degree of damage reaches the preset claim standard, it is determined that it meets the claim requirements.
  • the initial frames obtained in the above steps may be multiple, and each of the initial frames corresponds to a damaged location corresponding to one vehicle parameter of the claims vehicle, and according to the vehicle associated with the initial frame
  • the parameter calls the image data corresponding to the vehicle parameter, and the difference value between the two can be obtained by comparing each initial frame with the corresponding image data in the characteristic data, and then, detecting Whether the difference value between the two is greater than the preset first difference threshold, when the difference value between the two is greater than the first difference threshold, it means that the initial frame meets the claims requirements, and at this time, it is confirmed
  • the initial frame is a claims image, and the claim image is acquired.
  • S50 Evaluate the damage value of the claim settlement vehicle according to the claim settlement image and preset claim settlement rules.
  • the damage value refers to the damage to the claim vehicle after an insured accident
  • the damage refers to the damaged part (vehicle parameter) of the claim vehicle, and the The degree of damage at the location where the damage occurred.
  • the claim settlement rules can be preset according to requirements.
  • the damage value of each damaged part (such as low damage, medium damage, complete damage, etc.) is recorded in the claim settlement rules.
  • each of the claims settlement images records a certain damaged part of the claims settlement vehicle; in this embodiment, first, the claims settlement image is compared with the original image in the corresponding image data.
  • the damage value of the damaged part of the claim vehicle is added to the damage value of all the damaged parts of the claim vehicle to obtain the damage value of the claim vehicle.
  • the low degree of damage refers to the degree of damage that does not require replacement of the part, and only needs to be repaired to use
  • the moderate damage refers to the degree of damage to the extent that the part has been damaged to the extent that it cannot be repaired.
  • the complete damage refers to the damage of the part, which causes the entire claim vehicle to lose its original function and cannot be repaired. That is, this part cannot be repaired, and the entire claim vehicle has lost its value. Understandably, the damage value of the low-level damage is the money required to repair the part, and the damage value of the moderate damage is the money required to replace the part, and the complete damage The damage value of is the value of the claimed vehicle. It must be noted that the parts of the claim settlement vehicle recorded in each of the claims settlement images are replaced and the money spent for repair is recorded in the characteristic data of the claim settlement vehicle and the claim settlement rules.
  • the step S30 includes:
  • S301 Traverse each frame of the fixed-loss video, and detect whether each frame of the fixed-loss video matches the vehicle parameter.
  • the fixed-loss video consists of a plurality of static images, and each frame represents a static image; further, each frame of the fixed-loss video is traversed, and the image of the frame of the fixed-loss video is detected Whether it matches with the vehicle parameters is used to determine whether the image of the frame of the fixed-loss video is likely to meet the claims requirements, and when the image of the frame of the fixed-loss video matches the vehicle parameter, it represents the fixed-loss video It is possible that the image of the frame of the fixed-loss video meets the claim settlement requirements, and step S302 is entered to obtain the image of the frame of the fixed-loss video; It is impossible for the image of the frame of the fixed-loss video to meet the claim settlement requirement.
  • the claim settlement requirement is that the claim settlement image must have a front body view of the car.
  • the vehicle parameter in the frame of the fixed-loss video is the front body of the car or whether it is the side body of the car, it can be determined whether the image of the frame of the fixed-loss video matches the vehicle parameter.
  • S302 When the frame of the loss-fixed video matches the vehicle parameter, obtain all the frames corresponding to the same vehicle parameter, and use a preset image in all the obtained frames Request to confirm the car The initial frame corresponding to the vehicle parameter.
  • the image representing the frame of the fixed-loss video may meet the claims requirements.
  • the frame of the fixed-loss video is obtained, which is understandable
  • only one initial frame of the seat needs to be selected, and the confirmation of the initial frame can be confirmed according to the preset image requirements.
  • the shooting requirement in step S10 means that the damage determination video must include the entire vehicle image and the partial detail image corresponding to the damaged part of the claim vehicle required for the claim;
  • the step of confirming the initial frame corresponding to the vehicle parameter in all the acquired frames according to a preset image requirement includes:
  • [0075] Acquire whole vehicle images corresponding to the same vehicle parameter in all the frames in the fixed-loss video, and local detail images matching the whole vehicle image of the vehicle parameter; that is, the fixed-loss video
  • the captured frames in include the whole vehicle image and the partial detail image corresponding to the same damaged part (one damaged part corresponds to a vehicle parameter).
  • the whole vehicle image can be used to match the vehicle parameters to determine the damaged part, and when determining After the damaged part, the damage is determined according to the partial detail map of the damaged part (the partial detail image matching the whole vehicle image of the vehicle parameters), and the partial detail image of the damaged part used to judge the damage can be marked as the initial frame .
  • the captured frame in the fixed-loss video contains the entire vehicle image corresponding to the vehicle parameter, but does not contain the local detail map matching the entire vehicle image, it can be confirmed that the part corresponding to the vehicle parameter is not damaged. ; Only when the entire vehicle image corresponding to the same vehicle parameter has a matching partial detail image, the part corresponding to the vehicle parameter may be considered as a damaged part (or it may still not be a damaged part, so it is necessary to perform step S40 Further screening), at this time, the initial frames in the multiple partial detail images can be confirmed according to the preset image requirements.
  • the local detail image with the highest pixel or/and the highest definition in the local detail image matched with the entire vehicle image of the vehicle parameter is taken as the corresponding frame and recorded as the initial frame. That is, in this embodiment, the frame corresponding to the local detail image with the highest pixel or/and the highest definition can be selected and set as the initial frame that can be used to determine the damage of the damaged part.
  • the step S40 includes:
  • S401 Obtain a difference value between the retrieved image data and the initial frame, and detect whether the difference value is greater than a preset first difference threshold.
  • the first difference threshold may be preset according to requirements, and the first difference threshold is used to compare with the difference between the initial frame and the corresponding image data, and then determine Whether the part of the claim settlement vehicle corresponding to the vehicle parameter in the initial frame is damaged, that is, when the difference value between the initial frame and the corresponding image data is greater than the first difference threshold, it represents the If the part of the claim vehicle in the initial frame is damaged, at this time, confirm that the initial frame is a claim image; when the difference value between the initial frame and the corresponding image data is less than or equal to the first difference threshold, It means that the initial frame has not changed significantly compared with the corresponding image data, that is, the parts of the claim settlement vehicle recorded in the initial frame are not damaged.
  • S402 When the difference value is greater than the first difference threshold, confirm that the initial frame is a claims image.
  • the difference value between the initial frame and the corresponding image data is greater than the first difference threshold, it represents that the initial frame is compared with the corresponding image data,
  • the vehicle parameters that is, a certain part of the claim vehicle or a certain accessory
  • the initial frame is the claim settlement image, and the claim settlement image is acquired for use in the subsequent steps to evaluate the damage value of the claim settlement vehicle through the claims settlement image.
  • S403 when the difference value is less than or equal to the first difference threshold, confirm whether the difference value is greater than a preset second difference threshold, and when the difference value is greater than the second difference threshold , Sending the initial frame and the difference threshold to a preset reviewer; the second difference threshold is less than the first difference threshold.
  • the second difference threshold that is less than the first difference threshold is set, that is, When the server determines that the difference value corresponding to the damage of the part of the vehicle parameter corresponding to the initial frame is between the first difference threshold and the second difference threshold, it indicates that the damage is between the need for compensation and the need for compensation, and the difference is It is impossible to be automatically authenticated by the server. In this case, a preset reviewer can make a manual judgment to determine whether it needs to make a claim.
  • S404 When the difference value is less than or equal to the second difference threshold, mark the position of the vehicle parameter corresponding to the initial frame in the claims vehicle as a non-destructive position.
  • the difference value corresponding to the damage condition of the part of the vehicle parameter corresponding to the initial frame is less than or equal to the second difference threshold, it is indicated that it can be directly determined that no compensation is required, and the part is set as a non-damaged part.
  • the step S50 includes:
  • S501 Detect whether the difference value corresponding to the claims image is greater than the third difference threshold, where the third difference threshold is greater than the first difference threshold.
  • the damage value is the monetary value corresponding to the damage after a part of the claim settlement image is damaged; for example, the claim settlement rule is as follows: When the claim settlement image and the corresponding image data When the difference value between is greater than the third difference threshold, the damage value of the part of the claim vehicle in the image is the replacement value; otherwise, the damage value of the part of the claim vehicle in the image is the repair value.
  • the replacement value is the money required to replace the part recorded in the claims image; the repair value is the money required to repair the part recorded in the claims image.
