A method for intelligently issuing motorcycle insurance policies online

Abstract

The present invention discloses an online intelligent order processing method for motorcycle insurance, which relates to the field of image recognition technology. The method includes receiving multiple insurance images; extracting certificate images from multiple insurance images; determining target pixel points in each certificate image; sorting the target pixel points to obtain a target pixel point sequence of each certificate image; extracting the integral image values ​​corresponding to all target pixel points in the target pixel point sequence of each certificate image to obtain an integral image value sequence corresponding to each certificate image; determining the certificate type corresponding to each certificate image; determining the key information area of ​​each certificate image and the information type corresponding to the key information area based on the certificate type; identifying text information in the key information area of ​​each certificate image; generating a motorcycle insurance order based on the text information and the information type corresponding to the key information area. The method disclosed by the present invention can reduce the operating costs of insurance companies and improve the efficiency of issuing insurance orders.

Description

Technical Field

[0001] The present invention belongs to the technical field of image recognition, and in particular relates to an online intelligent motorcycle insurance policy processing method. Background Art

[0002] With the development of Internet technology, motorcycle insurance business has also shifted from the traditional offline paper-based issuance process to online issuance. When purchasing motorcycle insurance, car owners can upload insurance images containing images of various documents such as the owner's ID card, vehicle registration certificate, and vehicle license plate number. The reviewers will then conduct a manual review and enter some key information in the documents, such as the owner's name, ID number, and vehicle license plate number, into the system to generate a motorcycle insurance order.

[0003] However, this method requires manual entry of order information, a process that consumes a lot of manpower costs, and the speed of issuing insurance orders is slow, which increases the operating costs of the insurance company and reduces the customer experience.

[0004] Therefore, how to provide an effective solution to improve the efficiency of issuing insurance orders and reduce operating costs has become a difficult problem that needs to be solved urgently in the existing technology. Summary of the invention

[0005] The purpose of the present invention is to provide an online intelligent motorcycle insurance policy processing method to solve the above-mentioned problems existing in the prior art.

[0006] In order to achieve the above object, the present invention adopts the following technical solutions:

[0007] In a first aspect, the present invention provides a method for intelligently issuing motorcycle insurance policies online, comprising:

[0008] Receiving multiple insurance images uploaded by a motorcycle insurance owner, each insurance image including a document image;

[0009] Extracting the document image from the multiple insurance images through edge contour recognition;

[0010] Determine a target pixel point in each document image, wherein the target pixel point in any document image is a pixel point in any document image located on a specified diagonal line of any document image;

[0011] Based on the pixel coordinates, the target pixel points in each certificate image are sorted to obtain a target pixel point sequence of each certificate image;

[0012] Extracting integral image values ​​corresponding to all target pixel points in the target pixel point sequence of each certificate image to obtain an integral image value sequence corresponding to each certificate image;

[0013] Determine the document type corresponding to each document image based on the integral image value sequence corresponding to each document image;

[0014] Based on the certificate type corresponding to each certificate image, determine the key information area of ​​each certificate image and the information type corresponding to the key information area;

[0015] Through optical character recognition, the text information in the key information area of ​​each certificate image is identified;

[0016] A motorcycle insurance order is generated based on the text information in the key information area of ​​each certificate image and the information type corresponding to the key information area.

[0017] Based on the above disclosed content, the present invention receives multiple insured images uploaded by a car owner who purchases motorcycle insurance, each insured image includes an ID image; extracts the ID images from the multiple insured images through edge contour recognition; determines the target pixel points in each ID image, wherein the target pixel points in any ID image are the pixel points located on the specified diagonal line of any ID image; sorts the target pixel points in each ID image based on pixel coordinates to obtain a target pixel point sequence of each ID image; extracts the integral image values ​​corresponding to all target pixel points in the target pixel point sequence of each ID image to obtain an integral image value sequence corresponding to each ID image; determines the ID type corresponding to each ID image based on the integral image value sequence corresponding to each ID image; determines the key information area of ​​each ID image and the information type corresponding to the key information area based on the ID type corresponding to each ID image; identifies the text information in the key information area of ​​each ID image through optical character recognition; generates a motorcycle insurance order based on the text information in the key information area of ​​each ID image and the information type corresponding to the key information area. In this way, the document types corresponding to the various document images in the underwriting image can be automatically identified, and the text information in the various document images can be automatically extracted and the information type corresponding to the identified text information can be automatically generated, thereby automatically generating motorcycle insurance orders without manual entry, reducing the operating costs of insurance companies, improving the efficiency of issuing insurance orders, and enhancing the customer experience.

