TWI684157B - A Smart Claims System Based on Action Vehicle - Google Patents

Abstract

The invention relates to an intelligent claims system based on a mobile vehicle, which includes a mobile terminal device, a cloud analysis server and an insurance claim server, wherein the mobile terminal device includes an image capture module, a letter processing module, and a character recognition module Group and mobile communication module; the cloud analysis server includes cloud communication unit, cloud storage unit and cloud processing unit; and the insurance claim server includes claim communication unit, claim storage unit and claim processing unit to generate claims payable on the insurance policy The result and the amount to be claimed.

Description

A Smart Claims System Based on Action Vehicle

The invention relates to a self-service intelligent insurance claim settlement system. The insured can automatically recognize the text content of the paper document through the mobile device, and at the same time identify its authenticity, and the cloud analysis server parses out the corresponding name of the injury and illness in the paper document. International disease classification code, and then through the operation of the insurance claim server and bank server to transfer the claim amount to the insured account to improve the efficiency of insurance claims and reduce the risk of human judgment errors in the claim review operation.

In the past, insurers who wanted to apply for claims from insurance companies must prepare relevant paper documents (such as claim application information and diagnosis certificates) and send them to the insurance company, who would manually verify the authenticity of the paper documents After that, we manually judge the international disease classification code corresponding to the name of the injury in the paper document, and then enter the relevant information to complete the insurance claim operation procedure, but these manual operations are time-consuming and laborious, and have the problem of inefficiency.

In addition, there are many types of insurance companies' products, various claims clauses and a large number of claims cases, it is difficult to guarantee that the aforementioned claims processing procedures that use traditional manual methods to identify, judge and enter relevant data, etc., will not cause errors, which will result in insurance Loss of households or insurance companies. With the evolution of technology, how to complete the insurance claims process through new technologies, taking into account both efficiency and reliability, is currently the goal the insurance industry is actively seeking.

In order to achieve the above goals, the present inventor is actively studying hard, with a view to solving the related problems of insurance claims, and after continuous efforts and research, the present invention was finally developed.

Therefore, the object of the present invention is to provide an intelligent claims system based on mobile vehicles. The system includes a mobile terminal device, a cloud analysis server and an insurance claim server. The mobile terminal device (such as a notebook computer or a smart phone) , Tablet or smart wearable device) includes an image capture module, a letter processing module, a character recognition module and a mobile communication module, where the image capture module is used to capture claim application data, diagnostic certificates and Receipts and other paper documents, and converted into image files; the letter processing module is connected to image capture The module is used to preprocess the background color of the image file and identify the authenticity of the paper document according to the image file; the character recognition module is connected to the letter processing module to identify the image captured from the paper document Files and generate recognized data, which can include text data and anti-counterfeit image data; and the mobile communication module is connected to the character recognition module to transmit the recognized data; and the cloud analysis server is connected to the mobile terminal Equipment for processing the identified data, wherein the cloud analysis server includes a cloud communication unit, a cloud storage unit, and a cloud processing unit, wherein the cloud communication unit is used to receive the identified data; and the cloud storage unit is connected to the cloud communication unit And pre-store the International Classification of Disease (ICD); and the cloud processing unit is connected to the cloud communication unit and the cloud storage unit, respectively, for parsing at least one injury and illness name in the paper document and the corresponding international disease classification code ; The insurance claim server is connected to the cloud analysis server to process the insurer’s claims demand for the policy. The insurance claim server includes a claim communication unit, a claim storage unit and a claim processing unit, in which the claim communication unit is used to receive And transmit data related to the claims demand of the policy; the claims storage unit is connected to the claims communication unit for storing data related to the claims demand of the policy; and the claims processing unit is connected to the claims communication unit and claims storage unit respectively for Generate the claimable claims result and claimable amount of the policy, and send the claimable amount to the bank server by the claim communication unit, then transfer the claimable amount to the account of the insurer, and return the claim result to the insurance claim server, The claim storage unit stores the claim result, and the claim communication unit transmits the claim result to the mobile terminal device.

According to another embodiment of the present invention, the paper document may be one or more of claim application information, diagnosis certificate and receipt.

According to another embodiment of the present invention, the cloud communication unit is further used to transmit data stored in the cloud storage unit.