  • the third difference threshold is a threshold used to determine whether the part of the claim settlement vehicle recorded in the claim settlement image is moderately damaged or lowly damaged, when the claim settlement image and the image When the difference between the corresponding original images in the data is greater than the preset third difference threshold, it means that the part of the claim settlement vehicle recorded in the claim settlement image is moderately damaged; when the claim settlement image and the image data When the difference value between the corresponding original images in is less than or equal to the third difference threshold, it means that the part of the claim settlement vehicle recorded in the claim settlement image belongs to low damage.
  • each of the claims image and its corresponding image data will be compared to obtain a difference value, and the third difference threshold for comparison with each difference value is not necessarily Similarly, according to different claims settlement images, that is, according to different claims settlement vehicles, different third difference thresholds can be set.
  • S502 When the difference value corresponding to the claim settlement image is greater than the third threshold, confirm that the damage value of the damaged part of the claim settlement vehicle corresponding to the claim settlement image is corresponding to the Damage The replacement value of the damaged part.
  • the difference value between the claim image and the corresponding original image in the image data is greater than the second threshold, it means that the damage degree of the part recorded in the claim image is moderate damage. That is, this part cannot be repaired and needs to be replaced so that the claim settlement vehicle can achieve its original function, so the damage value of this part is evaluated as the replacement value.
  • S504 The sum of the confirmed damage values of all damaged parts of the claim settlement vehicle is confirmed as the damage value of the claim settlement vehicle.
  • the method further includes:
  • S505 When the difference value corresponding to the claims image is greater than the third difference threshold, detect whether the damaged part corresponding to the claims image is an irreplaceable part.
  • the non-replaceable part refers to a part that can only be repaired but cannot be replaced when the part is damaged. If the damage of the part reaches the level that it cannot be replaced, it can be considered that the claim vehicle is The degree of damage has reached complete damage, that is, the claim settlement vehicle has been unable to restore its original function by repairing or replacing parts (accessories).
  • S506 When the corresponding damaged part in the claim settlement image is a non-replaceable part, confirm that the damage value of the claim settlement vehicle is the value of the claim settlement vehicle.
  • This application obtains multiple initial frames that are likely to meet the claims requirements by performing framing processing on the fixed-loss video and preliminary comparison with vehicle parameters. Further, according to the characteristic data, the initial frames are obtained from the initial frames.
  • the claim settlement image is required, and the damage value of the claim settlement vehicle is evaluated through the claim settlement image, which can improve the accuracy of collecting the claim settlement image, and at the same time improve the efficiency of loss determination, thereby improving the efficiency of claim settlement and enhancing the user experience.
  • a vehicle damage measurement device is provided, and the vehicle damage measurement device corresponds to the vehicle damage measurement method in the foregoing embodiment one-to-one.
  • the vehicle loss assessment device includes a claim settlement module 11, a data acquisition module 12, an image capture module 13, an image comparison module 14, and a value evaluation module 15.
  • the detailed description of each functional module is as follows:
  • the claim settlement module 11 is configured to receive a claim settlement instruction containing a unique identification of a claim settlement vehicle sent by a user, and obtain, according to the unique identification, a damage assessment video of the claim settlement vehicle shot according to preset shooting requirements
  • the data acquisition module 12 is configured to acquire characteristic data of the claims vehicle; the characteristic data includes a plurality of vehicle parameters of the claims vehicle, and the vehicle parameters before the claim vehicle is damaged Multiple corresponding image data;
  • the image capture module 13 is configured to capture an initial frame from the fixed-loss video according to each of the vehicle parameters, and associate the initial frame with the vehicle parameters;
  • the image comparison module 14 is configured to retrieve the image data corresponding to the vehicle parameter according to the vehicle parameter associated with the initial frame, and compare the retrieved image data with the initial Frame comparison and Acquiring a comparison result, and using the initial frame in which the comparison result meets a preset claim settlement requirement as a claims image;
  • the value evaluation module 15 is configured to evaluate the damage value of the claim settlement vehicle according to the claim settlement image and preset claim settlement rules.
  • the image capture module 13 includes:
  • the image detection unit 131 is configured to traverse each frame of the fixed-loss video and detect whether each frame of the fixed-loss video matches the vehicle parameter;
  • the image acquiring unit 132 is configured to acquire all the frames corresponding to the same vehicle parameter when the frame of the fixed-loss video matches the vehicle parameter, and among all the acquired frames Confirm the initial frame corresponding to the vehicle parameter according to the preset image requirements.
  • the image comparison module 14 includes:
  • the first detection unit 141 is configured to obtain a difference value between the retrieved image data and the initial frame, and detect whether the difference value is greater than a preset first difference threshold;
  • the confirming unit 142 is configured to confirm that the initial frame is a claims image when the difference value is greater than the first difference threshold;
  • the second detection unit 143 is configured to, when the difference value is less than or equal to the first difference threshold, confirm whether the difference value is greater than a second difference threshold, and when the difference value is greater than the second difference threshold, When the difference threshold is different, send the initial frame and the difference threshold to a preset reviewer; the second difference threshold is less than the first difference threshold;
  • the marking unit 144 is configured to mark the position of the vehicle parameter corresponding to the initial frame in the claims vehicle as a non-destructive position when the difference value is less than or equal to the second difference threshold.
  • each of the claims settlement images records a damaged part of the claims settlement vehicle
  • the claims settlement rule includes at least one correspondence between a third difference threshold and a damage value
  • the value assessment module is also used to:
  • the value evaluation module is further used to:
  • the vehicle damage assessment device For the specific definition of the vehicle damage assessment device, please refer to the above definition of the vehicle damage assessment method, which will not be repeated here.
  • the various modules in the vehicle damage assessment device described above can be implemented in whole or in part by software, hardware, and combinations thereof.
  • the above-mentioned modules may be embedded in the form of hardware or independent of the processor in the computer equipment, or may be stored in the memory of the computer equipment in the form of software, so that the processor can call and execute the corresponding operations of the above-mentioned modules.
  • a computer device is provided.
  • the computer device may be a server, and an internal structure diagram thereof may be as shown in FIG. 10.
  • the computer equipment includes a processor, a memory, a network interface, and a database connected through a system bus. Among them, the processor of the computer device is used to provide calculation and control capabilities.
  • the memory of the computer device includes a non-volatile storage medium and an internal memory.
  • the non-volatile storage medium stores an operating system, computer readable instructions, and a database.
  • the internal memory provides an environment for the operation of the operating system and computer-readable instructions in the non-volatile storage medium.
  • the network interface of the computer equipment is used to communicate with an external terminal through a network connection.
  • the computer-readable instructions are executed by the processor to realize a vehicle damage method.
  • a computer device including a memory, a processor, and computer readable instructions stored in the memory and running on the processor, and the processor executes the following steps when the computer readable instructions are executed :
  • the characteristic data includes multiple vehicle parameters of the claim settlement vehicle, and multiple image data corresponding to each of the vehicle parameters before the claim settlement vehicle is damaged
  • one or more non-volatile readable storage media storing computer readable instructions are provided, and the non-volatile readable storage medium stores computer readable instructions, and the computer When the readable instructions are executed by one or more processors, the one or more processors implement the following steps:
  • the characteristic data includes multiple vehicle parameters of the claims vehicle, and multiple image data corresponding to each of the vehicle parameters before the claim vehicle is damaged
  • [0137] retrieve the image data corresponding to the vehicle parameter according to the vehicle parameter associated with the initial frame, compare the retrieved image data with the initial frame, and obtain a comparison result, Using the initial frame in which the comparison result meets the preset claim settlement requirement as a claim settlement image;
  • Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory.
  • Volatile memory may include random access memory (RAM) or external cache memory.
  • RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous chain Channel (Synchlink) DRAM (SLDRAM), memory bus (Rambus), direct RAM (RDRAM), direct memory bus dynamic RAM (DRDR AM), and memory bus dynamic RAM (RDRAM), etc.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Business, Economics & Management (AREA)
  • Data Mining & Analysis (AREA)
  • Finance (AREA)
  • Accounting & Taxation (AREA)
  • Multimedia (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Development Economics (AREA)
  • Evolutionary Computation (AREA)
  • Evolutionary Biology (AREA)
  • Computing Systems (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Artificial Intelligence (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • General Engineering & Computer Science (AREA)
  • Economics (AREA)
  • Marketing (AREA)
  • Strategic Management (AREA)
  • Technology Law (AREA)
  • General Business, Economics & Management (AREA)
  • Financial Or Insurance-Related Operations Such As Payment And Settlement (AREA)
  • Image Processing (AREA)