[0018] In a possible design, the designated diagonal line of any document image is a diagonal line from the upper left corner to the lower right corner of any document image;

[0019] The method of sorting target pixel points in each document image based on pixel coordinates includes:

[0020] Based on the pixel coordinates, the target pixels in each document image are sorted in the order that the target pixel with a smaller pixel horizontal coordinate precedes the target pixel with a larger pixel horizontal coordinate, and among two target pixels with the same pixel horizontal coordinate, the target pixel with a smaller pixel vertical coordinate precedes the target pixel with a larger pixel vertical coordinate.

[0021] In a possible design, extracting integral image values ​​corresponding to all target pixel points in a target pixel point sequence of each document image to obtain an integral image value sequence corresponding to each document image includes:

[0022] Extracting the integral image values ​​corresponding to all target pixel points in the target pixel point sequence of each document image;

[0023] Based on the sorting order of the target pixel points in the target pixel point sequence, the integral image values ​​corresponding to all the target pixel points of each document image are combined to obtain the integral image value sequence corresponding to each document image;

[0024] Among them, the integral image value corresponding to the target pixel point in the document image is , Indicates the integral image value corresponding to the target pixel point with pixel coordinates (I, J) in the document image. Represents the grayscale value or binary value corresponding to the pixel point with pixel coordinates (i, j) in the document image.

[0025] In a possible design, determining the certificate type corresponding to each certificate image based on the integral image value sequence corresponding to each certificate image includes:

[0026] Based on the integral image value sequence corresponding to each certificate image, determining the integral image value distribution curve corresponding to each certificate image;

[0027] Performing similarity calculation on the integral image value distribution curve corresponding to each certificate image and the integral image value distribution curve corresponding to a sample certificate image of a known certificate type, to determine the certificate type corresponding to each certificate image; or

[0028] The integral image value sequence corresponding to each certificate image is used as the input of the pre-trained certificate type recognition model to perform calculations to obtain the certificate type corresponding to each certificate image;

[0029] The certificate type recognition model is obtained by training with the integral image value sequence corresponding to the sample certificate images of known certificate types as sample input and the confidence of the sample certificate images of various certificate types as output.

[0030] In a possible design, before determining the target pixel points in each document image, the method further includes:

[0031] Each document image is converted to the specified size after adjusting the angle.

[0032] In a possible design, before extracting the integral image values ​​corresponding to all target pixels in the target pixel sequence of each document image, the method further includes:

[0033] Convert each document image into a grayscale image or a binary image.

[0034] In one possible design, the document images in the multiple insurance images include an identity card image, a business license image, a driver's license image, a vehicle registration certificate image and / or a vehicle certificate image.

[0035] In a second aspect, the present invention provides an online intelligent motorcycle insurance policy processing device, comprising:

[0036] A receiving unit, used for receiving a plurality of insurance images uploaded by a motorcycle insurance owner, each of which includes a certificate image;

[0037] A first extraction unit, configured to extract a document image from the plurality of insurance images by edge contour recognition;

[0038] A first determining unit is used to determine a target pixel point in each certificate image, wherein the target pixel point in any certificate image is a pixel point in any certificate image located on a specified diagonal line of the any certificate image;

[0039] A sorting unit, used to sort the target pixel points in each document image based on the pixel coordinates to obtain a target pixel point sequence of each document image;

[0040] The second extraction unit is used to extract the integral image values ​​corresponding to all target pixel points in the target pixel point sequence of each certificate image, and obtain the integral image value sequence corresponding to each certificate image;

[0041] A second determination unit, configured to determine the certificate type corresponding to each certificate image based on the integral image value sequence corresponding to each certificate image;

[0042] A third determining unit, configured to determine the key information area of ​​each certificate image and the information type corresponding to the key information area based on the certificate type corresponding to each certificate image;

[0043] A recognition unit, used to recognize text information in a key information area of ​​each certificate image through optical character recognition;

[0044] A generating unit is used to generate a motorcycle insurance low order based on the text information in the key information area of ​​each certificate image and the information type corresponding to the key information area.

