KR20220036308A - Method and system for guiding car repair based on image analysis - Google Patents

Abstract

According to an embodiment of the present invention, a method for a quotation information server to provide a guide for vehicle repair based on an image includes the steps of: (a) receiving a captured image of a damaged part of the vehicle from a user terminal or a repair shop terminal; (b) calculating an estimate range for vehicle repair by analyzing the image based on previously learned data; and (c) based on the calculated quotation range, comparing the total cost required for insurance processing and the quotation range to determine whether insurance processing is necessary. A method of providing a vehicle repair guide is provided.

Description

Method and system for providing vehicle repair guide based on image analysis {METHOD AND SYSTEM FOR GUIDING CAR REPAIR BASED ON IMAGE ANALYSIS}

The present invention relates to a method and system for providing a vehicle repair guide based on image analysis, and to a method and system for determining whether insurance treatment is necessary and providing guidance on an appropriate repair estimate by simply taking images of an accident vehicle.

When an accident occurs while driving a vehicle, the automobile insurance company is generally contacted and an insurance company employee is called to handle the accident.

If the damage to the vehicle is not serious, the driver drives the vehicle himself and moves the vehicle. If the damage to the vehicle is serious, it is towed by a tow truck.

Drivers have no choice but to visually determine vehicle damage, but unless you are an expert, it is not easy to determine the approximate repair cost.

In addition, when signing up for car insurance, you sign up for your own vehicle damage insurance. If you repair damage to your car through this insurance, you will incur a deductible, and there may be an insurance premium surcharge when renewing your car insurance in the future.

It is difficult for drivers to accurately determine the actual cost of repairing their vehicle through damage insurance and the cost of repairing it at a private company at their own expense.

Therefore, from the driver's perspective, it is not easy to make a decision about whether to apply insurance for repairs to current vehicle damage.

The purpose of the present invention is to solve the problems of the prior art described above.

The purpose of the present invention is to enable drivers to easily make decisions about whether or not to take out insurance in relation to repairs to vehicle damage.

The purpose of the present invention is to make it possible to easily determine the approximate range of repair costs based on captured images of vehicle damage.

The purpose of the present invention is to facilitate the collection of learning data required to perform image analysis-based repair cost calculation using machine learning techniques.

The objects of the present invention are not limited to the objects mentioned above, and other objects not mentioned can be clearly understood from the description below.

According to an embodiment of the present invention for achieving the above-described object, a quotation information server is a method for providing a guide for vehicle repair based on an image, (a) sending a message to the damaged part of the vehicle from a user terminal or a repair shop terminal; receiving a captured image; (b) calculating an estimate range for vehicle repair by analyzing the image based on previously learned data; and (c) based on the calculated quotation range, comparing the total cost required for insurance processing and the quotation range to determine whether insurance processing is necessary. A method of providing a vehicle repair guide is provided.

The method of providing a vehicle repair guide further includes, prior to step (a), performing learning based on captured image information about the damaged part of the vehicle and actual repair cost estimate information from a plurality of user terminals or repair shop terminals. can do.

The method of providing a vehicle repair guide includes, after step (a), receiving repair estimate information for the vehicle from the user terminal or a repair shop terminal; And it may further include determining whether the repair estimate information is an appropriate estimate based on previously learned data.

Step (c) may include receiving the user's insurance subscription information from the insurance company server based on the personal information received from the user terminal, and then calculating the total cost required for insurance processing.

According to another embodiment of the present invention, an image receiving unit that receives a captured image of a damaged part of a vehicle from a user terminal or a repair shop terminal; An estimate calculation unit that analyzes the image and calculates an estimate range for vehicle repair based on previously learned data; and an insurance processing determination unit that determines whether insurance processing is necessary by comparing the total cost required for insurance processing with the quotation range based on the calculated quotation range. A vehicle repair guide providing system is provided.

According to an embodiment of the present invention, it is possible to calculate an estimate range for the approximate repair cost based only on image information about the damaged part of the vehicle.

According to an embodiment of the present invention, the cost required for insurance processing is automatically calculated based on the calculated quotation information, making it possible to determine whether insurance processing is necessary.

According to an embodiment of the present invention, it is possible to recommend a repair shop that can perform repairs at a lower amount than the cost required for insurance processing.

1 is a diagram showing the schematic configuration of a vehicle repair guide system according to an embodiment of the present invention.
Figure 2 is a block diagram for explaining the operation of the quotation information server 200 according to an embodiment of the present invention.