According to another embodiment of the present invention, the character recognition module is an optical character recognition module.

According to another embodiment of the present invention, the identified data includes text data and anti-counterfeiting image data (print letter image data), where the text data may include the applicant's name, date, pathological description and name of the diagnosis, and various data of the policy Wait. Medical institutions often cover the text of the diagnosis certificate with the seal of the guard and/or the doctor’s seal, etc., causing some of the above-mentioned text information to be obscured by the printed letter. This phenomenon is not conducive to the character recognition of the diagnosis certificate. To solve this problem, The letter processing module can preprocess the background color of the image file before entering the character recognition module, especially when the text data is obscured by the letter and it is difficult to directly recognize the image file. Since most of the inks or inks used in general letter printing are of a single color (such as blue or red), the present invention can choose to be close to the original color of the letter in the image file (such as one of blue, red, and green). The background color is used to reduce the influence of the text data being obscured by the printed letter, and is used to improve the recognition range of the character recognition module for the text data of the image file.

In addition, in order to prevent interested persons from scanning or copying other documents with genuine security seals and/or physician seals, the image files of the security seals and/or physician seals are extracted and retouched or printed. Add or paste on the diagnosis certificate to be forged. Another embodiment of the present invention uses the edge detection algorithm in the letter processing module to define the detection edge of the letter image data as the basis for determining the authenticity of the diagnosis certificate. If the letter image data captured by scanning or copying is used, the letter image is likely to have problems of incomplete removal, residual color or jagged edges on the edges. The letter processing module of the present invention can use the edge detection algorithms such as Sobel operator, Roberts Cross operator, Prewitt operator, Canny operator, compass operator, Marr-Hildreth operator, Laplacian operator and wavelet transform to perform edge detection Detection, first define the detection edge of the letter image, and then set the identification boundary outside the detection edge of the letter image, and calculate the degree of variation of the color code value in the area between the detection edge and the identification boundary, and judge whether it is higher than A threshold value, used to identify the authenticity of image files with printed image data. If it is higher than the aforementioned threshold value, it means that the image data of the seal letter is most likely obtained by scanning, so it can be judged as a forger, otherwise it is judged as not a forger.

According to another embodiment of the present invention, the cloud processing unit of the cloud resolution server further includes an artificial intelligence sub-unit, which can improve the resolution of international diseases corresponding to the name of the injury through machine learning models and a large amount of training data in the cloud storage unit Classification code accuracy.

According to another embodiment of the present invention, the system performs the following steps when a user uses a mobile terminal device to submit a claim for a policy claim: In step S101, the user initiates a claim application process in the application program of the mobile terminal device; Step S102, the mobile terminal device captures the diagnosis certificate containing the letter image data through the image capture module, and converts it into an image file; step S103, the letter processing module first performs background color preprocessing on the image file, and then uses the edge Detect the identification seal image data, and generate anti-counterfeiting identification results; Step S104, If the anti-counterfeiting identification result is true, the character recognition module recognizes the image file and generates text data and letter image data; Step S105, the mobile communication module sends the text data and letter image data to the cloud of the cloud analysis server Communication unit; Step S106, the cloud communication unit receives text data and letter image data, and stores it in the cloud storage unit; Step S107, the cloud processing unit parses at least one injury name in the text data, and the international disease pre-stored by the cloud storage unit Find the corresponding international disease classification code in the classification code, and then store it in the cloud storage unit, and send it from the cloud communication unit to the claims communication unit of the insurance claims server; Step S108, the claims processing unit of the insurance claims server according to the policy and at least 1. The international disease classification code corresponding to the name of the injury and illness generates the claimable result and the claimable amount, and the claimable communication unit transmits the claimable amount to the bank server; in step S109, the bank server transfers the claimable amount to the insured Account and return the claim result to the insurance claim server, and the claim storage unit stores the claim result; step S110, the claim communication unit transmits the claim result to the mobile communication module of the mobile terminal device, so that the application of the mobile terminal device ends the claim Application Procedures.