[0045] In a third aspect, the present invention provides an electronic device comprising a memory, a processor and a transceiver which are communicatively connected in sequence, wherein the memory is used to store a computer program, the transceiver is used to send and receive messages, and the processor is used to read the computer program to execute the method for intelligent online issuance of motorcycle insurance policies as described in the first aspect or any possible design of the first aspect.

[0046] In a fourth aspect, the present invention provides a computer-readable storage medium having instructions stored thereon, which, when executed on a computer, executes the method for intelligent online issuance of motorcycle insurance policies as described in the first aspect or any possible design of the first aspect.

[0047] In a fifth aspect, the present invention provides a computer program product comprising instructions, which, when executed on a computer, causes the computer to execute the method for intelligent online processing of motorcycle insurance policies as described in the first aspect or any possible design of the first aspect.

[0048] Beneficial effects:

[0049] The present invention can automatically identify the document types corresponding to various document images in the insurance image, and can automatically extract text information in various document images and identify the information type corresponding to the text information, thereby automatically generating a motorcycle insurance order without manual entry, thereby reducing the operating costs of the insurance company, improving the efficiency of issuing insurance orders, enhancing the customer experience, and facilitating practical application and promotion. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 A flowchart of a method for intelligently issuing motorcycle insurance policies online provided in an embodiment of the present application;

[0051] Figure 2 A schematic block diagram of an online intelligent motorcycle insurance policy processing device provided in an embodiment of the present application;

[0052] Figure 3 A schematic block diagram of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0053] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the present invention will be briefly introduced below in combination with the drawings and the description of the embodiments or the prior art. Obviously, the following description of the structure of the drawings is only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. It should be noted that the description of these embodiments is used to help understand the present invention, but does not constitute a limitation of the present invention.

[0054] It should be understood that although the terms first, second, etc. may be used herein to describe various units, these units should not be limited by these terms. These terms are only used to distinguish one unit from another unit. For example, a first unit can be referred to as a second unit, and similarly, a second unit can be referred to as a first unit without departing from the scope of the exemplary embodiments of the present invention.

[0055] It should be understood that the term "and / or" that may appear in this article is merely a description of the association relationship between associated objects, indicating that three relationships may exist. For example, A and / or B can represent three situations: A exists alone, B exists alone, and A and B exist at the same time. The term " / and" that may appear in this article describes another type of association object relationship, indicating that two relationships may exist. For example, A / and B can represent two situations: A exists alone, and A and B exist alone. In addition, the character " / " that may appear in this article generally indicates that the previous and next associated objects are in an "or" relationship.

[0056] In order to improve the efficiency of issuing insurance orders and reduce operating costs, an embodiment of the present application provides an online intelligent policy processing method for motorcycle insurance, which can reduce the operating costs of insurance companies and improve the efficiency of issuing insurance orders.

[0057] The motorcycle insurance online intelligent bill processing method provided in the embodiment of the present application can be applied to the backend server of the insurance company. It can be understood that the execution subject does not constitute a limitation on the embodiment of the present application.

[0058] The following is a detailed description of the online intelligent motorcycle insurance policy processing method provided in the embodiment of the present application.

[0059] like Figure 1 As shown, it is a flowchart of the method for intelligent online policy issuance processing of motorcycle insurance provided in the first aspect of the embodiment of the present application. The method for intelligent online policy issuance processing of motorcycle insurance may include but is not limited to the following steps S101-S109.

[0060] Step S101: receiving a plurality of insurance images uploaded by a motorcycle insurance owner.

[0061] In the embodiment of the present application, when the owner purchases motorcycle insurance online, he can manually upload multiple insurance images, each of which includes a document image. The document images in the multiple insurance images may include, but are not limited to, ID card images, business license images, driver's license images, vehicle registration images, and / or vehicle certificate images.

[0062] Step S102: extracting document images from multiple insurance images through edge contour recognition.

[0063] In the embodiment of the present application, edge detection can be performed on the certificate images in the multiple insurance images by using, but not limited to, the Canny or Sobel algorithm, etc., to extract the certificate images in the multiple insurance images, which will not be elaborated here.

[0064] Step S103: Determine the target pixel points in each document image.

[0065] The target pixel point in any document image is a pixel point located on a specified diagonal line of any document image, for example, a diagonal line from the upper left corner to the lower right corner of the document image, or a diagonal line from the lower left corner to the upper right corner of the document image.

