WO2018233171A1 - 单证信息录入的方法、装置、计算机设备及存储介质 - Google Patents

单证信息录入的方法、装置、计算机设备及存储介质 Download PDF

Info

Publication number
WO2018233171A1
WO2018233171A1 PCT/CN2017/108079 CN2017108079W WO2018233171A1 WO 2018233171 A1 WO2018233171 A1 WO 2018233171A1 CN 2017108079 W CN2017108079 W CN 2017108079W WO 2018233171 A1 WO2018233171 A1 WO 2018233171A1
Authority
WO
WIPO (PCT)
Prior art keywords
document
document image
information
image
template
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Ceased
Application number
PCT/CN2017/108079
Other languages
English (en)
French (fr)
Inventor
李斌
吴海波
姜云鹏
凌剑
马向东
丁杰
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Ping An Technology Shenzhen Co Ltd
Original Assignee
Ping An Technology Shenzhen Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Ping An Technology Shenzhen Co Ltd filed Critical Ping An Technology Shenzhen Co Ltd
Publication of WO2018233171A1 publication Critical patent/WO2018233171A1/zh
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Images

Classifications

    • GPHYSICS
    • G06COMPUTING OR CALCULATING; COUNTING
    • G06QINFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q40/00Finance; Insurance; Tax strategies; Processing of corporate or income taxes
    • G06Q40/08Insurance
    • GPHYSICS
    • G06COMPUTING OR CALCULATING; COUNTING
    • G06VIMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00Arrangements for image or video recognition or understanding
    • G06V10/20Image preprocessing
    • G06V10/24Aligning, centring, orientation detection or correction of the image
    • G06V10/243Aligning, centring, orientation detection or correction of the image by compensating for image skew or non-uniform image deformations
    • GPHYSICS
    • G06COMPUTING OR CALCULATING; COUNTING
    • G06VIMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00Arrangements for image or video recognition or understanding
    • G06V10/20Image preprocessing
    • G06V10/25Determination of region of interest [ROI] or a volume of interest [VOI]
    • GPHYSICS
    • G06COMPUTING OR CALCULATING; COUNTING
    • G06VIMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00Arrangements for image or video recognition or understanding
    • G06V10/20Image preprocessing
    • G06V10/26Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion
    • G06V10/267Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion by performing operations on regions, e.g. growing, shrinking or watersheds
    • GPHYSICS
    • G06COMPUTING OR CALCULATING; COUNTING
    • G06VIMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V30/00Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
    • G06V30/40Document-oriented image-based pattern recognition
    • GPHYSICS
    • G06COMPUTING OR CALCULATING; COUNTING
    • G06VIMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V30/00Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
    • G06V30/40Document-oriented image-based pattern recognition
    • G06V30/41Analysis of document content
    • G06V30/414Extracting the geometrical structure, e.g. layout tree; Block segmentation, e.g. bounding boxes for graphics or text
    • GPHYSICS
    • G06COMPUTING OR CALCULATING; COUNTING
    • G06VIMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V30/00Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
    • G06V30/10Character recognition

Definitions

  • the present application relates to the field of computer processing, and in particular, to a method, device, computer device and storage medium for document information entry.
  • a method, apparatus, computer device, and storage medium for document information entry are provided.
  • a method for document information entry comprising:
  • the data information is stored according to the corresponding field information, and information entry is completed.
  • a device for entering information of a document comprising:
  • a type determining module configured to obtain a document image of the information to be entered, and determine a document type of the document image
  • a template determining module configured to determine, according to the document type, a document template corresponding to the document image
  • a processing module configured to align the document image with the document template
  • a cutting module configured to cut the aligned document image according to the document template to obtain a target data area corresponding to each field
  • An identification module configured to identify information in the target data area to obtain editable data information
  • the storage module is configured to store the data information according to the corresponding field information, and complete information entry.
  • a computer apparatus comprising a memory and a processor, the memory storing computer readable instructions, the computer readable instructions being executed by the processor such that the processor performs the following steps:
  • One or more non-transitory readable storage mediums storing computer readable instructions, when executed by one or more processors, cause the one or more processors to perform the following steps:
  • the data information is stored according to the corresponding field information, and information entry is completed.
  • FIG. 1 is a block diagram showing the internal structure of a computer device in an embodiment
  • FIG. 2 is a flow chart of a method for recording information of a document in an embodiment
  • FIG. 3 is a flow chart of a method for performing alignment processing according to a document image and a document template in an embodiment
  • FIG. 4 is a diagram showing an image of an aligned document cut according to a document template in one embodiment. Method flow diagram to the target data area corresponding to each field;
  • FIG. 5 is a flow chart of a method for identifying information in a target data area to obtain editable data information in an embodiment
  • FIG. 6 is a structural block diagram of an apparatus for recording information of a document in an embodiment
  • Figure 7 is a block diagram showing the structure of a processing module in one embodiment.