The detailed description of the present invention described below refers to the accompanying drawings, which show by way of example specific embodiments in which the present invention may be practiced. These embodiments are described in sufficient detail to enable those skilled in the art to practice the invention. It should be understood that the various embodiments of the invention are different from one another but are not necessarily mutually exclusive. For example, specific shapes, structures and characteristics described herein with respect to one embodiment may be implemented in other embodiments without departing from the spirit and scope of the invention. Additionally, it should be understood that the location or arrangement of individual components within each disclosed embodiment may be changed without departing from the spirit and scope of the invention. Accordingly, the detailed description that follows is not intended to be taken in a limiting sense, and the scope of the invention is limited only by the appended claims, together with all equivalents to what those claims assert, if properly described. Similar reference numbers in the drawings refer to identical or similar functions across various aspects.

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the attached drawings in order to enable those skilled in the art to easily practice the present invention.

1 is a diagram showing the schematic configuration of a vehicle repair estimate guidance system according to an embodiment of the present invention.

Referring to FIG. 1 , a vehicle repair quotation guidance system according to an embodiment may be configured to include a user terminal 100, an quotation guidance server 200, an insurance company server 300, and a repair shop terminal 400.

The user terminal 100, the quotation information server 200, the insurance company server 300, and the repair shop terminal 400 are connected to a mutual communication network, for example, a mobile communication network, a local area network (LAN), and a metropolitan area network (MAN). They can communicate with each other through the Metropolitan Area Network (WAN), Wide Area Network (WWW), and Wireless Fidelity (WiFi).

The user terminal 100 and the repair shop terminal 400 have calculation functions and may be implemented in any form as long as they are devices capable of communicating with the outside. For example, it may be implemented as a smartphone, tablet PC, desktop, laptop, PDA, etc., but is not limited thereto.

Meanwhile, the user terminal 100 and the repair shop terminal 400 implement the functions described below and may include an application for receiving services provided by the quotation information server 200. These applications may be downloaded and installed from the quotation information server 200 or another app store server (not shown).

The user terminal 100 performs a function of transmitting a captured image of the damaged part of the vehicle to the quotation information server 200. The captured image of the damaged area may be input to the user terminal 100 from outside, or may be captured by a photographing module included in the user terminal 100.

The quotation information server 200 analyzes the image received from the user terminal 100, determines the degree of damage to the vehicle, and calculates a repair quotation range accordingly. This process can be performed through machine learning algorithms. For example, after extracting a feature vector from an image, the extent of damage can be recognized through an artificial neural network learning method, and the extent of damage and repair estimate range can be calculated. Artificial neural networks can be divided into unsupervised learning and supervised learning. The supervised learning algorithm requires a teacher signal to indicate whether the output result of the input data matches the intended result, allowing artificial control and reliability of the learned system. is high. Therefore, this supervised learning algorithm can be utilized.

For this purpose, the quotation information server 200 requires a plurality of learning data. This learning data includes image information about the damaged part of the vehicle received from the user terminal 100, the user terminal 100, or the repair shop terminal 400. It may be actual repair estimate information for damage to the vehicle received from .

When receiving image information from the user terminal 100, information on the name of the damaged part (e.g., bumper, fender, bonnet, door, etc.) directly input by the user may also be received, and learning is performed through this data. If performed, labeling for each image becomes possible and learning is accumulated, it will be possible to identify the damaged area even if only image information is received without the user inputting the name of the damaged area.

When receiving actual vehicle repair estimate information from the user terminal 100 or the repair shop terminal 400, identification information about the user or vehicle identification information may also be received. As a result, it is possible to match the previously received image information about the damaged area with the corresponding estimate information.

The vehicle repair quotation information may be a scanned image of the quotation, and the quotation information server 200 can obtain quotation-related text information included in the image through a text recognition algorithm (OCR: Optical Character Reader).

The quotation information server 200 determines whether insurance processing for the repair is necessary by linking with the insurance company server 400 based on the repair quotation information received from the user terminal 100 or the repair shop terminal 400 or the repair quotation information determined by itself. A judgment can be made and the results can be provided to the user terminal 100.

The insurance company server 300 stores the insurance subscription information of each subscriber and, based on the information received from the quotation information server 200, transmits information about costs incurred during insurance processing to the quotation information server 200. .

In Figure 1, the quotation information server 200 and the insurance company server 300 are illustrated as being implemented as separate servers, but they may be implemented as a single server.