[0066] In one or more embodiments, before determining the target pixel points in each certificate image, the angle of each certificate image may be adjusted so that the rotation angle of each certificate image is 0 degrees, and then the rotated certificate images are converted to the specified size to make the sizes of each certificate image consistent, so as to avoid affecting the accuracy of subsequent recognition due to inconsistent angles and sizes.

[0067] Step S104: based on the pixel coordinates, sort the target pixel points in each certificate image to obtain a target pixel point sequence of each certificate image.

[0068] In the embodiment of the present application, the designated diagonal line of any document image is the diagonal line from the upper left corner to the lower right corner of any document image. At this time, the target pixels in each document image are sorted, and the target pixels in each document image can be sorted based on the pixel coordinates, in the order that the target pixel with a smaller pixel horizontal coordinate precedes the target pixel with a larger pixel horizontal coordinate, and in the order that the target pixel with a smaller pixel vertical coordinate of two target pixels with the same pixel horizontal coordinate precedes the target pixel with a larger pixel vertical coordinate.

[0069] It can be understood that if the designated diagonal line of any ID image is the diagonal line from the lower left corner to the upper right corner of any ID image, then when sorting the target pixel points in each ID image, the target pixel points in each ID image can be sorted based on the pixel coordinates, in the order that the target pixel point with a larger pixel horizontal coordinate precedes the target pixel point with a smaller pixel horizontal coordinate, and among two target pixel points with the same pixel horizontal coordinate, the target pixel point with a larger pixel vertical coordinate precedes the target pixel point with a smaller pixel vertical coordinate.

[0070] Step S105: extracting integral image values ​​corresponding to all target pixel points in the target pixel point sequence of each certificate image, and obtaining an integral image value sequence corresponding to each certificate image.

[0071] In an embodiment of the present application, the integral image values ​​corresponding to all target pixel points in the target pixel point sequence of each certificate image can be first extracted, and then based on the sorting order of the target pixel points in the target pixel point sequence, the integral image values ​​corresponding to all target pixel points of each certificate image can be combined in sequence to obtain the integral image value sequence corresponding to each certificate image.

[0072] The integral image value corresponding to the target pixel in the document image can be expressed as ,in Indicates the integral image value corresponding to the target pixel point with pixel coordinates (I, J) in the document image. Represents the grayscale value or binary value corresponding to the pixel point with pixel coordinates (i, j) in the document image.

[0073] In order to avoid the influence of light on the integral image value corresponding to the pixel point, The binary value corresponding to the pixel point with pixel coordinates (i, j) in the certificate image can be taken. It can be understood that if the integral image value corresponding to the target pixel point is calculated based on the binary value corresponding to the pixel point, then each certificate image can be converted into a binary image before extracting the integral image values ​​corresponding to all target pixels in the target pixel point sequence of each certificate image. If the integral image value corresponding to the target pixel point is calculated based on the grayscale value corresponding to the pixel point, then each certificate image can be converted into a grayscale image before extracting the integral image values ​​corresponding to all target pixels in the target pixel point sequence of each certificate image.

[0074] Step S106: Based on the integral image value sequence corresponding to each certificate image, determine the certificate type corresponding to each certificate image.

[0075] Since different types of certificate images have obvious differences in layout and content on the certificates, the integral image value sequences corresponding to different certificate images will also be obviously different. Therefore, in one or more embodiments, the integral image value sequences corresponding to sample certificate images of various known certificate types can be determined in the aforementioned manner, and the integral image value distribution curves corresponding to the sample certificate images of various known certificate types can be determined based on this. When determining the certificate type corresponding to each certificate image, the integral image value distribution curve corresponding to each certificate image can be determined based on the integral image value sequence corresponding to each certificate image (that is, the corresponding integral image value distribution curve is drawn according to the integral image value sequence corresponding to each certificate image), and then the integral image value distribution curve corresponding to each certificate image is calculated for similarity with the integral image value distribution curve corresponding to the sample certificate image of the known certificate type, so as to determine the certificate type corresponding to each certificate image.

[0076] In one or more embodiments, a document type recognition model for document type recognition may also be pre-trained. When determining the document type corresponding to each document image, the integral image value sequence corresponding to each document image may be used as the input of the pre-trained document type recognition model for operation to obtain the document type corresponding to each document image. The document type recognition model may be trained by using the integral image value sequence corresponding to a large number of sample document images of known document types as sample inputs and the confidence of the sample document images being various document types as outputs.