  • the computer device can be a terminal or a server.
  • the terminal may be a personal computer or a mobile electronic device including at least one of a mobile phone, a tablet, a personal digital assistant, or a wearable device.
  • the server can be a standalone server or a server cluster.
  • the computer device includes a processor connected by a system bus, a non-volatile storage medium, an internal memory, and a network interface.
  • the non-volatile storage medium of the computer device can store an operating system and computer readable instructions that, when executed, cause the processor to perform a method of document information entry.
  • the processor of the computer device is used to provide computing and control capabilities to support the operation of the entire computer device.
  • the internal memory can store computer readable instructions that, when executed by the processor, cause the processor to perform a method of document information entry.
  • the network interface of the computer device is used for network communication. It will be understood by those skilled in the art that the structure shown in FIG. 1 is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation of the computer device to which the solution of the present application is applied.
  • the specific computer device may It includes more or fewer components than those shown in the figures, or some components are combined, or have different component arrangements.
  • a method for document information entry is proposed.
  • the method may be applied to a terminal or a server, and specifically includes the following steps:
  • Step 202 Acquire a document image of the information to be entered, and determine a document type of the document image.
  • the document image refers to a document that exists in the form of a picture, such as a scanned copy of a document, a photo, and the like.
  • a document is a document that can be used as a voucher, such as an insurance policy, a policy, and the like. Because the functions of different documents are different, the corresponding layouts and specifications are different. Therefore, the documents can be divided into different types according to the layout specifications of the documents. After obtaining the document image of the information to be entered, it is first necessary to determine the type of the document corresponding to the document image, so as to facilitate the subsequent determination of the document template corresponding to the document type according to the type of the document. There are various ways to determine the type of the document image.
  • the locations of different document types are stored in different folders, and then according to the obtained document image.
  • the location can directly determine the type of document corresponding to the document image.
  • the document type of the document image may be determined according to the document image number, wherein the document image number is used to uniquely identify a document image, and the representative image number may be included in the document image number.
  • the number of the certificate type for example, the second-to-last digit of the document image number can be set to represent the type of the document.
  • Step 204 Determine a document template corresponding to the document image according to the type of the document.
  • the document images are classified into a plurality of document types according to the document layout in advance, and the document images of the same document type have the same template layout. Therefore, the document template corresponding to different document types can be stored in advance. After obtaining the document type of the document image, the corresponding document template can be obtained directly according to the correspondence between the document type and the document template.
  • step 206 the document image is aligned with the document template.
  • the document image and the document template are all in the form of a rectangle, and there are many methods for aligning the document image with the document template.
  • the coordinate position of the document template is preset. The document template is loaded to a fixed position, so that the positions of the four corners of the document template are correspondingly determined, and then the frame of the document image is positioned to obtain the coordinate positions of the four corners of the document image.
  • the frame coordinate positioning of the document image can use the Hough transform line method to detect the border of the document image, and then determine the coordinate position of the four corners, of course, other methods can also be used.
  • the document images are mostly obtained by scanning or photographing, it is inevitable that there will be problems of zooming, skewing, and even inverting. In order to accurately cut, the document image needs to be scaled before the alignment, and the document image of the same size as the document template is obtained, and then the deviation and the reverse problem are corrected, so that the document image can be accurately Align with the document template.
  • Step 208 Cut the aligned document image according to the document template to obtain a target data region corresponding to each field.
  • the document image is cut in units of fields to obtain a target data region corresponding to each field.
  • a field is the content of a topical information contained in a document. For example, the items "name”, “gender”, and “date of birth” in a policy are so-called “fields”.
  • the target data area refers to the area where the data information corresponding to the field is located. For example, if the data information corresponding to the "name” field is "Zhang San", then the area where "Zhang San" is located is the target data area corresponding to the "Name" field.
  • the fields contained in the same are the same, and the specifications of the fields are the same, so the document template corresponding to the document type can be set in advance.
  • the cutting mode corresponding to each document template in advance. Because once the document template is determined, the location of each field in the field is determined, and the location of the target data region corresponding to each field is also determined, so the corresponding cutting mode can be set according to the target data region corresponding to each field. That is, the document image is cut according to the cutting method corresponding to the document template, and the target data region corresponding to each field in the document image is obtained.