10‧‧‧Intelligent claims system based on mobile vehicle

11‧‧‧Mobile terminal equipment

12‧‧‧ Cloud analysis server

13‧‧‧ Insurance Claim Server

14‧‧‧Bank server

111‧‧‧Image capture module

112‧‧‧ letter processing module

113‧‧‧ character recognition module

114‧‧‧Mobile communication module

121‧‧‧ cloud communication unit

122‧‧‧ cloud storage unit

123‧‧‧ cloud processing unit

131‧‧‧ claims communication unit

132‧‧‧ claims storage unit

133‧‧‧ claims processing unit

1231‧‧‧Artificial Intelligence Subunit

21‧‧‧Image file

22‧‧‧ letter image data

31‧‧‧ letter image data

311‧‧‧ Detect edge

32‧‧‧Forensic boundary

S101-S110‧‧‧Step

FIG. 1 is a block diagram showing a schematic structure of an intelligent claims system based on a mobile vehicle of the present invention.

FIG. 2(A) is a schematic diagram showing the image file of the present invention before background color preprocessing.

FIG. 2(B) is a schematic diagram showing the image file of the present invention after background color preprocessing.

FIG. 3 is a schematic diagram showing the authenticity of the printed letter image data of the present invention.

FIG. 4 is a flowchart showing an example of implementation of the smart claims process of the present invention.

The specific form of the present invention will be described with reference to the attached drawings.

As shown in FIG. 1, the present invention is an intelligent claims system based on mobile vehicles, which is applied to insurance companies to settle claims. The intelligent claims system 10 of this embodiment includes: a mobile terminal device 11 (such as a note) with an application program loaded on it Computer, smart phone, tablet computer or smart watch), which includes: an image capture module 111, which is used to capture a paper document to be converted into an image file, the image capture module 111 may be the A camera on the mobile terminal device 11; a letter processing module 112 connected to the image capture module 111 for preprocessing the background color of the image file and identifying the authenticity of the paper document based on the image file Pseudo; a character recognition module 113, which is connected to the letter processing module 112, for recognizing the image file, for example, optical character recognition (Optical Character Recognition, OCR) technology to recognize the text in the image file, Generate a recognized data; and a mobile communication module 114, which is connected to the character recognition module 113 for transmitting the recognized data, which can use GSM, 3G, 4G, Wi-Fi and other communication interfaces, but The category is not limited to this; a cloud analysis server 12, which is connected to the mobile terminal device 11, the cloud analysis server 12 is used to process the identified data, including: a cloud communication unit 121, which can use GSM, 3G , 4G, Wi-Fi, Ethernet and other communication interfaces (not limited to this) are used to receive the identified data, wherein the cloud analysis server 12 is connected to the mobile terminal device 11 through the cloud communication unit 121 The mobile communication module 114; a cloud storage unit 122, which is connected to the cloud communication unit 121, and pre-stores the international disease classification code, and the cloud communication unit 121 is used to transmit the data stored in the cloud storage unit 122; and a A cloud processing unit 123, which is connected to the cloud communication unit 121 and the cloud storage unit 122, respectively, for parsing at least one injury and illness name and the corresponding international disease classification code in the paper document; and an insurance claim server 13, which Connected to the cloud parsing server 12, it is used to process a policyholder's claims for a policy, including: a claims communication unit 131, which can use GSM, 3G, 4G, Wi-Fi, Ethernet and other communication interfaces (not (This is limited), used to receive and transmit data related to the claims demand of the policy, wherein the insurance claim server 13 is connected to the cloud communication unit 121 of the cloud analysis server 12 via the claim communication unit 131; a A claims storage unit 132 connected to the claims communication unit 131 for storing data related to the claims demand of the policy; and a claims processing unit 133 connected to the claims communication unit 131 and the claims storage unit 132 respectively for Generate a claimable result and a claimable amount regarding the policy, and the claim communication unit 131 transmits the claimable amount to a bank server 14, and the bank server 14 transfers the claimable amount to the policy The account of the user, and returns a claim result to the insurance claim server 13, the claim storage unit 132 stores the claim result, and the claim communication unit 131 transmits the claim result to the mobile terminal device 11.

In addition, the paper document may be one or more of the claim application information, diagnosis certificate and receipt.

In addition, the character recognition module 113 is an optical character recognition module, wherein the recognized The data includes a text data and an anti-counterfeit image data, the mobile terminal device 11 can use the optical character recognition module to identify the text data, wherein the text data can include the applicant's name, date, pathological description and name of diagnosis, Information about the policy, etc.