Abstract

一种车辆定损方法、装置、计算机设备及存储介质,所述方法包括:接收用户发送的包含理赔车辆唯一标识的理赔指令,根据唯一标识获取按照预设的拍摄要求拍摄的理赔车辆的定损视频(S10);获取理赔车辆的特征数据(S20);自定损视频中根据各车辆参数抓取初始帧,并将初始帧与车辆参数关联(S30);根据与初始帧关联的车辆参数调取与该车辆参数对应的图像数据,将调取的图像数据与初始帧进行比对并获取比对结果,将比对结果符合理赔要求的初始帧作为理赔图像(S40);根据理赔图像以及预设的理赔规则评估理赔车辆的损毁价值(S50)。该方法可以通过准确的采集理赔图像,并根据理赔图像进行快速定损,进而提高理赔效率,增强用户体验。

Description

车辆定损方法、 装置、 计算机设备及存储介质
[0001] 本申请以 2019年 4月 9日提交的申请号为 201910280404.3, 名称为“车辆定损方法
、 装置、 计算机设备及存储介质”的中国发明专利申请为基础, 并要求其优先权 技术领域
[0002] 本申请涉及图像检测领域, 尤其涉及一种车辆定损方法、 装置、 计算机设备及 存储介质。
[0003]
[0004] 背景技术
[0005] 现今的车险勘验过程中, 是通过拍摄现场场景的图片, 再让查勘人员根据所述 图片来判断物体受损情况, 再由理赔人员根据受损情况对客户进行理赔报价, 而对于非专业的查勘人员来说, 仅根据现场拍摄的照片将会很难符合判断受损 的具体情况, 进而导致理赔人员无法根据查勘人员确定的受损情况进行理赔判 定, 因此需要查勘人员或理赔人员再次联系客户, 进行车辆受损图片的二次采 集, 使理赔的时间延长, 降低理赔效率。
[0006]
[0007] 发明内容
[0008] 基于此, 本申请提供一种车辆定损方法、 装置、 计算机设备及存储介质, 用于 准确的采集理赔图像, 并根据理赔图像进行快速定损, 进而提高理赔效率, 增 强用户体验。
[0009] 一种车辆定损方法, 包括:
[0010] 接收用户发送的包含理赔车辆唯一标识的理赔指令, 根据所述唯一标识获取按 照预设的拍摄要求拍摄的所述理赔车辆的定损视频;
[0011] 获取所述理赔车辆的特征数据; 所述特征数据包含所述理赔车辆的多个车辆参 数, 以及所述理赔车辆未损毁之前与各所述车辆参数分别对应的多个图像数据 [0012] 自所述定损视频中根据各所述车辆参数抓取初始帧, 并将所述初始帧与所述车 辆参数关联;
[0013] 根据与所述初始帧关联的所述车辆参数调取与该车辆参数对应的所述图像数据 , 将调取的所述图像数据与所述初始帧进行比对并获取比对结果, 将所述比对 结果符合预设的理赔要求的所述初始帧作为理赔图像;
[0014] 根据所述理赔图像以及预设的理赔规则评估所述理赔车辆的损毁价值。
[0015] 一种车辆定损装置, 包括:
[0016] 理赔接收模块, 用于接收用户发送的包含理赔车辆唯一标识的理赔指令, 根据 所述唯一标识获取按照预设的拍摄要求拍摄的所述理赔车辆的定损视频;
[0017] 数据获取模块, 用于获取所述理赔车辆的特征数据; 所述特征数据包含所述理 赔车辆的多个车辆参数, 以及所述理赔车辆未损毁之前与各所述车辆参数分别 对应的多个图像数据;
[0018] 图像抓取模块, 用于自所述定损视频中根据各所述车辆参数抓取初始帧, 并将 所述初始帧与所述车辆参数关联;
[0019] 图像比对模块, 用于根据与所述初始帧关联的所述车辆参数调取与该车辆参数 对应的所述图像数据, 将调取的所述图像数据与所述初始帧进行比对并获取比 对结果, 将所述比对结果符合预设的理赔要求的所述初始帧作为理赔图像;
[0020] 价值评估模块, 用于根据所述理赔图像以及预设的理赔规则评估所述理赔车辆 的损毁价值。
[0021] 一种计算机设备, 包括存储器、 处理器以及存储在所述存储器中并可在所述处 理器上运行的计算机可读指令, 所述处理器执行所述计算机可读指令时实现如 下步骤:
[0022] 接收用户发送的包含理赔车辆唯一标识的理赔指令, 根据所述唯一标识获取按 照预设的拍摄要求拍摄的所述理赔车辆的定损视频;
[0023] 获取所述理赔车辆的特征数据; 所述特征数据包含所述理赔车辆的多个车辆参 数, 以及所述理赔车辆未损毁之前与各所述车辆参数分别对应的多个图像数据
[0024] 自所述定损视频中根据各所述车辆参数抓取初始帧, 并将所述初始帧与所述车 辆参数关联;
[0025] 根据与所述初始帧关联的所述车辆参数调取与该车辆参数对应的所述图像数据 , 将调取的所述图像数据与所述初始帧进行比对并获取比对结果, 将所述比对 结果符合预设的理赔要求的所述初始帧作为理赔图像;
[0026] 根据所述理赔图像以及预设的理赔规则评估所述理赔车辆的损毁价值。 一个或 多个存储有计算机可读指令的非易失性可读存储介质, 其特征在于, 所述计算 机可读指令被一个或多个处理器执行时, 使得所述一个或多个处理器执行如下 步骤:
[0027] 接收用户发送的包含理赔车辆唯一标识的理赔指令, 根据所述唯一标识获取按 照预设的拍摄要求拍摄的所述理赔车辆的定损视频;
[0028] 获取所述理赔车辆的特征数据; 所述特征数据包含所述理赔车辆的多个车辆参 数, 以及所述理赔车辆未损毁之前与各所述车辆参数分别对应的多个图像数据
[0029] 自所述定损视频中根据各所述车辆参数抓取初始帧, 并将所述初始帧与所述车 辆参数关联;
[0030] 根据与所述初始帧关联的所述车辆参数调取与该车辆参数对应的所述图像数据 , 将调取的所述图像数据与所述初始帧进行比对并获取比对结果, 将所述比对 结果符合预设的理赔要求的所述初始帧作为理赔图像;
[0031] 根据所述理赔图像以及预设的理赔规则评估所述理赔车辆的损毁价值。
[0032] 本申请的一个或多个实施例的细节在下面的附图和描述中提出, 本申请的其他 特征和优点将从说明书、 附图以及权利要求变得明显。
[0033]
[0034]
[0035] 附图说明
[0036] 为了更清楚地说明本申请实施例的技术方案, 下面将对本申请实施例的描述中 所需要使用的附图作简单地介绍, 显而易见地, 下面描述中的附图仅仅是本申 请的一些实施例, 对于本领域普通技术人员来讲, 在不付出创造性劳动的前提 下, 还可以根据这些附图获得其他的附图。 [0037] 图 1是本申请一实施例中车辆定损方法的一应用环境不意图;
[0038] 图 2是本申请一实施例中车辆定损方法的流程图;
[0039] 图 3是本申请一实施例中车辆定损方法的步骤 S30的流程图;
[0040] 图 4是本申请一实施例中车辆定损方法的步骤 S 40的流程图;
[0041] 图 5是本申请一实施例中车辆定损方法的步骤 S50的流程图;
[0042] 图 6是本申请另一实施例中车辆定损方法的步骤 S 50的流程图;
[0043] 图 7是本申请一实施例中车辆定损装置的示意图;
[0044] 图 8是本申请一实施例中车辆定损装置的图像抓取模块的示意图;
[0045] 图 9是本申请一实施例中车辆定损装置的图像比对模块的示意图;
[0046] 图 10是本申请一实施例中计算机设备的一示意图。
[0047]
[0048] 具体实施方式
[0049] 下面将结合本申请实施例中的附图, 对本申请实施例中的技术方案进行清楚、 完整地描述, 显然, 所描述的实施例是本申请一部分实施例, 而不是全部的实 施例。 基于本申请中的实施例, 本领域普通技术人员在没有作出创造性劳动前 提下所获得的所有其他实施例, 都属于本申请保护的范围。
[0050] 本申请提供的车辆定损方法, 可应用在如图 1的应用环境中, 其中, 客户端 ( 计算机设备 /终端设备) 通过网络与服务器 (服务端) 进行通信。 对定损视频进 行分帧处理, 通过与车辆参数进行初步比对, 获取多个有可能符合理赔要求的 初始帧, 进一步地, 根据所述特征数据从初始帧中获取符合要求的理赔图像, 并通过理赔图像评估出理赔车辆的损毁价值。 其中, 客户端 (计算机设备 /终端 设备) 包括但不限于各种个人计算机、 笔记本电脑、 智能手机、 平板电脑和便 携式可穿戴设备。 服务器可以用独立的服务器或者是多个服务器组成的服务器 集群来实现。
[0051] 在一实施例中, 如图 2所示, 提供一种车辆定损方法, 以该方法应用在图 1中的 服务器为例进行说明, 包括如下步骤:
[0052] S10, 接收用户发送的包含理赔车辆唯一标识的理赔指令, 根据所述唯一标识 获取按照预设的拍摄要求拍摄的所述理赔车辆的定损视频。 [0053] 其中, 所述拍摄要求是指所述定损视频中必须包含理赔所需的所述理赔车辆的 损毁部位对应的整车图像和局部细节图像。 所述理赔指令可以由用户通过点击 预设的按钮发送至服务器。 所述理赔车辆是指预先进行投保, 并发生了保单上 记载的保险事故且造成了价值损毁的车辆, 可理解地, 本申请可以应用于对理 赔车辆进行定损的情况, 还可以应用于对除所述理赔车辆之外的其他理赔物进 行定损的情况, 比如, 还可以应用于对发生火灾 (保险事故) 后的房屋, 发生 生产事故并损毁的生产机器等进行定损。 所述定损视频为记录所述理赔车辆发 生保险事故后的车辆状况, 以及现场场景的视频。 所述唯一标识为区分不同理 赔车辆的具有唯一性的标识, 如车牌号等。