  • Step 210 Identify information in the target data area to obtain editable data information.
  • the information in the target data area needs to be identified to obtain editable data information.
  • the data information includes text information, digital information, and symbol information. Identifying the information in the target data area is mainly divided into two steps. The first step is to detect the number of targets. According to the area where the data information in the area is located, the data information in the target data area is positioned to determine the corresponding target area. In the second step, the data information in the target area is identified to obtain editable data information.
  • the identification method can adopt the existing picture text recognition technology. For example, OCR (Optical Character Recognition) method can be used for identification. The recognition process is to identify the data information in the picture as editable data information.
  • OCR Optical Character Recognition
  • Step 212 Store the data information according to the corresponding field information, and complete information entry.
  • each document image has a document number for uniquely identifying the document image, and after identifying the data information corresponding to each field information, in addition to storing the data information corresponding to the corresponding field information.
  • the data information corresponding to the same document image is also stored in association, thereby completing information entry.
  • the method of inputting the information of the document is completed automatically, and the data information in the document image can be recorded into the system without manual participation, which not only saves time and labor, but also can process multiple documents in parallel and improve The efficiency of the entry.
  • the aligned image of the document is cut to obtain the target data area corresponding to each field, and then the information in the target data area is identified, and the editable data information is obtained, and then the data information is correspondingly matched.
  • the field information is stored to complete the information entry.
  • the method automatically cuts out the target data area corresponding to each field in the document image according to the document template, and then identifies each target data area to obtain editable data information, and the method can automatically complete the information in the document image.
  • the entry without human intervention, not only saves time and effort, but also improves the efficiency of entry.
  • the step 206 of aligning the document image with the document template includes:
  • Step 206A determining the location of the title in the document image.
  • each document image contains a body title, such as "XX policy”.
  • the font used in the body title is larger than the size of other fonts in the body text, so the connection area of the black pixels contained in the body title is larger than that of other fonts in comparison with other fonts in the body text, and the feature is utilized.
  • step 206B the tilt angle of the document image is calculated according to the position where the title is located.
  • the image is tilted, for example, if If the position of the title in the picture is detected to be on the left side of the image, then the angle of the image of the document is 90 degrees to the left. Specifically, by scanning the image, an area including a black pixel (body heading) of a preset specification is obtained, and a line is directly under the area containing the black pixel of the preset specification, and the black pixel is obtained.
  • the underline corresponding to the area determines the tilt angle of the document image based on the detected head position and the angle between the underline and the horizontal line.
  • the direction of the image tilt can be qualitatively determined. For example, if the title is below, the image is reversed, if the title is on the left, the image is tilted to the left, and if the title is on the right, The image is tilted to the right.
  • the degree of tilt is specifically determined by the angle between the underline and the positive direction of the horizontal line.
  • the tilt angle is 0 degrees. If it is detected that the title position is above the image, but the angle between the underline and the positive direction of the horizontal line is 15 degrees, then the left angle of the document image is 15 degrees. If it is detected that the title position is below the image, and the angle of inclination of the underline with the positive direction of the horizontal line is 15 degrees, then the leftward tilt angle of the document image can be calculated to be 195 degrees (that is, shifted to the right by 165 degrees).
  • Step 206C Rotate the document image according to the tilt angle to determine the coordinate position corresponding to the four corners of the rotated document image.
  • the single angle is compared according to the tilt angle.
  • the image is rotated so that it is tilted at an angle of 0 degrees. For example, if it is detected that the angle of the image is 90 degrees to the left, then the picture needs to be rotated 90 degrees clockwise to make it in the correct position.
  • the document image can be added to the Matlab to obtain an image matrix corresponding to the document image, and the position of the image edge can be determined according to the image matrix, and the coordinate positions corresponding to the four corners of the image edge are automatically determined.
  • step 206D the document image is aligned with the document template according to the coordinate position corresponding to the four corners of the document image.
  • the offset between the coordinate position of the document image and the coordinate position of the document template can be calculated, since both are Rectangular images, all in the positive direction, so you only need to calculate the offset between the coordinates of an angle and the coordinates of its corresponding angle to determine the offset of the document image and the document template.
  • the offset between the document image and the coordinates of the upper left corner of the document template is calculated separately, and then according to the offset, the document image is moved to a position coincident with the document template, that is, the document image and the document are The template is strictly aligned.
  • the step 208 of cutting the aligned document images according to the document template to obtain the target data regions corresponding to the respective fields includes:
  • Step 208A Obtain a field cutting rule corresponding to the document template.