As shown in FIG. 2(A), the anti-counterfeit image data generated according to the image file 21 is a letter image data 22. Since the character recognition module 113 generates the recognized data, the text in the image file 21 may be obscured by the printed image data 22, which is not conducive to the operation of the character recognition module 113. To solve this problem, the letter processing module 112 may perform background color pre-processing before the image file 20 enters the character recognition module 113. As shown in FIG. 2(B), the background color pre-processing selects the primary color close to the letter image data 22 in the image file 21 (as shown in red in FIG. 2(B)), and sets the background color of the image file 21 , To dilute the effect of the text in the image file 21 being obscured by the seal image data 22.

The letter processing module 112 not only performs background color preprocessing on the image file 21, but also recognizes the letter image data 22 through edge detection, wherein the letter processing module 112 can use the Sobel operator, Roberts Cross operator, Edge detection algorithms such as Prewitt operator, Canny operator, Compass operator, Marr-Hildreth operator, Laplacian operator and wavelet transform are used for edge detection to identify the image file 21 with the seal image data 22 Authenticity. As shown in FIG. 3, the aforementioned edge detection algorithm is used to define a detection edge 311 of the printed image data 31, and then set an identification boundary 32 outside the detection edge 311, and from the detection edge 311 to the identification The degree of variation of the color code value is calculated between the boundaries 32, and it is judged whether it is higher than a threshold value. If it is higher than the threshold value, it means that the detection edge 311 still has residual color or jagged patterns and other problems. There is a very high possibility that it is obtained by scanning, so it can be determined that the printed image data 31 is a counterfeiter , Otherwise it is determined that it is not a forger.

In addition, the cloud processing unit 123 of the cloud parsing server 12 further includes an artificial intelligence sub-unit 1231 which can be machine-learned to parse the international disease classification code corresponding to the at least one injury name to improve the accuracy of the analysis.

Referring to FIG. 4, the following describes an example flow of implementation of the usage mode of the present invention.

As shown in FIG. 4, when the user of the system 10 uses the mobile terminal device 11 to submit a claim for insurance policy, the intelligent claim system 10 performs the following steps: Step S101, the user selects one of the mobile terminal devices 11 Start a claims application process in the application; steps S102, the mobile terminal device 11 retrieves the diagnosis certificate containing the seal letter image data through the image capture module 111, and converts it to the image file; step S103, the seal letter processing module 112 first targets the image file Perform background color preprocessing, and then use edge detection to identify the printed image data and generate an anti-counterfeiting identification result; step S104, if the anti-counterfeiting identification result is true, the character recognition module 113 recognizes the image file, and Generate the text data and the printed image data; Step S105, the mobile communication module 114 transmits the text data and the printed image data to the cloud communication unit 121 of the cloud parsing server 12; Step S106, the cloud communication The unit 121 receives the text data and the printed image data, and stores them in the cloud storage unit 122; Step S107, the cloud processing unit 123 parses the at least one injury name in the text data, and the cloud storage unit 122 pre-stores The corresponding international disease classification code is found in the international disease classification code, and then stored in the cloud storage unit 123, and transmitted from the cloud communication unit 121 to the claim communication unit 131 of the insurance claim server 13; step S108, the The claim processing unit 133 of the insurance claim server 13 generates a claimable result and a claimable amount according to the insurance policy and the international disease classification code corresponding to the at least one injury and illness name, and the claim communication unit 131 transmits the claimable claim Amount to a bank server 14; Step S109, the bank server 14 transfers the claim amount to the account of the insured, and returns a claim result to the insurance claim server 13 (if the account of the insured If it is correct, it is a message of remittance success; otherwise, it is a message of remittance failure), the claim storage unit 132 stores the claim result; step S110, the claim communication unit 131 transmits the claim result to the mobile terminal device 11 The mobile communication module 114 causes the application program of the mobile terminal device 11 to end the claim application process.

The above are only preferred specific embodiments of the present invention, which are not intended to limit the present invention. Any modification, equivalent disassembly, improvement, etc. made within the spirit and principle of the present invention should be included in this Within the scope of protection of the invention.