[0054] 可理解地, 当所述理赔车辆发生保险事故并且造成了价值损毁时, 用户即可通 过发起理赔请求进行理赔, 在本实施例中, 当接收到用户发送的理赔请求指令 后, 即根据所述唯一标识采集所述理赔车辆的定损视频, 具体地, 所述定损视 频, 可以由所述用户自行拍摄后发送至服务器, 亦可由查勘员拍摄后发送至所 述服务器, 接着, 所述服务器获取所述理赔车辆的定损视频, 以供在后续步骤 中进行理赔。 可理解地, 所述局部细节图是指包含理赔车辆的损毁部位的细节 图, 而所述整车图像是指包含所述局部细节图像的整车图像。
[0055] S20, 获取所述理赔车辆的特征数据; 所述特征数据包含所述理赔车辆的多个 车辆参数, 以及在所述理赔车辆未损毁之前与各所述车辆参数分别对应的图像 数据。
[0056] 其中, 所述特征数据为所述理赔车辆各种特征和组成物的数据, 所述车辆参数 是指车辆的部位或配件, 以及其在车辆中的位置, 举例说明, 前挡风玻璃即为 一车辆参数, 又比如, 前左轮胎。 所述图像数据为所述理赔车辆各角度的图像 、 或各组成物的细节图像等, 所述图像数据中的图像均为所述理赔车辆的原始 图像, 即未损毁时的图像; 可理解地, 每一所述图像数据对应于一个所述车辆 参数, 比如, 前挡风玻璃对应于前挡风玻璃图像。 必须说的是所述特征数据还 包括每一所述车辆数据所对应的维修参考价格、 更换参考价格, 以供在后续步 骤中用于评估所述理赔车辆的损毁价值。
[0057] 所述特征数据可以预先存储在服务器的数据库中, 获取所述特征数据时只需从 所述服务器的数据库中调取; 亦可以从第三方服务器中获取, 获取所述特征数 据, 以供在后续步骤中通过所述特征数据对所述定损视频进行抓帧, 以及通过 所述特征数据, 评估所述理赔车辆的损毁价值。
[0058] S30, 自所述定损视频中根据各所述车辆参数抓取初始帧, 并将所述初始帧与 所述车辆参数关联。
[0059] 其中, 所述初始帧为自所述定损视频中根据车辆参数抓取出来的帧, 每一所述 初始帧即代表一幅图像。
[0060] 根据各所述车辆参数, 从所述定损视频中抓取包含有所述车辆参数的初始帧, 比如, 当各所述车辆参数包括车身、 轮胎、 玻璃等, 则从所述定损视频中抓取 出包含有车身、 玻璃、 轮胎等的初始帧。
[0061] 在一实施例中, 当理赔要求 (所述理赔要求是指各保险公司所订立的用于理赔 的图像必须满足的要求) 为理赔图像须有理赔车辆的正面车身图, 侧面车身图 (整车图像) , 以及损毁部位的局部细节图时, 可理解地, 按照所述拍摄要求 拍摄的定损视频可能不会包含所有车辆参数所对应的图像, 而是仅包含损毁部 位的整车图像和局部细节图即可, 因此, 在进行定损时, 只须抓取出所述定损 视频中包含的车辆参数对应的图像即可, 比如说, 特征数据中的车辆参数包括 车身表面的各部位 (车身、 玻璃、 轮胎等) , 以及车内的各部位 (座椅、 方向 盘、 发动机) , 而所述定损视频中没有包含车内各部位对应的图像, 而只有包 含车身表面各部位的图像, 此时只需抓取所述定损视频中包含车身表面各部位 的图像即可, 所抓取出的图像即为所述初始帧, 接着, 将抓取的所述初始帧与 其对应的车辆参数相关联, 以供在后续步骤中获取理赔图像。
[0062] S40, 根据与所述初始帧关联的所述车辆参数调取与该车辆参数对应的所述图 像数据, 将调取的所述图像数据与所述初始帧进行比对并获取比对结果, 将所 述比对结果符合预设的理赔要求的所述初始帧作为理赔图像。
[0063] 其中, 所述理赔图像为用于理赔的图像, 所述理赔要求是指各保险公司所订立 的用于理赔的图像必须满足的要求, 举例来说, 所述理赔要求为通过将与该车 辆参数对应的初始帧 (损毁之后该车辆参数对应的部位的图像) 与图像数据 ( 损毁之前该车辆参数对应的部位的图像) 进行比对, 确定损毁情况的轻重, 在 损毁程度达到预设的理赔标准时, 确定其符合理赔要求。
[0064] 上述步骤中获取的所述初始帧可以为多个, 每一个所述初始帧对应于一个所述 理赔车辆的一个车辆参数对应的损毁部位, 根据与所述初始帧关联的所述车辆 参数调取与该车辆参数对应的所述图像数据, 可以通过将每一所述初始帧与对 应的所述特征数据中的图像数据进行比对, 获取二者之间的差异值, 接着, 检 测二者之间的差异值是否大于预设的第一差异阈值, 当二者之间的差异值大于 所述第一差异阈值时, 代表所述初始帧符合所述理赔要求, 此时, 即确认所述 初始帧为理赔图像, 并获取所述理赔图像, 通过上述步骤, 将所述初始帧中符 合所述理赔要求的所有帧选取出来, 并确认它们为所述理赔图像, 以供在后续 步骤中通过所述理赔图像评估所述理赔车辆的损毁价值。
[0065] S50, 根据所述理赔图像以及预设的理赔规则评估所述理赔车辆的损毁价值。
[0066] 其中, 所述损毁价值是指发生保险事故后导致理赔车辆损毁, 该损毁所对应的 金钱价值, 所述损毁是指所述理赔车辆的出现损毁的部位 (车辆参数) , 以及 所述出现损毁的部位的损毁程度。 所述理赔规则可以根据需求预先设定。 理赔 规则中记录了每一个损毁部位的损毁程度 (比如: 低度损毁、 中度损毁、 完全 损毁等) 所对应的损毁价值。 比如, 每一所述理赔图像中, 均记载着所述理赔 车辆的某一损毁部位; 在本实施例中, 首先将所述理赔图像与其对应的所述图 像数据中的原始图像进行比对, 根据二者之间的差异程度, 评估所述理赔图像 中的理赔车辆的损毁部位的损毁程度是低度损毁, 中度损毁还是完全损毁, 从 而自理赔规则中获取所述理赔图像中的理赔车辆的损毁部位的损毁价值, 将所 述理赔车辆所有损毁部位的损毁价值相加, 进而得到所述理赔车辆的损毁价值 。 其中, 所述低度损毁是指损毁的程度不需要更换该部位, 只需进行修复即可 使用的程度; 所述中度损毁是指损毁的程度为该部位已经损毁到无法进行修复 的程度, 必须要对该部位进行更换, 才能使所述理赔车辆回复到原来的状态, 拥有原来的功能; 所述完全损毁是指该部位损毁, 导致整个理赔车辆失去了原 有功能, 且无法修复, 也即该部位无法进行修复, 并且使整个理赔车辆都失去 了价值。 可理解地, 所述低度损毁的损毁价值即为修复该部位的所需耗费的金 钱, 所述中度损毁的损毁价值即为更换该部位所需耗费的金钱, 所述完全损毁 的损毁价值为所述理赔车辆的价值。 必须说明的是, 每一所述理赔图像中所记 载的所述理赔车辆的部位进行更换, 修复所需耗费的金钱, 均在所述理赔车辆 的特征数据以及所述理赔规则中有记载。
[0067] 本实施例通过对定损视频进行分帧处理, 通过与车辆参数进行初步比对, 获取 多个有可能符合理赔要求的初始帧, 进一步地, 根据所述特征数据从初始帧中 获取符合要求的理赔图像, 并通过理赔图像评估出理赔车辆的损毁价值, 可以 提升采集理赔图像的准确率, 同时提高定损效率, 进而提高理赔效率, 增强用 户体验。
[0068] 在一实施例中, 如图 3所示, 所述步骤 S30包括:
[0069] S301 , 遍历所述定损视频的每一帧, 检测所述定损视频的各所述帧与所述车辆 参数是否匹配。
[0070] 首先, 对所述定损视频进行分帧, 比如设定帧长和帧宽, 通过分帧函数对所述 定损视频进行分帧处理; 进而获取所述定损视频的帧, 可理解地, 所述定损视 频由多个静态图像组成, 每一帧均代表一幅静态图像; 进一步地, 遍历所述定 损视频的每一帧, 通过检测所述定损视频的帧的图像与所述车辆参数是否匹配 , 来判定所述定损视频的帧的图像是否有可能满足理赔要求, 当所述定损视频 的帧的图像与所述车辆参数匹配时, 代表所述定损视频的帧的图像有可能满足 所述理赔要求, 此时进入步骤 S302, 获取所述定损视频的帧的图像; 当所述定 损视频的帧的图像与所述车辆参数不匹配时, 代表所述定损视频的帧的图像不 可能满足所述理赔要求, 此时, 检测下一所述定损视频的帧的图像是否与所述 车辆参数相匹配, 直至所有所述定损视频的帧均检测完毕, 也即组成所述定损 视频的每一幅静态图像均已被检测。
[0071] 检测所述定损视频的帧的图像与所述车辆参数是否匹配, 举例说明, 所述理赔 要求为理赔图像须有车的正面车身图, 该车的侧面车身图时, 即检测所述定损 视频的帧中的车辆参数是否为车的正面车身, 或者是否为车的侧面车身, 即可 判定所述定损视频的帧的图像与所述车辆参数是否匹配。
[0072] S302, 当所述定损视频的帧与所述车辆参数匹配时, 获取对应于同一个所述车 辆参数的所有所述帧, 并在获取的所有所述帧中根据预设的图像要求确认该车 辆参数对应的初始帧。
[0073] 当所述定损视频的帧与所述车辆参数匹配时, 代表所述定损是的帧的图像可能 符合所述理赔要求, 此时, 获取所述定损视频的帧, 可理解地, 获取的对应于 同一个所述车辆参数的帧可能有多个, 此时, 仅需选取其中的一个座位初始帧 即可, 该初始帧的确认可以根据预设的图像要求进行确认。
[0074] 在一实施例中, 步骤 S10中的所述拍摄要求是指所述定损视频中必须包含理赔 所需的所述理赔车辆的损毁部位对应的整车图像和局部细节图像; 此时, 所述 在获取的所有所述帧中根据预设的图像要求确认该车辆参数对应的初始帧, 包 括:
[0075] 获取所述定损视频中的所有所述帧中对应于同一个所述车辆参数的整车图像, 以及与该车辆参数的整车图像匹配的局部细节图像; 也即, 定损视频中抓取的 帧包括对应于同一个损毁部位 (一个损毁部位对应于一个车辆参数) 的整车图 像和局部细节图像, 整车图像可以用于和车辆参数进行匹配以确定损毁部位, 而在确定损毁部位之后, 根据该损毁部位的局部细节图 (与该车辆参数的整车 图像匹配的局部细节图) 确定损毁情况, 该损毁部位用于评判损毁情况的局部 细节图即可被标记为初始帧。 若该定损视频中抓取的帧中包含对应于所述车辆 参数的整车图像, 但是不包含与该整车图像匹配的局部细节图时, 可以确认该 车辆参数对应的部位并未出现损毁; 仅有在对应于同一个所述车辆参数的整车 图像具有与其匹配的局部细节图像时, 才认为该车辆参数对应的部位可能为损 毁部位 (也有可能依旧不是损毁部位, 因此需要在步骤 S40中进一步筛选) , 此 时可以根据预设的图像要求确认多个所述局部细节图像中的初始帧。
[0076] 将与该车辆参数的整车图像匹配的所述局部细节图像中像素最高或 /和清晰度 最高的所述局部细节图像作为对应的所述帧, 记录为初始帧。 也即, 在该实施 例中, 可以选取对像素最高或 /和清晰度最高的所述局部细节图像对应的帧, 将 其设定为可以用于定损该损毁部位的损毁情况的初始帧。
[0077] 本实施例通过比对所述定损视频的帧的图像和所述车辆参数是否匹配, 可以更 精确的抓取出所述定损视频中, 有可能符合所述理赔要求的帧, 以供在后续步 骤中获取符合理赔要求的理赔图像。 [0078] 在一实施例中, 如图 4所示, 所述步骤 S40包括:
[0079] S401, 获取调取的所述图像数据与所述初始帧之间的差异值, 并检测所述差异 值是否大于预设的第一差异阈值。
[0080] 其中, 所述第一差异阈值可以根据需求预先设定, 所述第一差异阈值是用于与 所述初始帧和与其对应的所述图像数据之间的差异值进行比较, 进而判定所述 初始帧中的车辆参数所对应的理赔车辆的部位是否出现损毁, 即当所述初始帧 与其对应的所述图像数据之间的差异值大于所述第一差异阈值时, 即代表所述 初始帧中的理赔车辆的部位出现损毁, 此时, 确认该初始帧为理赔图像; 当该 初始帧和与其对应的所述图像数据之间的差异值小于或等于所述第一差异阈值 时, 代表该初始帧和与其对应的所述图像数据相比, 并未发生明显的变化, 也 即该初始帧中记载的所述理赔车辆的部位并未出现损毁, 此时, 选取下一未被 选取进行检测的初始帧, 并进行上述检测, 后续过程如上述, 直至所有所述初 始帧均被检测完毕。 必须说明的是, 在上述检测过程中, 每一所述初始帧和与 其对应的所述图像数据进行比对后均会获得不同的差异值, 与各差异值进行比 较的第一差异阈值并非一定相同, 根据不同的初始帧也即根据不同的理赔车辆 的部位, 可以设定不同的第一差异阈值。
[0081] S402, 当所述差异值大于所述第一差异阈值时, 确认所述初始帧为理赔图像。
[0082] 当所述初始帧与其对应的所述图像数据之间的差异值大于所述第一差异阈值时 , 代表所述初始帧与其对应的所述图像数据相比, 所述初始帧中的车辆参数 ( 即理赔车辆的某部位或某配件) 发生了明显的变化, 比如变形、 弯曲、 出现划 痕, 各部位或配件之间产生位移等; 代表该初始帧为符合理赔要求的理赔图像 , 此时, 确认该初始帧为所述理赔图像, 并获取该理赔图像, 以供在后续步骤 中通过所述理赔图像评估所述理赔车辆的损毁价值。
[0083] S403 , 当所述差异值小于或等于所述第一差异阈值时, 确认所述差异值是否大 于预设的第二差异阈值, 并在所述差异值大于所述第二差异阈值时, 将所述初 始帧和所述差异阈值发送至预设的审核人员; 所述第二差异阈值小于所述第一 差异阈值。
[0084] 也即, 在本实施例中, 设定小于所述第一差异阈值的所述第二差异阈值, 也即 , 在服务器判定该初始帧对应的车辆参数所属部位损毁情况对应的差异值位于 第一差异阈值和第二差异阈值之间时, 说明其损毁情况介于需要理赔和不需要 理赔之间, 其差异不可能通过服务器被自动鉴定出来, 此时, 可以通过预设的 审核人员进行人工判定, 来确定其是否需要进行理赔。
[0085] S404, 在所述差异值小于或等于所述第二差异阈值时, 将所述初始帧对应的所 述车辆参数在所述理赔车辆中的部位标记为非损毁部位。 在该初始帧对应的车 辆参数所述部位的损毁情况对应的差异值小于或等于所述第二差异阈值时, 说 明可直接判定其无需进行理赔, 此时将该部位设定为非损毁部位。
[0086] 在一实施例中, 如图 5所示, 所述步骤 S50包括:
[0087] S501 , 检测所述理赔图像对应的所述差异值是否大于所述第三差异阈值, 所述 第三差异阈值大于所述第一差异阈值。
[0088] 其中, 所述损毁价值为所述理赔图像中的部位出现损毁后, 该损毁对应的金钱 价值; 举例说明, 一所述理赔规则如下: 当所述理赔图像与其对应的所述图像 数据之间的差异值大于第三差异阈值时, 该图像中的理赔车辆的部位的损毁价 值为更换价值; 反之, 该图像中的理赔车辆的部位的损毁价值为修复价值。 所 述更换价值为所述理赔图像中记载的部位进行更换所需耗费的金钱; 所述修复 价值为所述理赔图像中记载的部位进行修复时所需耗费的金钱。
[0089] 其中, 所述第三差异阈值是用于判定所述理赔图像中记载的所述理赔车辆的部 位是属于中度损毁或是低度损毁的阈值, 当所述理赔图像与所述图像数据中对 应的原始图像之间的差异值大于预设的第三差异阈值时, 代表所述理赔图像中 记载的所述理赔车辆的部位属于中度损毁; 当所述理赔图像与所述图像数据中 对应的原始图像之间的差异值小于或等于所述第三差异阈值时, 代表所述理赔 图像中记载的所述理赔车辆的部位属于低度损毁。 必须说明的是, 在检测过程 中, 每一所述理赔图像与其对应的所述图像数据进行比对后均会获得一差异值 , 而与各差异值进行比较的所述第三差异阈值并非一定相同, 根据不同的理赔 图像也即根据不同的理赔车辆的部位, 可以设定不同的第三差异阈值。
[0090] S502, 当所述理赔图像对应的所述差异值大于所述第三阈值时, 确认所述理赔 图像对应的所述理赔车辆的损毁部位的损毁价值为所述理赔规则中对应于该损 毁部位的更换价值。
[0091] 当所述理赔图像与所述图像数据中对应的原始图像之间的差异值大于所述第二 阈值时, 代表所述理赔图像中记载的部位的损毁程度是属于中度损毁, 也即该 部位已经无法进行修复, 须进行更换, 才能使所述理赔车辆达到原有的功能, 故评估该部位的损毁价值为所述更换价值。
[0092] S503 , 当所述理赔图像对应的所述差异值小于或等于所述第三差异阈值时, 确 认所述理赔图像对应的所述理赔车辆的损毁部位的损毁价值为所述理赔规则中 对应于该损毁部位的修复价值。
[0093] 当所述理赔图像与所述图像数据中对应的原始图像之间的差异值小于或等于所 述第二阈值时, 代表所述理赔图像中记载的部位的损毁程度是属于低度损毁, 也即要向恢复所述理赔车辆原有的功能, 只需对该部位进行修复, 故评估该部 位的损毁价值为修复价值。
[0094] S504, 将已确认的所述理赔车辆的所有损毁部位的所述损毁价值之和, 确认为 所述理赔车辆的损毁价值。
[0095] 可理解地, 所述理赔车辆发生损毁的部位可能有多处, 故将上述步骤中所评估 的所有所述理赔车辆出现损毁的部位的损毁价值相加, 也即将所有所述更换价 值或所述修复价值相加, 所得的结果即为所述理赔车辆的损毁价值。
[0096] 在一实施例中, 如图 6所示, 所述步骤 S501之后, 还包括:
[0097] S505 , 当所述理赔图像对应的所述差异值大于所述第三差异阈值时, 检测所述 理赔图像对应的损毁部位是否为不可更换部位。
[0098] 其中, 所述不可更换部位是指当该部位出现损毁时, 只能进行修复而不能进行 更换的部位, 若该部位的损毁程度达到无法更换的程度, 则可以认为所述理赔 车辆的损毁程度已经达到完全损毁, 即所述理赔车辆已经无法通过修复或更换 部位 (配件) 而恢复原有的功能。
[0099] S506 , 当所述理赔图像中对应的损毁部位为不可更换部位时, 确认所述理赔车 辆的损毁价值为所述理赔车辆的价值。
[0100] 也即, , 当所述理赔图像和与其对应的所述图像数据之间的差异值大于所述第 三差异阈值时, 代表所述理赔图像中记载的部位的损毁程度已经达到中度损毁 , 即该损毁部位不仅无法进行修复, 而且又不可更换, 此时, 可判定所述理赔 车辆的损毁程度达到完全损毁, 即所述理赔车辆已经无法通过修复或更换部位 (配件) 而恢复原有的功能, 此时, 确认所述理赔图像对应的所述理赔车辆的 损毁部位的损毁价值为整个所述理赔车辆的价值, 此时无需评估该理赔车辆的 其他部位的损毁价值, 直接将所述理赔车辆的损毁价值确认为所述理赔车辆的 价值即可。
[0101] 本申请通过对定损视频进行分帧处理, 通过与车辆参数进行初步比对, 获取多 个有可能符合理赔要求的初始帧, 进一步地, 根据所述特征数据从初始帧中获 取符合要求的理赔图像, 并通过理赔图像评估出理赔车辆的损毁价值, 可以提 升采集理赔图像的准确率, 同时提高定损效率, 进而提高理赔效率, 增强用户 体验。