The repair shop terminal 400 is a terminal provided at each vehicle repair shop, and can transmit a captured image of vehicle damage to the quotation information server 200 on behalf of the user terminal 100, and provide a repair quotation for the vehicle in the form of an image. It can also be transmitted to the quotation information server 200.

Figure 2 is a block diagram for explaining the operation of the quotation information server 200 according to an embodiment of the present invention.

Referring to FIG. 2, the quotation information server 200 includes an image receiving unit 210, a quotation receiving unit 220, a data learning unit 230, a quotation calculating unit 240, an insurance processing determination unit 250, and a repair shop. It may be configured to include a recommendation unit 260.

The image receiving unit 210, the quotation receiving unit 220, the data learning unit 230, the quotation calculating unit 240, the insurance processing determination unit 250, and the repair shop recommendation unit 260 are programs that can communicate with external devices. It may be a module or hardware. These program modules or hardware may be included in the quotation guidance server 200 or other devices capable of communicating therewith in the form of operating systems, application modules, and other program modules, and may be physically stored on various known storage devices. there is. Meanwhile, such program modules or hardware include, but are not limited to, routines, subroutines, programs, objects, components, data structures, etc. that perform specific tasks or execute specific abstract data types according to the present invention.

The image receiver 210 receives a captured image of vehicle damage from the user terminal 100 or the repair shop terminal 400. In addition, information on the name of the damaged part, vehicle model information, manufacturer information, model name information, etc. can also be received.

The quotation receiving unit 220 performs a function of receiving actual repair quotation information for the vehicle for which the captured image has been received from the user terminal 100 or the repair shop terminal 400. In addition, it performs the function of identifying the contents of the quotation received in image form through a text recognition algorithm (e.g., OCR, etc.).

The data learning unit 230 functions to perform artificial neural network learning based on images of damaged parts and estimate information for each of a plurality of vehicles. Through quote analysis information, it is possible to determine information about repair items and cost information for each repair item. By learning image information about the damaged area and the quote analysis information, the degree of damage to the vehicle is determined based on image analysis. It becomes possible to identify, identify necessary repair items according to the degree of damage, and calculate estimates for each repair item. In other words, by simply receiving image information about the damaged part of the vehicle through learning, it can be possible to calculate an estimate range for the approximate repair cost. Additionally, as learning data accumulates, it will be possible to determine where the damaged part of the vehicle is and what the manufacturer and model name of the vehicle are based on image analysis.

The estimate calculation unit 240 determines the approximate range of the repair estimate for the vehicle based on the image information received from the user terminal 100 or the repair shop terminal 400 and the data learned by the data learning unit 230. It can be calculated.

In addition, after the learning by the data learning unit 230 is accumulated, the quotation calculation unit 240 analyzes the image received from the user terminal 100 or the repair shop terminal 400 and the actual quotation, and It may be possible to determine whether the quoted estimate is an appropriate estimate.

For example, when learning is accumulated, when the front bumper of a BMW 520d is partially dented, the repair item is sheet metal painting, and the repair cost is 300,000 to 400,000 won, and data will be obtained from the specific user terminal 100. If image data for the same part of the same vehicle type is obtained, and as a result of analyzing the quote actually received from the user terminal 100, it is determined that a specific repair shop gave a quote of 600,000 won, the user terminal 100 is considered to be an excessive quote. and can provide information on appropriate estimates.

The insurance processing determination unit 250 includes information on the quotation range calculated based on the analysis result of the received image, quotation information actually received from the user terminal 100 or the repair shop terminal 400, or the quotation calculation unit 240. ) performs the function of determining whether the user needs to insure the repair of the vehicle based on information about the appropriate estimate calculated by ).

For this purpose, when registering as a member or receiving an image of a vehicle damage area from the user terminal 100, the user's personal information (name, phone number, date of birth, information on the insurance company of the subscribed car insurance, etc.) can be received, and such personal information can be transmitted to the specific insurance company server 300 to receive the user's car insurance subscription information from the insurance company server 300.

Car insurance subscription information includes information about your vehicle damage insurance. Additionally, this information may include information on the vehicle value set when signing up for insurance, information on the standard material accident surcharge amount, and information on deductibles.

The insurance processing decision unit 250 calculates the cost incurred when insurance is processed based on the repair estimate obtained, compares this cost with the repair estimate, and determines whether insurance processing is necessary. Derive the result value.