[0077] At the same time, the above-mentioned recognition method can easily and quickly identify the certificate type corresponding to each certificate image with relatively low computing resources, thereby improving the efficiency of issuing motorcycle insurance orders.

[0078] Step S107: Based on the certificate type corresponding to each certificate image, determine the key information area of ​​each certificate image and the information type corresponding to the key information area.

[0079] Since the layout of ID cards, business licenses, driver's licenses, vehicle registration certificates, and vehicle certificates are fixed, the positions of key information such as name, address, ID number, and license plate number on the certificates are also fixed. Therefore, the key information area of ​​each certificate image and the information type corresponding to the key information area can be directly determined according to the certificate type corresponding to each certificate image. Among them, key information refers to some information that needs to be entered when generating a motorcycle insurance order, such as the owner's name, address, ID number, vehicle model, and license plate number, etc.

[0080] Step S108: Identify text information in the key information area of ​​each certificate image through optical character recognition.

[0081] After determining the key information area of ​​each certificate image and the information type corresponding to the key information area, the text information in the key information area of ​​each certificate image can be directly recognized by optical character recognition (OCR), so that various information contents of the car owner can be recognized. The principle of optical character recognition will not be described in detail in the embodiments of this application.

[0082] Step S109: Generate a motorcycle insurance order based on the text information in the key information area of ​​each certificate image and the information type corresponding to the key information area.

[0083] In an embodiment of the present application, after identifying the text information in the key information area of ​​each certificate image, the text information can be added to the location where the corresponding information is recorded in the insurance order template based on the text information in the key information area of ​​each certificate image and the information type corresponding to the key information area to generate a motorcycle insurance order.

[0084] The motorcycle insurance online intelligent policy processing method provided by the present invention receives multiple insurance images uploaded by a motorcycle insurance owner, each of which includes an ID image; extracts the ID image from the multiple insurance images through edge contour recognition; determines the target pixel points in each ID image, wherein the target pixel point in any ID image is a pixel point in any ID image that is located on a specified diagonal line of the ID image; sorts the target pixel points in each ID image based on pixel coordinates to obtain a target pixel point sequence for each ID image; extracts the target pixel points of each ID image In the point sequence, the integral image values ​​corresponding to all target pixel points are used to obtain the integral image value sequence corresponding to each certificate image; based on the integral image value sequence corresponding to each certificate image, the certificate type corresponding to each certificate image is determined; based on the certificate type corresponding to each certificate image, the key information area of ​​each certificate image and the information type corresponding to the key information area are determined; through optical character recognition, the text information in the key information area of ​​each certificate image is identified; based on the text information in the key information area of ​​each certificate image and the information type corresponding to the key information area, a motorcycle insurance order is generated. In this way, the certificate types corresponding to various certificate images in the insured image can be automatically identified, and the text information in various certificate images can be automatically extracted and the information type corresponding to the identified text information can be automatically generated, thereby automatically generating a motorcycle insurance order without manual entry, reducing the operating costs of insurance companies, improving the efficiency of issuing insurance orders, and enhancing the customer experience, which is convenient for practical application and promotion.

[0085] See also Figure 2 The second aspect of the embodiment of the present application provides an online intelligent motorcycle insurance policy processing device, the online intelligent motorcycle insurance policy processing device comprising:

[0086] A receiving unit, used for receiving a plurality of insurance images uploaded by a motorcycle insurance owner, each of which includes a certificate image;

[0087] A first extraction unit, configured to extract a document image from the plurality of insurance images by edge contour recognition;

[0088] A first determining unit is used to determine a target pixel point in each certificate image, wherein the target pixel point in any certificate image is a pixel point in any certificate image located on a specified diagonal line of the any certificate image;

[0089] A sorting unit, used to sort the target pixel points in each document image based on the pixel coordinates to obtain a target pixel point sequence of each document image;

[0090] The second extraction unit is used to extract the integral image values ​​corresponding to all target pixel points in the target pixel point sequence of each certificate image, and obtain the integral image value sequence corresponding to each certificate image;

[0091] A second determination unit, configured to determine the certificate type corresponding to each certificate image based on the integral image value sequence corresponding to each certificate image;

[0092] A third determining unit, configured to determine the key information area of ​​each certificate image and the information type corresponding to the key information area based on the certificate type corresponding to each certificate image;

[0093] A recognition unit, used to recognize text information in a key information area of ​​each certificate image through optical character recognition;

[0094] A generating unit is used to generate a motorcycle insurance low order based on the text information in the key information area of ​​each certificate image and the information type corresponding to the key information area.