[0102] 应理解, 上述实施例中各步骤的序号的大小并不意味着执行顺序的先后, 各过 程的执行顺序应以其功能和内在逻辑确定, 而不应对本申请实施例的实施过程 构成任何限定。
[0103] 在一实施例中, 提供一种车辆定损装置, 该车辆定损装置与上述实施例中车辆 定损方法一一对应。 如图 7所示, 该车辆定损装置包括理赔接收模块 11、 数据获 取模块 12、 图像抓取模块 13、 图像比对模块 14和价值评估模块 15。 各功能模块 详细说明如下:
[0104] 所述理赔接收模块 11, 用于接收用户发送的包含理赔车辆唯一标识的理赔指令 , 根据所述唯一标识获取按照预设的拍摄要求拍摄的所述理赔车辆的定损视频
[0105] 所述数据获取模块 12, 用于获取所述理赔车辆的特征数据; 所述特征数据包含 所述理赔车辆的多个车辆参数, 以及所述理赔车辆未损毁之前与各所述车辆参 数分别对应的多个图像数据;
[0106] 所述图像抓取模块 13 , 用于自所述定损视频中根据各所述车辆参数抓取初始帧 , 并将所述初始帧与所述车辆参数关联;
[0107] 所述图像比对模块 14, 用于根据与所述初始帧关联的所述车辆参数调取与该车 辆参数对应的所述图像数据, 将调取的所述图像数据与所述初始帧进行比对并 获取比对结果, 将所述比对结果符合预设的理赔要求的所述初始帧作为理赔图 像;
[0108] 所述价值评估模块 15 , 用于根据所述理赔图像以及预设的理赔规则评估所述理 赔车辆的损毁价值。
[0109] 在一实施例中, 如图 8所示, 所述图像抓取模块 13包括:
[0110] 图像检测单元 131, 用于遍历所述定损视频的每一帧, 检测所述定损视频的各 所述帧与所述车辆参数是否匹配;
[0111] 图像获取单元 132, 用于当所述定损视频的帧与所述车辆参数匹配时, 获取对 应于同一个所述车辆参数的所有所述帧, 并在获取的所有所述帧中根据预设的 图像要求确认该车辆参数对应的初始帧。
[0112] 在一实施例中, 如图 9所示, 所述图像比对模块 14包括:
[0113] 第一检测单元 141, 用于获取调取的所述图像数据与所述初始帧之间的差异值 , 并检测所述差异值是否大于预设的第一差异阈值;
[0114] 确认单元 142, 用于当所述差异值大于所述第一差异阈值时, 确认所述初始帧 为理赔图像;
[0115] 第二检测单元 143 , 用于当所述差异值小于或等于所述第一差异阈值时, 确认 所述差异值是否大于第二差异阈值, 并在所述差异值大于所述第二差异阈值时 , 将所述初始帧和所述差异阈值发送至预设的审核人员; 所述第二差异阈值小 于所述第一差异阈值;
[0116] 标记单元 144, 用于在所述差异值小于或等于所述第二差异阈值时, 将所述初 始帧对应的所述车辆参数在所述理赔车辆中的部位标记为非损毁部位。
[0117] 在一实施例中, 每一所述理赔图像均记载着所述理赔车辆的一个损毁部位, 所 述理赔规则至少包含一项第三差异阈值与损毁价值之间的对应关系, 所述价值 评估模块还用于:
[0118] 检测所述理赔图像对应的所述差异值是否大于所述第三差异阈值, 所述第三差 异阈值大于所述第一差异阈值;
[0119] 当所述理赔图像对应的所述差异值大于所述第三差异阈值时, 确认所述理赔图 像对应的所述理赔车辆的损毁部位的损毁价值为所述理赔规则中对应于该损毁 部位的更换价值;
[0120] 当所述理赔图像对应的所述差异值小于或等于所述第三差异阈值时, 确认所述 理赔图像对应的所述理赔车辆的损毁部位的损毁价值为所述理赔规则中对应于 该损毁部位的修复价值;
[0121] 将已确认的所述理赔车辆的所有损毁部位的所述损毁价值之和, 确认为所述理 赔车辆的损毁价值。
[0122] 在一实施例中, 所述价值评估模块还用于:
[0123] 当所述理赔图像对应的所述差异值大于所述第三差异阈值时, 检测所述理赔图 像对应的损毁部位是否为不可更换部位;
[0124] 当所述理赔图像中对应的损毁部位为不可更换部位时, 确认所述理赔车辆的损 毁价值为所述理赔车辆的价值。
[0125] 关于车辆定损装置的具体限定可以参见上文中对于车辆定损方法的限定, 在此 不再赘述。 上述车辆定损装置中的各个模块可全部或部分通过软件、 硬件及其 组合来实现。 上述各模块可以硬件形式内嵌于或独立于计算机设备中的处理器 中, 也可以以软件形式存储于计算机设备中的存储器中, 以便于处理器调用执 行以上各个模块对应的操作。
[0126] 在一个实施例中, 提供了一种计算机设备, 该计算机设备可以是服务器, 其内 部结构图可以如图 10所示。 该计算机设备包括通过系统总线连接的处理器、 存 储器、 网络接口和数据库。 其中, 该计算机设备的处理器用于提供计算和控制 能力。 该计算机设备的存储器包括非易失性存储介质、 内存储器。 该非易失性 存储介质存储有操作系统、 计算机可读指令和数据库。 该内存储器为非易失性 存储介质中的操作系统和计算机可读指令的运行提供环境。 该计算机设备的网 络接口用于与外部的终端通过网络连接通信。 该计算机可读指令被处理器执行 时以实现一种车辆定损方法。
[0127] 在一个实施例中, 提供了一种计算机设备, 包括存储器、 处理器及存储在存储 器上并可在处理器上运行的计算机可读指令, 处理器执行计算机可读指令时实 现以下步骤:
[0128] 接收用户发送的包含理赔车辆唯一标识的理赔指令, 根据所述唯一标识获取按 照预设的拍摄要求拍摄的所述理赔车辆的定损视频;
[0129] 获取所述理赔车辆的特征数据; 所述特征数据包含所述理赔车辆的多个车辆参 数, 以及所述理赔车辆未损毁之前与各所述车辆参数分别对应的多个图像数据
[0130] 自所述定损视频中根据各所述车辆参数抓取初始帧, 并将所述初始帧与所述车 辆参数关联;
[0131] 根据与所述初始帧关联的所述车辆参数调取与该车辆参数对应的所述图像数据
, 将调取的所述图像数据与所述初始帧进行比对并获取比对结果, 将所述比对 结果符合预设的理赔要求的所述初始帧作为理赔图像;
[0132] 根据所述理赔图像以及预设的理赔规则评估所述理赔车辆的损毁价值。
[0133] 在一个实施例中, 提供了一个或多个存储有计算机可读指令的非易失性可读存 储介质, 该非易失性可读存储介质上存储有计算机可读指令, 该计算机可读指 令被一个或多个处理器执行时, 使得一个或多个处理器实现以下步骤:
[0134] 接收用户发送的包含理赔车辆唯一标识的理赔指令, 根据所述唯一标识获取按 照预设的拍摄要求拍摄的所述理赔车辆的定损视频;
[0135] 获取所述理赔车辆的特征数据; 所述特征数据包含所述理赔车辆的多个车辆参 数, 以及所述理赔车辆未损毁之前与各所述车辆参数分别对应的多个图像数据
[0136] 自所述定损视频中根据各所述车辆参数抓取初始帧, 并将所述初始帧与所述车 辆参数关联;
[0137] 根据与所述初始帧关联的所述车辆参数调取与该车辆参数对应的所述图像数据 , 将调取的所述图像数据与所述初始帧进行比对并获取比对结果, 将所述比对 结果符合预设的理赔要求的所述初始帧作为理赔图像;
[0138] 根据所述理赔图像以及预设的理赔规则评估所述理赔车辆的损毁价值。
[0139] 本领域普通技术人员可以理解实现上述实施例方法中的全部或部分流程, 是可 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 储于一非易失性计算机可读取存储介质中, 该计算机可读指令在执行时, 可包 括如上述各方法的实施例的流程。 其中, 本申请所提供的各实施例中所使用的 对存储器、 存储、 数据库或其它介质的任何引用, 均可包括非易失性和 /或易失 性存储器。 非易失性存储器可包括只读存储器 (ROM) 、 可编程 ROM (PROM ) 、 电可编程 ROM (EPROM) 、 电可擦除可编程 ROM (EEPROM) 或闪存。 易失性存储器可包括随机存取存储器 (RAM) 或者外部高速缓冲存储器。 作为 说明而非局限, RAM以多种形式可得, 诸如静态 RAM (SRAM) 、 动态 RAM ( DRAM) 、 同步 DRAM (SDRAM) 、 双数据率 SDRAM (DDRSDRAM) 、 增强 型 SDRAM (ESDRAM) 、 同步链路 (Synchlink) DRAM (SLDRAM) 、 存储 器总线 (Rambus) 直接 RAM (RDRAM) 、 直接存储器总线动态 RAM (DRDR AM) 、 以及存储器总线动态 RAM (RDRAM) 等。
[0140] 所属领域的技术人员可以清楚地了解到, 为了描述的方便和简洁, 仅以上述各 功能单元、 模块的划分进行举例说明, 实际应用中, 可以根据需要而将上述功 能分配由不同的功能单元、 模块完成, 即将所述装置的内部结构划分成不同的 功能单元或模块, 以完成以上描述的全部或者部分功能。
[0141] 以上所述实施例仅用以说明本申请的技术方案, 而非对其限制; 尽管参照前述 实施例对本申请进行了详细的说明, 本领域的普通技术人员应当理解: 其依然 可以对前述各实施例所记载的技术方案进行修改, 或者对其中部分技术特征进 行等同替换; 而这些修改或者替换, 并不使相应技术方案的本质脱离本申请各 实施例技术方案的精神和范围, 均应包含在本申请的保护范围之内。
发明概述
技术问题
问题的解决方案
发明的有益效果