10‧‧‧Intelligent claims system based on mobile vehicle

11‧‧‧Mobile terminal equipment

12‧‧‧ Cloud analysis server

13‧‧‧ Insurance Claim Server

14‧‧‧Bank server

111‧‧‧Image capture module

112‧‧‧ letter processing module

113‧‧‧ character recognition module

114‧‧‧Mobile communication module

121‧‧‧ cloud communication unit

122‧‧‧ cloud storage unit

123‧‧‧ cloud processing unit

131‧‧‧ claims communication unit

132‧‧‧ claims storage unit

133‧‧‧ claims processing unit

1231‧‧‧Artificial Intelligence Subunit

Claims (6)

An intelligent claims system based on mobile vehicles, including: a mobile terminal device, which includes: an image capture module, which is used to capture a paper document to be converted into an image file; a letter processing module, It is connected to an image capture module for preprocessing the background color of the image file and identifying the authenticity of the paper document based on the image file; a character recognition module, which is connected to the letter processing module, is used to To recognize the image file and generate a recognized data, wherein the recognized data includes a text data and a letter image data, and the letter processing module uses edge detection to define one of the letter image data to detect the edge, and An identification boundary is set outside the detection edge, and a color code value variation of the printed image data is calculated in an area between the detection edge and the identification boundary; and a mobile communication module connected to the character identification A module for transmitting the identified data; a cloud analysis server connected to the mobile terminal device for processing the identified data, which includes: a cloud communication unit for receiving the identified data Data; a cloud storage unit, which is connected to the cloud communication unit and pre-stores the international disease classification code; and a cloud processing unit, which is connected to the cloud communication unit and the cloud storage unit, respectively, to analyze the paper document At least one injury and illness name and corresponding international disease classification code; and an insurance claims server, which is connected to the cloud analysis server and is used to process a policyholder's claims for a policy, including: a claims communication unit, which uses To receive and transmit data related to the claim demand of the policy; a claim storage unit connected to the claim communication unit for storing data related to the claim demand of the policy; and A claim processing unit, which is connected to the claim communication unit and the claim storage unit, respectively, to generate a claimable result and a claimable amount of the insurance policy. For example, in the system of claim 1, the paper document may be a claim application, a diagnosis certificate or a receipt. For example, in the system of claim 1, the cloud communication unit is further used to transmit the data stored in the cloud storage unit. For example, the system of claim 2 or 3, wherein the character recognition module is an optical character recognition module. For example, in the system of claim 4, the cloud processing unit of the cloud parsing server further includes an artificial intelligence sub-unit, which can learn and parse the international disease classification code corresponding to the at least one injury name by machine learning. For example, in the system of claim 5, if a user uses the mobile terminal device to make claims for the policy, the system performs the following steps: Step S101, the user is in an application of the mobile terminal device Start a claim application process; step S102, the mobile terminal device retrieves the diagnosis certificate containing the seal letter image data through the image capture module, and converts it into the image file; step S103, the seal letter processing module first Perform background color preprocessing on the image file, and then use edge detection to define the detected edge of the printed image data, calculate the variation of the color code value of the printed image data, and determine whether it is higher than a threshold, To generate an anti-counterfeiting identification result; step S104, if the variation of the color code value is lower than the threshold value, the anti-counterfeiting identification result is true, the character recognition module recognizes the image file, and generates the text resource Data and the printed image data; step S105, the mobile communication module transmits the text data and the printed image data to the cloud communication unit of the cloud parsing server; step S106, the cloud communication unit receives the text data and the Print letter image data and store it in the cloud storage unit; Step S107, the cloud processing unit parses the at least one injury name in the text data, and finds the corresponding international in the international disease classification code pre-stored by the cloud storage unit The disease classification code is then stored in the cloud storage unit, and transmitted from the cloud communication unit to the claims communication unit of the insurance claims server; step S108, the claims processing unit of the insurance claims server according to the policy and the at least An international disease classification code corresponding to the name of an injury or illness generates a claimable result and a claimable amount, and the claimant communication unit transmits the claimable amount to a bank server; in step S109, the bank server sends the claimable claim The amount is transferred to the account of the insurer, and a claim result is returned to the insurance claim server, and the claim storage unit stores the claim result; in step S110, the claim communication unit transmits the claim result to the mobile terminal device The mobile communication module enables the application program of the mobile terminal device to end the claim application process.

Images