[0095] The working process, working details and technical effects of the online intelligent processing device for motorcycle insurance provided in the second aspect of this embodiment can be found in the first aspect of the embodiment and will not be repeated here.

[0096] like Figure 3 As shown, the third aspect of an embodiment of the present application provides an electronic device, comprising a memory, a processor and a transceiver which are communicatively connected in sequence, wherein the memory is used to store computer programs, the transceiver is used to send and receive messages, and the processor is used to read the computer program to execute the online intelligent order processing method for motorcycle insurance as described in the first aspect of the embodiment.

[0097] For specific examples, the memory may include but is not limited to random access memory (RAM), read-only memory (ROM), flash memory, first-in-first-out memory (FIFO) and / or first-in-last-out memory (FILO), etc.; the processor may be but is not limited to a microprocessor of the STM32F105 series, an ARM (Advanced RISC-Machines), an X86 or other architecture processor, or a processor with an integrated NPU (neural-network processing units); the transceiver may be but is not limited to a WiFi (Wireless Fidelity) wireless transceiver, a Bluetooth wireless transceiver, a General Packet Radio Service (GPRS) wireless transceiver, a ZigBee protocol (a low-power local area network protocol based on the IEEE802.15.4 standard, ZigBee) wireless transceiver, a 3G transceiver, a 4G transceiver and / or a 5G transceiver, etc.

[0098] The fourth aspect of this embodiment provides a computer-readable storage medium storing instructions for the method for intelligently issuing motorcycle insurance online as described in the first aspect of the embodiment, that is, the computer-readable storage medium stores instructions, and when the instructions are run on a computer, the method for intelligently issuing motorcycle insurance online as described in the first aspect is executed. The computer-readable storage medium refers to a carrier for storing data, which may include but is not limited to a floppy disk, an optical disk, a hard disk, a flash memory, a USB flash drive, and / or a memory stick, etc., and the computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices.

[0099] The fifth aspect of this embodiment provides a computer program product containing instructions, which, when executed on a computer, enables the computer to execute the method for intelligent online motorcycle insurance policy processing as described in the first aspect of the embodiment, wherein the computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device.

[0100] It should be understood that certain details are provided in the following description to facilitate a complete understanding of the example embodiments. However, it should be understood by those of ordinary skill in the art that the example embodiments can be implemented without these certain details. For example, the system can be shown in a block diagram to avoid obscuring the example with unnecessary details. In other examples, well-known processes, structures, and techniques may not be shown in unnecessary detail to avoid obscuring the example embodiments.

[0101] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the protection scope of the present invention. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for intelligent online motorcycle insurance policy issuance, characterized in that: include: Receiving multiple insurance images uploaded by a motorcycle insurance owner, each insurance image including a document image; Extracting the document image from the multiple insurance images through edge contour recognition; Determine a target pixel point in each document image, wherein the target pixel point in any document image is a pixel point in any document image located on a specified diagonal line of any document image; Based on the pixel coordinates, the target pixel points in each certificate image are sorted to obtain a target pixel point sequence of each certificate image; Extracting integral image values ​​corresponding to all target pixel points in the target pixel point sequence of each certificate image to obtain an integral image value sequence corresponding to each certificate image; Determine the document type corresponding to each document image based on the integral image value sequence corresponding to each document image; Based on the certificate type corresponding to each certificate image, determine the key information area of ​​each certificate image and the information type corresponding to the key information area; Through optical character recognition, the text information in the key information area of ​​each certificate image is identified; A motorcycle insurance order is generated based on the text information in the key information area of ​​each certificate image and the information type corresponding to the key information area.

2. The method for intelligently issuing motorcycle insurance policies online according to claim 1, characterized in that: The specified diagonal line of any certificate image is the diagonal line from the upper left corner to the lower right corner of any certificate image; The method of sorting target pixel points in each document image based on pixel coordinates includes: Based on the pixel coordinates, the target pixels in each document image are sorted in the order that the target pixel with a smaller pixel horizontal coordinate precedes the target pixel with a larger pixel horizontal coordinate, and among two target pixels with the same pixel horizontal coordinate, the target pixel with a smaller pixel vertical coordinate precedes the target pixel with a larger pixel vertical coordinate.