Claims

权利要求书
[权利要求 1] 一种车辆定损方法, 其特征在于, 包括:
接收用户发送的包含理赔车辆唯一标识的理赔指令, 根据所述唯一标 识获取按照预设的拍摄要求拍摄的所述理赔车辆的定损视频; 获取所述理赔车辆的特征数据; 所述特征数据包含所述理赔车辆的多 个车辆参数, 以及所述理赔车辆未损毁之前与各所述车辆参数分别对 应的多个图像数据;
自所述定损视频中根据各所述车辆参数抓取初始帧, 并将所述初始帧 与所述车辆参数关联; 根据与所述初始帧关联的所述车辆参数调取与该车辆参数对应的所述 图像数据, 将调取的所述图像数据与所述初始帧进行比对并获取比对 结果, 将所述比对结果符合预设的理赔要求的所述初始帧作为理赔图 像;
根据所述理赔图像以及预设的理赔规则评估所述理赔车辆的损毁价值
[权利要求 2] 如权利要求 1所述的车辆定损方法, 其特征在于, 所述自所述定损视 频中根据各所述车辆参数抓取初始帧, 包括:
遍历所述定损视频的每一帧, 检测所述定损视频的各所述帧与所述车 辆参数是否匹配;
当所述定损视频的帧与所述车辆参数匹配时, 获取对应于同一个所述 车辆参数的所有所述帧, 并在获取的所有所述帧中根据预设的图像要 求确认该车辆参数对应的初始帧。
[权利要求 3] 如权利要求 1所述的车辆定损方法, 其特征在于, 所述将调取的所述 图像数据与所述初始帧进行比对并获取比对结果, 将所述比对结果符 合理赔要求的初始帧作为理赔图像, 包括:
获取调取的所述图像数据与所述初始帧之间的差异值, 并检测所述差 异值是否大于预设的第一差异阈值;
当所述差异值大于所述第一差异阈值时, 确认所述初始帧为理赔图像 当所述差异值小于或等于所述第一差异阈值时, 确认所述差异值是否 大于第二差异阈值, 并在所述差异值大于所述第二差异阈值时, 将所 述初始帧和所述差异阈值发送至预设的审核人员; 所述第二差异阈值 小于所述第一差异阈值;
在所述差异值小于或等于所述第二差异阈值时, 将所述初始帧对应的 所述车辆参数在所述理赔车辆中的部位标记为非损毁部位。
[权利要求 4] 如权利要求 3所述的车辆定损方法, 其特征在于, 每一所述理赔图像 均记载着所述理赔车辆的一个损毁部位, 所述理赔规则至少包含一项 第三差异阈值与损毁价值之间的对应关系, 所述根据所述理赔图像以 及预设的理赔规则评估所述理赔车辆的损毁价值, 包括:
检测所述理赔图像对应的所述差异值是否大于所述第三差异阈值, 所 述第三差异阈值大于所述第一差异阈值;
当所述理赔图像对应的所述差异值大于所述第三差异阈值时, 确认所 述理赔图像对应的所述理赔车辆的损毁部位的损毁价值为所述理赔规 则中对应于该损毁部位的更换价值;
当所述理赔图像对应的所述差异值小于或等于所述第三差异阈值时, 确认所述理赔图像对应的所述理赔车辆的损毁部位的损毁价值为所述 理赔规则中对应于该损毁部位的修复价值;
将已确认的所述理赔车辆的所有损毁部位的所述损毁价值之和, 确认 为所述理赔车辆的损毁价值。
[权利要求 5] 如权利要求 4所述的车辆定损方法, 其特征在于, 所述检测所述理赔 图像对应的所述差异值是否大于所述第三差异阈值, 所述第三差异阈 值大于所述第一差异阈值之后, 还包括:
当所述理赔图像对应的所述差异值大于所述第三差异阈值时, 检测所 述理赔图像对应的损毁部位是否为不可更换部位; 当所述理赔图像中对应的损毁部位为不可更换部位时, 确认所述理赔 车辆的损毁价值为所述理赔车辆的价值。
[权利要求 6] 一种车辆定损装置, 其特征在于, 包括:
理赔接收模块, 用于接收用户发送的包含理赔车辆唯一标识的理赔指 令, 根据所述唯一标识获取按照预设的拍摄要求拍摄的所述理赔车辆 的定损视频;
数据获取模块, 用于获取所述理赔车辆的特征数据; 所述特征数据包 含所述理赔车辆的多个车辆参数, 以及所述理赔车辆未损毁之前与各 所述车辆参数分别对应的多个图像数据; 图像抓取模块, 用于自所述定损视频中根据各所述车辆参数抓取初始 帧, 并将所述初始帧与所述车辆参数关联;
图像比对模块, 用于根据与所述初始帧关联的所述车辆参数调取与该 车辆参数对应的所述图像数据, 将调取的所述图像数据与所述初始帧 进行比对并获取比对结果, 将所述比对结果符合预设的理赔要求的所 述初始帧作为理赔图像;
价值评估模块, 用于根据所述理赔图像以及预设的理赔规则评估所述 理赔车辆的损毁价值。
[权利要求 7] 如权利要求 6所述的车辆定损装置, 其特征在于, 所述图像抓取模块
, 包括:
图像检测单元, 用于遍历所述定损视频的每一帧, 检测所述定损视频 的各所述帧与所述车辆参数是否匹配;
图像获取单元, 用于当所述定损视频的帧与所述车辆参数匹配时, 获 取对应于同一个所述车辆参数的所有所述帧, 并在获取的所有所述帧 中根据预设的图像要求确认该车辆参数对应的初始帧。
[权利要求 8] 如权利要求 6所述的车辆定损装置, 其特征在于, 所述图像比对模块
, 包括:
第一检测单元, 用于获取调取的所述图像数据与所述初始帧之间的差 异值, 并检测所述差异值是否大于预设的第一差异阈值;
确认单元, 用于当所述差异值大于所述第一差异阈值时, 确认所述初 始帧为理赔图像; 第二检测单元, 用于当所述差异值小于或等于所述第一差异阈值时, 确认所述差异值是否大于第二差异阈值, 并在所述差异值大于所述第 二差异阈值时, 将所述初始帧和所述差异阈值发送至预设的审核人员 ; 所述第二差异阈值小于所述第一差异阈值;
标记单元, 用于在所述差异值小于或等于所述第二差异阈值时, 将所 述初始帧对应的所述车辆参数在所述理赔车辆中的部位标记为非损毁 部位。
[权利要求 9] 如权利要求 8所述的车辆定损装置, 其特征在于, 每一所述理赔图像 均记载着所述理赔车辆的一个损毁部位, 所述理赔规则至少包含一项 第三差异阈值与损毁价值之间的对应关系, 所述价值评估模块还用于 检测所述理赔图像对应的所述差异值是否大于所述第三差异阈值, 所 述第三差异阈值大于所述第一差异阈值;
当所述理赔图像对应的所述差异值大于所述第三差异阈值时, 确认所 述理赔图像对应的所述理赔车辆的损毁部位的损毁价值为所述理赔规 则中对应于该损毁部位的更换价值;
当所述理赔图像对应的所述差异值小于或等于所述第三差异阈值时, 确认所述理赔图像对应的所述理赔车辆的损毁部位的损毁价值为所述 理赔规则中对应于该损毁部位的修复价值;
将已确认的所述理赔车辆的所有损毁部位的所述损毁价值之和, 确认 为所述理赔车辆的损毁价值。
[权利要求 10] 如权利要求 9所述的车辆定损装置, 其特征在于, 所述价值评估模块 还用于:
当所述理赔图像对应的所述差异值大于所述第三差异阈值时, 检测所 述理赔图像对应的损毁部位是否为不可更换部位; 当所述理赔图像中对应的损毁部位为不可更换部位时, 确认所述理赔 车辆的损毁价值为所述理赔车辆的价值。
[权利要求 11] 一种计算机设备, 包括存储器、 处理器以及存储在所述存储器中并可 在所述处理器上运行的计算机可读指令, 其特征在于, 所述处理器执 行所述计算机可读指令时实现如下步骤:
接收用户发送的包含理赔车辆唯一标识的理赔指令, 根据所述唯一标 识获取按照预设的拍摄要求拍摄的所述理赔车辆的定损视频; 获取所述理赔车辆的特征数据; 所述特征数据包含所述理赔车辆的多 个车辆参数, 以及所述理赔车辆未损毁之前与各所述车辆参数分别对 应的多个图像数据;
自所述定损视频中根据各所述车辆参数抓取初始帧, 并将所述初始帧 与所述车辆参数关联; 根据与所述初始帧关联的所述车辆参数调取与该车辆参数对应的所述 图像数据, 将调取的所述图像数据与所述初始帧进行比对并获取比对 结果, 将所述比对结果符合预设的理赔要求的所述初始帧作为理赔图 像;
根据所述理赔图像以及预设的理赔规则评估所述理赔车辆的损毁价值
[权利要求 12] 如权利要求 11所述的计算机设备, 其特征在于, 所述自所述定损视频 中根据各所述车辆参数抓取初始帧, 包括:
遍历所述定损视频的每一帧, 检测所述定损视频的各所述帧与所述车 辆参数是否匹配;
当所述定损视频的帧与所述车辆参数匹配时, 获取对应于同一个所述 车辆参数的所有所述帧, 并在获取的所有所述帧中根据预设的图像要 求确认该车辆参数对应的初始帧。
[权利要求 13] 如权利要求 11所述的计算机设备, 其特征在于, 所述将调取的所述图 像数据与所述初始帧进行比对并获取比对结果, 将所述比对结果符合 理赔要求的初始帧作为理赔图像, 包括:
获取调取的所述图像数据与所述初始帧之间的差异值, 并检测所述差 异值是否大于预设的第一差异阈值;
当所述差异值大于所述第一差异阈值时, 确认所述初始帧为理赔图像 当所述差异值小于或等于所述第一差异阈值时, 确认所述差异值是否 大于第二差异阈值, 并在所述差异值大于所述第二差异阈值时, 将所 述初始帧和所述差异阈值发送至预设的审核人员; 所述第二差异阈值 小于所述第一差异阈值;
在所述差异值小于或等于所述第二差异阈值时, 将所述初始帧对应的 所述车辆参数在所述理赔车辆中的部位标记为非损毁部位。
[权利要求 14] 如权利要求 13所述的计算机设备, 其特征在于, 每一所述理赔图像均 记载着所述理赔车辆的一个损毁部位, 所述理赔规则至少包含一项第 三差异阈值与损毁价值之间的对应关系, 所述根据所述理赔图像以及 预设的理赔规则评估所述理赔车辆的损毁价值, 包括:
检测所述理赔图像对应的所述差异值是否大于所述第三差异阈值, 所 述第三差异阈值大于所述第一差异阈值;
当所述理赔图像对应的所述差异值大于所述第三差异阈值时, 确认所 述理赔图像对应的所述理赔车辆的损毁部位的损毁价值为所述理赔规 则中对应于该损毁部位的更换价值;
当所述理赔图像对应的所述差异值小于或等于所述第三差异阈值时, 确认所述理赔图像对应的所述理赔车辆的损毁部位的损毁价值为所述 理赔规则中对应于该损毁部位的修复价值;
将已确认的所述理赔车辆的所有损毁部位的所述损毁价值之和, 确认 为所述理赔车辆的损毁价值。
[权利要求 15] 如权利要求 14所述的计算机设备, 其特征在于, 所述检测所述理赔图 像对应的所述差异值是否大于所述第三差异阈值, 所述第三差异阈值 大于所述第一差异阈值之后, 所述处理器执行所述计算机可读指令时 还实现如下步骤:
当所述理赔图像对应的所述差异值大于所述第三差异阈值时, 检测所 述理赔图像对应的损毁部位是否为不可更换部位; 当所述理赔图像中对应的损毁部位为不可更换部位时, 确认所述理赔 车辆的损毁价值为所述理赔车辆的价值。
[权利要求 16] 一个或多个存储有计算机可读指令的非易失性可读存储介质, 其特征 在于, 所述计算机可读指令被一个或多个处理器执行时, 使得所述一 个或多个处理器执行如下步骤:
接收用户发送的包含理赔车辆唯一标识的理赔指令, 根据所述唯一标 识获取按照预设的拍摄要求拍摄的所述理赔车辆的定损视频; 获取所述理赔车辆的特征数据; 所述特征数据包含所述理赔车辆的多 个车辆参数, 以及所述理赔车辆未损毁之前与各所述车辆参数分别对 应的多个图像数据;
自所述定损视频中根据各所述车辆参数抓取初始帧, 并将所述初始帧 与所述车辆参数关联; 根据与所述初始帧关联的所述车辆参数调取与该车辆参数对应的所述 图像数据, 将调取的所述图像数据与所述初始帧进行比对并获取比对 结果, 将所述比对结果符合预设的理赔要求的所述初始帧作为理赔图 像;
根据所述理赔图像以及预设的理赔规则评估所述理赔车辆的损毁价值
[权利要求 17] 如权利要求 16所述的非易失性可读存储介质, 其特征在于, 所述自所 述定损视频中根据各所述车辆参数抓取初始帧, 包括:
遍历所述定损视频的每一帧, 检测所述定损视频的各所述帧与所述车 辆参数是否匹配;
当所述定损视频的帧与所述车辆参数匹配时, 获取对应于同一个所述 车辆参数的所有所述帧, 并在获取的所有所述帧中根据预设的图像要 求确认该车辆参数对应的初始帧。
[权利要求 18] 如权利要求 16所述的非易失性可读存储介质, 其特征在于, 所述将调 取的所述图像数据与所述初始帧进行比对并获取比对结果, 将所述比 对结果符合理赔要求的初始帧作为理赔图像, 包括:
获取调取的所述图像数据与所述初始帧之间的差异值, 并检测所述差 异值是否大于预设的第一差异阈值;
当所述差异值大于所述第一差异阈值时, 确认所述初始帧为理赔图像 当所述差异值小于或等于所述第一差异阈值时, 确认所述差异值是否 大于第二差异阈值, 并在所述差异值大于所述第二差异阈值时, 将所 述初始帧和所述差异阈值发送至预设的审核人员; 所述第二差异阈值 小于所述第一差异阈值;
在所述差异值小于或等于所述第二差异阈值时, 将所述初始帧对应的 所述车辆参数在所述理赔车辆中的部位标记为非损毁部位。
[权利要求 19] 如权利要求 18所述的非易失性可读存储介质, 其特征在于, 每一所述 理赔图像均记载着所述理赔车辆的一个损毁部位, 所述理赔规则至少 包含一项第三差异阈值与损毁价值之间的对应关系, 所述根据所述理 赔图像以及预设的理赔规则评估所述理赔车辆的损毁价值, 包括: 检测所述理赔图像对应的所述差异值是否大于所述第三差异阈值, 所 述第三差异阈值大于所述第一差异阈值;
当所述理赔图像对应的所述差异值大于所述第三差异阈值时, 确认所 述理赔图像对应的所述理赔车辆的损毁部位的损毁价值为所述理赔规 则中对应于该损毁部位的更换价值;
当所述理赔图像对应的所述差异值小于或等于所述第三差异阈值时, 确认所述理赔图像对应的所述理赔车辆的损毁部位的损毁价值为所述 理赔规则中对应于该损毁部位的修复价值;
将已确认的所述理赔车辆的所有损毁部位的所述损毁价值之和, 确认 为所述理赔车辆的损毁价值。
[权利要求 20] 如权利要求 19所述的非易失性可读存储介质, 其特征在于, 所述检测 所述理赔图像对应的所述差异值是否大于所述第三差异阈值, 所述第 三差异阈值大于所述第一差异阈值之后, 所述计算机可读指令被一个 或多个处理器执行时, 使得所述一个或多个处理器执行如下步骤: 当所述理赔图像对应的所述差异值大于所述第三差异阈值时, 检测所 述理赔图像对应的损毁部位是否为不可更换部位;
当所述理赔图像中对应的损毁部位为不可更换部位时, 确认所述理赔 车辆的损毁价值为所述理赔车辆的价值。
PCT/CN2019/101910 2019-04-09 2019-08-22 车辆定损方法、装置、计算机设备及存储介质 WO2020206900A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201910280404.3A CN110147719A (zh) 2019-04-09 2019-04-09 车辆定损方法、装置、计算机设备及存储介质
CN201910280404.3 2019-04-09