  • each document template contains several fields, in order to Extract the data information corresponding to each field in the document image, and set the corresponding field cutting rule according to the layout of the field in the document template in advance, that is, set the cutting mode corresponding to each field, and extract the target corresponding to each field according to the cutting mode.
  • Data area For example, if the target data area corresponding to the "Name" field is directly below the "Name" field, and the shape specification size of the target data area is: rectangle, 1 cm x 3 cm, then the corresponding cutting method is to set a rectangle that clips the target data area. frame. Then, the field cutting rule of the document template is stored in one-to-one correspondence with the document template. Therefore, after obtaining the document template, the corresponding field cutting rule can be obtained according to the document template.
  • step 208B the document image is cut according to the field cutting rule, and the target data area corresponding to each field is obtained.
  • the field cutting rule corresponding to each field in the document template is obtained according to the document template, and then the document image is cut according to the field cutting rule, and the target data region corresponding to each field is obtained.
  • the target data area refers to an area containing data information corresponding to the field.
  • the target data area corresponding to the “name” field is an area containing the data information corresponding to the “name” field, for example, including a specific name, such as The area of "Zhang San".
  • the step 210 of identifying information in the target data region to obtain editable data information includes:
  • step 210A the data information in the target data area is located.
  • the data information in the target data region image is located, so that the located data information is included in the minimum circumscribed rectangle, that is, the target data.
  • the data information in the area is separately extracted by a minimum circumscribed rectangle to obtain a target area picture, wherein the four sides of the minimum circumscribed rectangle are tangent to the uppermost, lowermost, leftmost, and rightmost ends of the data information.
  • the target area picture containing the data information is subsequently identified as the identification object.
  • step 210B the data information is identified by using a picture text recognition technology to obtain editable data information.
  • the extracted minimum circumscribed rectangular image is used as the identification object, and the image character recognition technology is used to include
  • the data information is identified to obtain editable data information.
  • the image text recognition technology can adopt the existing recognition method.
  • the OCR (Optical Character Recognition) method can be used for recognition, and the recognition process is to identify the data information in the image as editable data information.
  • the area containing the data information is segmented to form a plurality of sub-images, each of which contains partial data information, By recognizing multiple sub-images in parallel, the speed of text recognition is improved.
  • the step of obtaining the document image of the information to be entered, and determining the document type of the document image includes: acquiring the document image of the information to be recorded, and extracting the image number of the document image, according to the preset image number. The correspondence between the type of the document and the type of the document determines the type of the document of the document image.
  • the image number is used to uniquely identify a document image, wherein the image number may be directly used as a document number or may be a number assigned separately for the image.
  • the image number is associated with the document type in advance, for example, the second digit of the image number can be associated with the document type, for example, if the image number The second digit in the figure is 1, indicating that the insurance is accident insurance. If the second digit is 2, it represents auto insurance, etc., by presetting the correspondence between the image number and the document type, according to the obtained order.
  • the image number of the image can be used to determine the type of document for the document image.
  • an apparatus for document information entry comprising:
  • the type determining module 602 is configured to obtain a document image of the information to be entered, and determine a document type of the document image.
  • the template determining module 604 is configured to determine a document template corresponding to the document image according to the type of the document.
  • the processing module 606 is configured to perform alignment processing on the document image and the document template.
  • the cutting module 608 is configured to cut the aligned document image according to the document template to obtain a target data region corresponding to each field.
  • the identification module 610 is configured to identify information in the target data area to obtain editable data information.
  • the storage module 612 is configured to store the data information according to the corresponding field information, and complete the information entry.
  • the processing module 606 includes:
  • the location determining module 606A is configured to determine a location of the title in the document image.
  • An angle calculation module 606B configured to calculate a tilt angle of the document image according to the location of the title degree.
  • the coordinate determining module 606C is configured to rotate the document image according to the tilt angle to determine a coordinate position corresponding to the four corners of the rotated document image.
  • the alignment module 606D is configured to align the document image with the document template according to the coordinate position corresponding to the four corners of the document image.
  • the cutting module is further configured to acquire a field cutting rule corresponding to the document template, and cut the document image according to the field cutting rule to obtain a target data region corresponding to each field.
  • the identification module is further configured to locate data information in the target data area, and use the image text recognition technology to identify the data information to obtain editable data information.
  • the type determining module is further configured to acquire a document image of the information to be recorded, extract an image number of the document image, and determine a document image according to a correspondence between the preset image number and the document type.
  • Document type is further configured to acquire a document image of the information to be recorded, extract an image number of the document image, and determine a document image according to a correspondence between the preset image number and the document type.