For example, it is assumed that the repair estimate is 1 million won, and as a result of obtaining the driver's insurance subscription information, the insurance premium for the year is 1.5 million won, the standard material accident surcharge is 500,000 won, and the deductible is 20%. If you try it, when processing insurance, the deductible will be 200,000 won (=1 million won

If insurance is not processed, next year's insurance premium will be discounted to approximately 1.35 million won (assuming a 10% discount), so the amount lost in insurance premiums when insurance is processed will be 300,000 won (=1.65 million won - 1.35 million won).

In other words, the total amount you need to handle when processing insurance is 500,000 won (=200,000 won + 300,000 won).

If insurance is not processed, the repair cost will be 1 million won, so the insurance decision determination unit 250 concludes that it is economically advantageous to insure, and transmits the relevant information to the user terminal 100.

Meanwhile, the depreciation cost of the vehicle may be added to the cost incurred in processing the insurance. Specifically, when insurance is processed for vehicle repairs, an insurance history is left. Considering the vehicle type, current vehicle value, repair cost information that remains in the history when insurance is processed, and depreciation costs, the amount to be processed by the person is the amount to be processed when processing insurance. By adding these depreciation costs, you can compare them to the repair costs that would have been incurred if insurance was not taken care of.

The repair shop recommendation unit 260 provides the user terminal 100 with information about a repair shop that can perform repairs at a lower amount than the total cost required when insurance is processed, based on the judgment result by the insurance processing determination unit 250. It performs the function it provides.

To this end, the repair shop recommendation unit 260 can transmit the image of the damaged vehicle and information on the total cost required for insurance processing to a plurality of repair shop terminals 400, and receive an amount lower than the total cost from the repair shop terminal 400. By receiving information about whether repairs are possible or quotation information, information about a repair shop that provides a quotation at a lower amount than the total cost can be provided to the user terminal 100.

According to an embodiment of the present invention, it is possible to calculate an estimate range for the approximate repair cost based only on image information about the damaged part of the vehicle.

According to an embodiment of the present invention, the cost required for insurance processing is automatically calculated based on the calculated quotation information, making it possible to determine whether insurance processing is necessary.

According to an embodiment of the present invention, it is possible to recommend a repair shop that can perform repairs at a lower amount than the cost required for insurance processing.

The description of the present invention described above is for illustrative purposes, and those skilled in the art will understand that the present invention can be easily modified into other specific forms without changing the technical idea or essential features of the present invention. will be. Therefore, the embodiments described above should be understood in all respects as illustrative and not restrictive. For example, each component described as single may be implemented in a distributed manner, and similarly, components described as distributed may also be implemented in a combined form.

The scope of the present invention is indicated by the patent claims described below, and all changes or modified forms derived from the meaning and scope of the claims and their equivalent concepts should be construed as being included in the scope of the present invention.

100: user terminal
200: Quotation information server
210: Image receiving unit
220: Quotation receiving unit
230: Data learning unit
240: Quotation calculation unit
250: Insurance processing determination unit
260: Repair shop recommendation section

Claims (5)

As a method for the quotation information server to provide a guide for vehicle repair based on an image,
(a) receiving a captured image of the damaged part of the vehicle from a user terminal or a repair shop terminal;
(b) calculating an estimate range for vehicle repair by analyzing the image based on previously learned data; and
(c) A method of providing a vehicle repair guide, including the step of determining whether insurance processing is necessary by comparing the total cost required for insurance processing and the estimate range based on the calculated quotation range. According to paragraph 1,
Before step (a) above,
A method of providing a vehicle repair guide, further comprising performing learning based on image information taken about a damaged part of a vehicle from a plurality of user terminals or a repair shop terminal and actual repair cost estimate information therefor. According to paragraph 1,
After step (a) above,
Receiving repair estimate information for the vehicle from the user terminal or repair shop terminal; and
A method of providing a vehicle repair guide, further comprising determining whether the repair estimate information is an appropriate estimate based on previously learned data. According to paragraph 1,
In step (c),
A method of providing a vehicle repair guide, comprising receiving the user's insurance subscription information from an insurance company server based on personal information received from the user terminal, and then calculating the total cost required for insurance processing. An image receiving unit that receives a captured image of a damaged part of a vehicle from a user terminal or a repair shop terminal;
An estimate calculation unit that analyzes the image based on previously learned data to calculate an estimate range for vehicle repair; and
A vehicle repair guide providing system including an insurance processing determination unit that determines whether insurance processing is necessary by comparing the total cost required for insurance processing with the quotation range based on the calculated quotation range.