Publications (1)

Publication Number Publication Date
WO2020206900A1 true WO2020206900A1 (zh) 2020-10-15

Family

ID=67588953

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2019/101910 WO2020206900A1 (zh) 2019-04-09 2019-08-22 车辆定损方法、装置、计算机设备及存储介质

Country Status (2)

Country Link
CN (1) CN110147719A (zh)
WO (1) WO2020206900A1 (zh)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113361424A (zh) * 2021-06-11 2021-09-07 爱保科技有限公司 一种车辆智能定损图像获取方法、装置、介质和电子设备
CN114241398A (zh) * 2022-02-23 2022-03-25 深圳壹账通科技服务有限公司 基于人工智能的车辆定损方法、装置、设备及存储介质

Families Citing this family (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110147719A (zh) * 2019-04-09 2019-08-20 平安科技(深圳)有限公司 车辆定损方法、装置、计算机设备及存储介质
CN111489433B (zh) * 2020-02-13 2023-04-25 北京百度网讯科技有限公司 车辆损伤定位的方法、装置、电子设备以及可读存储介质
CN112053250A (zh) * 2020-09-03 2020-12-08 中国平安财产保险股份有限公司 基于视频的车险自助理赔方法、装置、设备
CN112085721A (zh) * 2020-09-07 2020-12-15 中国平安财产保险股份有限公司 基于人工智能的水淹车定损方法、装置、设备及存储介质
CN112348686B (zh) * 2020-11-24 2021-07-13 德联易控科技(北京)有限公司 理赔图片的采集方法、装置及通讯设备
CN115994910B (zh) * 2023-03-24 2023-06-06 邦邦汽车销售服务(北京)有限公司 基于数据处理的汽车损伤程度确定方法及系统

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20070159309A1 (en) * 2005-09-30 2007-07-12 Omron Corporation Information processing apparatus and information processing method, information processing system, program, and recording media
CN106600422A (zh) * 2016-11-24 2017-04-26 中国平安财产保险股份有限公司 一种车险智能定损方法和系统
CN107730485A (zh) * 2017-08-03 2018-02-23 上海壹账通金融科技有限公司 车辆定损方法、电子设备及计算机可读存储介质
CN108875648A (zh) * 2018-06-22 2018-11-23 深源恒际科技有限公司 一种基于手机视频流的实时车辆损伤和部件检测的方法
CN108961085A (zh) * 2018-07-09 2018-12-07 北京精友世纪软件技术有限公司 一种车险智能定损系统及方法
CN109146834A (zh) * 2017-06-13 2019-01-04 上海擎感智能科技有限公司 车辆定损方法及装置、计算机可读存储介质、终端
CN110147719A (zh) * 2019-04-09 2019-08-20 平安科技(深圳)有限公司 车辆定损方法、装置、计算机设备及存储介质

Family Cites Families (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106952165B (zh) * 2017-03-13 2021-08-24 平安科技(深圳)有限公司 车险理赔定损的方法及系统
CN107886436B (zh) * 2017-11-08 2021-12-14 广东翼卡车联网服务有限公司 一种车辆保险自动理赔的方法、存储介质及车载终端
CN108335217B (zh) * 2018-01-25 2020-11-13 中国平安财产保险股份有限公司 车辆出险的模拟理赔方法、装置、设备和计算机存储介质
CN108921068B (zh) * 2018-06-22 2020-10-20 深源恒际科技有限公司 一种基于深度神经网络的汽车外观自动定损方法及系统
CN108898504B (zh) * 2018-07-09 2021-12-07 北京精友世纪软件技术有限公司 一种移动查勘定损系统的智能训练及完善方法
CN109285079A (zh) * 2018-08-31 2019-01-29 阿里巴巴集团控股有限公司 终端屏幕保险的数据处理方法、装置、客户端及服务器

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20070159309A1 (en) * 2005-09-30 2007-07-12 Omron Corporation Information processing apparatus and information processing method, information processing system, program, and recording media
CN106600422A (zh) * 2016-11-24 2017-04-26 中国平安财产保险股份有限公司 一种车险智能定损方法和系统
CN109146834A (zh) * 2017-06-13 2019-01-04 上海擎感智能科技有限公司 车辆定损方法及装置、计算机可读存储介质、终端
CN107730485A (zh) * 2017-08-03 2018-02-23 上海壹账通金融科技有限公司 车辆定损方法、电子设备及计算机可读存储介质
CN108875648A (zh) * 2018-06-22 2018-11-23 深源恒际科技有限公司 一种基于手机视频流的实时车辆损伤和部件检测的方法
CN108961085A (zh) * 2018-07-09 2018-12-07 北京精友世纪软件技术有限公司 一种车险智能定损系统及方法
CN110147719A (zh) * 2019-04-09 2019-08-20 平安科技(深圳)有限公司 车辆定损方法、装置、计算机设备及存储介质

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113361424A (zh) * 2021-06-11 2021-09-07 爱保科技有限公司 一种车辆智能定损图像获取方法、装置、介质和电子设备
CN114241398A (zh) * 2022-02-23 2022-03-25 深圳壹账通科技服务有限公司 基于人工智能的车辆定损方法、装置、设备及存储介质

Also Published As

Publication number Publication date
CN110147719A (zh) 2019-08-20

Similar Documents

Publication Publication Date Title
WO2020206900A1 (zh) 车辆定损方法、装置、计算机设备及存储介质
US10803532B1 (en) Processing insured items holistically with mobile damage assessment and claims processing
US20230186401A1 (en) Feedback loop in mobile damage assessment and claims processing
CN108364253B (zh) 车辆定损方法、系统以及电子设备
US11915321B2 (en) Configuration and transfer of image data using a mobile device
CN111666995A (zh) 基于深度学习模型的车辆定损方法、装置、设备及介质
US10533849B2 (en) Analysis apparatus and analysis program
CN109598519B (zh) 车辆审核方法、装置、计算机设备和存储介质
KR102125999B1 (ko) Ai 자동 견적 장치, 방법 및 컴퓨터 프로그램
CN111325128A (zh) 违规操作检测方法、装置、计算机设备和存储介质
CN114113120A (zh) 一种电池包底部损伤的检测方法、系统、设备和介质
CN110909731A (zh) 车辆救援场景质检方法、装置、设备及存储介质
CN110704459B (zh) 停车场僵尸车辆数据的清理方法、介质及系统
CN112672102B (zh) 一种视频生成方法及装置
CN116645530A (zh) 基于图像比对的施工检测方法、装置、设备及存储介质
CN112529213A (zh) 派工校验方法、装置、系统及计算机可读存储介质
CN110599801B (zh) 停车管理方法、装置、计算机设备及存储介质
CN111768357B (zh) 一种图像检测的方法及装置
CN112052150A (zh) 页面加载时间检测方法、设备、存储介质及装置
WO2023181255A1 (ja) 情報提供装置、事故情報の提供方法及びプログラム記録媒体
CN111079466A (zh) 一种车辆识别方法、装置、电子设备及存储介质
WO2021125294A1 (ja) 貸与対象物管理システム、貸与対象物管理プログラム及び貸与対象物管理方法。
CN109767346B (zh) 车辆业务处理方法、装置、计算机设备和存储介质
EP4343714A1 (en) System and method for automated image analysis for damage analysis
CN117496149A (zh) 车险赔付方案生成方法、装置、计算机设备及存储介质

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: 19923784

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: 19923784

Country of ref document: EP

Kind code of ref document: A1