  • the various modules in the device for recording the above information may be implemented in whole or in part by software, hardware, and combinations thereof.
  • the network interface may be an Ethernet card or a wireless network card.
  • the above modules may be embedded in the hardware in the processor or in the memory in the server, or may be stored in the memory in the server, so that the processor calls the corresponding operations of the above modules.
  • the processor can be a central processing unit (CPU), a microprocessor, a microcontroller, or the like.
  • the apparatus for recording document information described above can be implemented in the form of a computer readable instruction that can be executed on a computer device as shown in FIG.
  • the embodiment of the present application provides a computer device.
  • the internal structure of the computer device may correspond to the structure shown in FIG. 1. That is, the computer device may be a server or a terminal, and includes a series of computers stored in the memory.
  • the readable instructions when the computer readable instructions are executed by the processor, can implement the method of document information entry proposed by the embodiments of the present application.
  • a computer device includes a memory, a processor, and computer readable instructions stored on the memory and operative on the processor, the processor implementing the computer readable instructions to implement the following steps : obtaining a document image of the information to be entered, determining the type of the document of the document image; Determining, by the document type, a document template corresponding to the document image; aligning the document image with the document template; and cutting the aligned document image according to the document template Obtaining a target data area corresponding to each field; identifying information in the target data area to obtain editable data information; storing the data information according to the corresponding field information, and completing information entry.
  • the step of the processor performing the alignment process of the document image with the document template comprises: determining a location of a title in the document image, according to the title Calculating a tilt angle of the document image, and rotating the document image according to the tilt angle to determine a coordinate position corresponding to four corners of the rotated document image, corresponding to four corners of the document image The coordinate position aligns the document image with the document template.
  • the step of performing, by the processing, cutting the aligned document image according to the document template to obtain a target data region corresponding to each field comprises: obtaining a correspondence with the document template
  • the field cutting rule is to cut the document image according to the field cutting rule to obtain a target data area corresponding to each field.
  • the step of the processor performing the identification of the information in the target data area to obtain editable data information comprises: locating data information in the target data area, The data information is identified by using picture text recognition technology to obtain editable data information.
  • the step of acquiring the document image of the document to be recorded by the processor, and determining the document type of the document image comprises: acquiring a document image of the information to be entered, and extracting the The image number of the document image determines the document type of the document image according to the correspondence between the preset image number and the document type.
  • a computer readable storage medium having computer instructions stored thereon, the instructions being executed by the processor to: obtain a document image of the information to be entered, and determine the image of the document a document type; determining, corresponding to the document image, according to the document type a document template; the document image is aligned with the document template; and the aligned document image is cut according to the document template to obtain a target data region corresponding to each field; and the target data is The information in the area is identified to obtain editable data information; the data information is stored according to the corresponding field information, and information entry is completed.
  • the step of the processor performing the alignment process of the document image with the document template comprises: determining a location of a title in the document image, according to the title Calculating a tilt angle of the document image, and rotating the document image according to the tilt angle to determine a coordinate position corresponding to four corners of the rotated document image, corresponding to four corners of the document image The coordinate position aligns the document image with the document template.
  • the step of performing, by the processing, cutting the aligned document image according to the document template to obtain a target data region corresponding to each field comprises: obtaining a correspondence with the document template
  • the field cutting rule is to cut the document image according to the field cutting rule to obtain a target data area corresponding to each field.
  • the step of the processor performing the identification of the information in the target data area to obtain editable data information comprises: locating data information in the target data area, The data information is identified by using picture text recognition technology to obtain editable data information.
  • the step of acquiring the document image of the document to be recorded by the processor, and determining the document type of the document image comprises: acquiring a document image of the information to be entered, and extracting the The image number of the document image determines the document type of the document image according to the correspondence between the preset image number and the document type.
  • the program when executed, may include the flow of an embodiment of the methods as described above.