3. The method for intelligently issuing motorcycle insurance policies online according to claim 1, characterized in that: The step of extracting integral image values ​​corresponding to all target pixel points in the target pixel point sequence of each document image to obtain an integral image value sequence corresponding to each document image includes: Extracting the integral image values ​​corresponding to all target pixel points in the target pixel point sequence of each document image; Based on the sorting order of the target pixel points in the target pixel point sequence, the integral image values ​​corresponding to all the target pixel points of each document image are combined to obtain the integral image value sequence corresponding to each document image; Among them, the integral image value corresponding to the target pixel point in the document image is , Indicates the integral image value corresponding to the target pixel point with pixel coordinates (I, J) in the document image. Represents the grayscale value or binary value corresponding to the pixel point with pixel coordinates (i, j) in the document image.

4. The method for intelligently issuing motorcycle insurance policies online according to claim 1, characterized in that: The step of determining the certificate type corresponding to each certificate image based on the integral image value sequence corresponding to each certificate image includes: Based on the integral image value sequence corresponding to each certificate image, determining the integral image value distribution curve corresponding to each certificate image; Performing similarity calculation on the integral image value distribution curve corresponding to each certificate image and the integral image value distribution curve corresponding to a sample certificate image of a known certificate type, to determine the certificate type corresponding to each certificate image; or The integral image value sequence corresponding to each certificate image is used as the input of the pre-trained certificate type recognition model to perform calculations to obtain the certificate type corresponding to each certificate image; The certificate type recognition model is obtained by training with the integral image value sequence corresponding to the sample certificate images of known certificate types as sample input and the confidence of the sample certificate images of various certificate types as output.

5. The method for intelligently issuing motorcycle insurance policies online according to claim 1, characterized in that: Before determining the target pixel points in each document image, the method further includes: Each document image is converted to the specified size after adjusting the angle.

6. The method for intelligently issuing motorcycle insurance policies online according to claim 1, characterized in that: Before extracting the integral image values ​​corresponding to all target pixel points in the target pixel point sequence of each document image, the method further includes: Convert each document image into a grayscale image or a binary image.

7. The method for intelligently issuing motorcycle insurance policies online according to claim 1, characterized in that: The document images in the multiple insurance images include an identity card image, a business license image, a driver's license image, a vehicle registration certificate image and / or a vehicle certificate image.

8. An online intelligent motorcycle insurance policy processing device, characterized in that: include: A receiving unit, used to receive a plurality of insurance images uploaded by a motorcycle owner who purchases motorcycle insurance, each insurance image including a certificate image; A first extraction unit, configured to extract a document image from the plurality of insurance images by edge contour recognition; A first determining unit is used to determine a target pixel point in each document image, wherein the target pixel point in any document image is a pixel point in any document image located on a specified diagonal line of the any document image; A sorting unit, used to sort the target pixel points in each document image based on the pixel coordinates to obtain a target pixel point sequence of each document image; The second extraction unit is used to extract the integral image values ​​corresponding to all target pixel points in the target pixel point sequence of each certificate image, and obtain the integral image value sequence corresponding to each certificate image; A second determination unit, configured to determine the certificate type corresponding to each certificate image based on the integral image value sequence corresponding to each certificate image; A third determining unit, configured to determine the key information area of ​​each certificate image and the information type corresponding to the key information area based on the certificate type corresponding to each certificate image; A recognition unit, used to recognize text information in a key information area of ​​each certificate image through optical character recognition; A generating unit is used to generate a motorcycle insurance low order based on the text information in the key information area of ​​each certificate image and the information type corresponding to the key information area.

9. An electronic device, characterized in that: It comprises a memory, a processor and a transceiver which are communicatively connected in sequence, wherein the memory is used to store computer programs, the transceiver is used to send and receive messages, and the processor is used to read the computer program to execute the online intelligent policy processing method for motorcycle insurance as described in any one of claims 1 to 7.

10. A computer program product comprising a computer program or instructions, characterized in that When the computer program or the instruction is executed by a computer, the method for intelligent online motorcycle insurance policy issuance and processing is implemented as described in any one of claims 1 to 7.

Images