  • the foregoing storage medium may be a non-volatile storage medium such as a magnetic disk or an optical disk.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Multimedia (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Business, Economics & Management (AREA)
  • Artificial Intelligence (AREA)
  • Accounting & Taxation (AREA)
  • Finance (AREA)
  • Geometry (AREA)
  • Computer Graphics (AREA)
  • Development Economics (AREA)
  • Economics (AREA)
  • Marketing (AREA)
  • Strategic Management (AREA)
  • Technology Law (AREA)
  • General Business, Economics & Management (AREA)
  • Processing Or Creating Images (AREA)
  • Character Input (AREA)

Abstract

一种单证信息录入的方法,包括:获取待录入信息的单证影像,确定所述单证影像的单证类型(202),根据所述单证类型确定与所述单证影像对应的单证模板(204),将所述单证影像与所述单证模板进行对齐处理(206),根据所述单证模板将对齐处理后的单证影像进行切割得到各个字段对应的目标数据区域(208),对所述目标数据区域中的信息进行识别,得到可编辑的数据信息(210);将所述数据信息按照所对应的字段信息进行存储,完成信息录入(212)。

Description

单证信息录入的方法、装置、计算机设备及存储介质
本申请要求于2017年6月23日提交中国专利局、申请号为201710487870X、发明名称为“单证信息录入的方法、装置、计算机设备及存储介质”的中国专利申请的优先权,其全部内容通过引用结合在本申请中。
技术领域
本申请涉及计算机处理领域,特别是涉及一种单证信息录入的方法、装置、计算机设备及存储介质。
背景技术
随着保险业务的发展,越来越多的人开始进行投保,用户申请投保会向保险公司提供相关资料,很多资料都是以影像的形式存在的,这种单证影像由于是以图片形式存在的,所以不能编辑,因此,需要将单证影像中的数据录入到系统中,传统的单证信息录入是采用人工方式通过界面录入各项信息,耗时耗力且如果某些录入错误被校验后,必须重新录入,所以录入效率比较低。
发明内容
根据本申请的各种实施例,提供了一种单证信息录入的方法、装置、计算机设备及存储介质。
一种单证信息录入的方法,包括:
获取待录入信息的单证影像,确定所述单证影像的单证类型;
根据所述单证类型确定与所述单证影像对应的单证模板;
将所述单证影像与所述单证模板进行对齐处理;
根据所述单证模板将对齐处理后的单证影像进行切割得到各个字段对应的目标数据区域;
对所述目标数据区域中的信息进行识别,得到可编辑的数据信息;及
将所述数据信息按照所对应的字段信息进行存储,完成信息录入。
一种单证信息录入的装置,包括:
类型确定模块,用于获取待录入信息的单证影像,确定所述单证影像的单证类型;
模板确定模块,用于根据所述单证类型确定与所述单证影像对应的单证模板;
处理模块,用于将所述单证影像与所述单证模板进行对齐处理;
切割模块,用于根据所述单证模板将对齐处理后的单证影像进行切割得到各个字段对应的目标数据区域;
识别模块,用于对所述目标数据区域中的信息进行识别,得到可编辑的数据信息;及
存储模块,用于将所述数据信息按照所对应的字段信息进行存储,完成信息录入。
一种计算机设备,包括存储器和处理器,所述存储器中存储有计算机可读指令,所述计算机可读指令被所述处理器执行时,使得所述处理器执行以下步骤:
获取待录入信息的单证影像,确定所述单证影像的单证类型;
根据所述单证类型确定与所述单证影像对应的单证模板;
将所述单证影像与所述单证模板进行对齐处理;
根据所述单证模板将对齐处理后的单证影像进行切割得到各个字段对应的目标数据区域;
对所述目标数据区域中的信息进行识别,得到可编辑的数据信息;及将所述数据信息按照所对应的字段信息进行存储,完成信息录入。
一个或多个存储有计算机可读指令的非易失性可读存储介质,所述计算机可读指令被一个或多个处理器执行时,使得所述一个或多个处理器执行以下步骤:
获取待录入信息的单证影像,确定所述单证影像的单证类型;
根据所述单证类型确定与所述单证影像对应的单证模板;
将所述单证影像与所述单证模板进行对齐处理;
根据所述单证模板将对齐处理后的单证影像进行切割得到各个字段对应的目标数据区域;
对所述目标数据区域中的信息进行识别,得到可编辑的数据信息;及
将所述数据信息按照所对应的字段信息进行存储,完成信息录入。
本申请的一个或多个实施例的细节在下面的附图和描述中提出。本申请的其它特征、目的和优点将从说明书、附图以及权利要求书变得明显。
附图说明
为了更清楚地说明本申请实施例中的技术方案,下面将对实施例描述中所需要使用的附图作简单地介绍,显而易见地,下面描述中的附图仅仅是本申请的一些实施例,对于本领域普通技术人员来讲,在不付出创造性劳动的前提下,还可以根据这些附图获得其他的附图。
图1为一个实施例中计算机设备的内部结构框图;
图2为一个实施例中单证信息录入的方法流程图;
图3为一个实施例中根据单证影像和单证模板进行对齐处理的方法流程图;
图4为一个实施例中根据单证模板将对齐处理后的单证影像进行切割得 到各个字段对应的目标数据区域的方法流程图;
图5为一个实施例中对目标数据区域中的信息进行识别,得到可编辑的数据信息的方法流程图;
图6为一个实施例中单证信息录入的装置结构框图;
图7为一个实施例中处理模块的结构框图。
具体实施方式
为了使本申请的目的、技术方案及优点更加清楚明白,以下结合附图及实施例,对本申请进行进一步详细说明。应当理解,此处所描述的具体实施例仅仅用以解释本申请,并不用于限定本申请。
如图1所示,为一个实施例中计算机设备的内部结构示意图。该计算机设备可以是终端也可以是服务器。终端可以是个人计算机或者移动电子设备,移动电子设备包括手机、平板电脑、个人数字助理或者穿戴式设备等中的至少一种。服务器可以是独立的服务器,也可以是服务器集群。参照图1,该计算机设备包括通过系统总线连接的处理器、非易失性存储介质、内存储器、网络接口。其中,该计算机设备的非易失性存储介质可存储操作系统和计算机可读指令,该计算机可读指令被执行时,可使得处理器执行一种单证信息录入的方法。该计算机设备的处理器用于提供计算和控制能力,支撑整个计算机设备的运行。该内存储器中可储存有计算机可读指令,该计算机可读指令被处理器执行时,可使得处理器执行一种单证信息录入的方法。计算机设备的网络接口用于进行网络通信。本领域技术人员可以理解,图1中示出的结构,仅仅是与本申请方案相关的部分结构的框图,并不构成对本申请方案所应用于其上的计算机设备的限定,具体的计算机设备可以包括比图中所示更多或更少的部件,或者组合某些部件,或者具有不同的部件布置。
如图2所示,在一个实施例中,提出了一种单证信息录入的方法,该方法可以应用于终端或服务器中,具体包括以下步骤:
步骤202,获取待录入信息的单证影像,确定单证影像的单证类型。
在本实施例中,单证影像是指以图片形式存在的单证,比如单证的扫描件、照片等。单证是指可以作为凭证的单据,比如,投保书、保单等。由于不同单证的作用不同,其相应的布局、规格都不尽相同,所以可以预先根据单证的布局规格将单证划分为不同的类型。获取到待录入信息的单证影像后,首先需要确定该单证影像对应的单证类型,便于后续根据单证类型确定与该单证类型对应的单证模板。确定单证影像类型的方式有多种,在一个实施例中,通过预先设置不同单证类型存储的位置,不同单证类型分别存储在不同的文件夹中,然后根据获取到的单证影像所在的位置就可以直接确定与该单证影像对应的单证类型。在另一个实施例中,可以根据单证影像编号来确定单证影像的单证类型,其中,单证影像编号用来唯一标识一个单证影像,可以在该单证影像编号中包含有代表单证类型的数字,比如,可以设置单证影像编号的倒数第二位代表单证类型。
步骤204,根据单证类型确定与单证影像对应的单证模板。
在本实施例中,预先根据单证布局将单证影像分为多种单证类型,相同单证类型的单证影像具有相同的模板布局。所以可以预先存储不同单证类型对应的单证模板。获取到单证影像的单证类型后,便可以直接根据单证类型与单证模板之间的对应关系获取相应的单证模板。
步骤206,将单证影像与单证模板进行对齐处理。
在本实施例中,为了便于后续根据单证模板进行切割,首先需要将单证影像与单证模板进行对齐处理,以保证单证影像的各个部分与单证模板中的各个部分进行一一对齐。具体地,单证影像和单证模板都是以矩形的形式存在的,将单证影像与单证模板进行对齐的方法有很多种,在一个实施例中,预先设置单证模板的坐标位置,即将单证模板加载到某个固定的位置,从而该单证模板的四个角的位置也就相应的确定了,之后对单证影像的边框进行定位,得到单证影像四个角的坐标位置,然后计算每个角与单证模板中相应角的坐标偏移量,根据计算得到的偏移量将单证影像移动到与单证模板重合 的位置。其中,单证影像的边框坐标定位可以采用霍夫变换直线方法检测单证影像的边框,进而来确定四个角的坐标位置,当然也可以采用别的方式。此外,由于单证影像大多是通过扫描的方式或拍照的方式获取的,所以难免会存在缩放、偏斜、甚至颠倒的问题。为了准确的进行切割,在对齐之前需要对单证影像进行放缩处理,得到与单证模板相同大小的单证影像,然后再针对偏斜、颠倒的问题进行纠偏,便于使单证影像能够精确的与单证模板对齐。
步骤208,根据单证模板将对齐处理后的单证影像进行切割得到各个字段对应的目标数据区域。
在本实施例中,将单证影像和单证模板进行对齐后,以字段为单位对该单证影像进行切割得到各个字段对应的目标数据区域。字段是指单证中包含的某一专题信息的内容。比如,一个保单中的“姓名”、“性别”、“出生日期”这些项就是所谓的“字段”。目标数据区域是指与字段对应的数据信息所在的区域。比如,假设“姓名”字段对应的数据信息为“张三”,那么“张三”所在的区域就是与“姓名”字段对应的目标数据区域。对于同一类单证来说,其里面包含的字段是相同的,且字段的规格也都是相同的,所以可以预先设置与该单证类型对应的单证模板。为了根据单证模板对单证影像进行切割得到各个字段对应的目标数据区域,需要预先设置每个单证模板对应的切割方式。因为一旦单证模板确定,其里面的各个字段所在的位置都是确定的,各个字段对应的目标数据区域所在的位置也是确定的,所以就可以根据各个字段对应的目标数据区域设置对应的切割方式,即根据单证模板对应的切割方式对单证影像进行切割,得到单证影像中各个字段对应的目标数据区域。
步骤210,对目标数据区域中的信息进行识别,得到可编辑的数据信息。
在本实施例中,由于提取到的目标数据区域是以图片形式存在的,所以提取到目标数据区域后,还需要对目标数据区域中的信息进行识别,以便得到可编辑的数据信息。其中,数据信息包括文字信息、数字信息以及符号信息。对目标数据区域中的信息进行识别主要分为两步,第一步,检测目标数 据区域中数据信息所在的区域,即对目标数据区域中的数据信息进行定位确定相应的目标区域。第二步,对目标区域中的数据信息进行识别得到可编辑的数据信息。识别的方法可以采用现有的图片文字识别技术,比如,可以采用OCR(Optical Character Recognition,光学字符识别)方法进行识别,识别的过程就是将图片中的数据信息识别为可编辑的数据信息。
步骤212,将数据信息按照所对应的字段信息进行存储,完成信息录入。
在本实施例中,识别得到字段对应的可编辑的数据信息后,将数据信息根据对应的字段信息自动存入到相应的位置,并且将同一单证影像中的数据信息需要进行关联存储。具体地,每个单证影像都有一个单证编号,用于唯一标识该单证影像,识别得到各个字段信息对应的数据信息后,除了要将各个数据信息与其对应的字段信息进行对应存储外,还需要将同一单证影像对应的数据信息进行关联存储,从而完成信息录入。该单证信息录入的方法整个过程都在全自动完成,无需人工参与,即可将单证影像中的数据信息录入到系统中,不仅省时省力,而且可以并行处理多张单证影像,提高了录入的效率。
在本实施例中,通过获取待录入信息的单证影像,确定单证影像的单证类型,根据单证类型确定与单证影像对应的单证模板,将单证影像与单证模板进行对齐处理,根据单证模板将对齐处理后的单证影像进行切割得到各个字段对应的目标数据区域,继而对目标数据区域中的信息进行识别,得到可编辑的数据信息,进而将数据信息按照所对应的字段信息进行存储,完成信息录入。该方法根据单证模板自动对单证影像中的各个字段对应的目标数据区域切割出来,然后对各个目标数据区域进行识别得到可编辑的数据信息,该方法能够自动完成对单证影像中信息的录入,无须人工参与,不但省时省力,而且提高了录入效率。
如图3所示,在一个实施例中,将单证影像与单证模板进行对齐处理的步骤206包括:
步骤206A,确定单证影像中标题所在的位置。
在本实施例中,每个单证影像中都包含有正文标题,比如“XX保单”。而且正文标题采用的字体比正文中其他字体的尺寸都要大一些,所以相对于正文中的其他字体,该正文标题所包含的黑色像素的连通区域就会比别的字体的大,利用该特点,通过检测图片中文字的黑色像素就可以确定单证影像中标题所在的位置。
步骤206B,根据标题所在的位置计算单证影像的倾斜角度。
在本实施例中,因为单证中标题所在的位置都是在单证的正上方中间的位置,那么如果检测到标题所在的位置偏离了正上方的位置,就说明图片存在倾斜,比如,如果检测到图片中标题位置在图像的正左侧,那么说明该单证影像的倾斜角度为左偏90度。具体地,通过对图像进行扫描,获取包含有预设规格的黑色像素(正文标题)的区域,并在该含有预设规格的黑色像素的区域的正下方进行划线,得到与该黑色像素的区域对应的下划线(下划线与正文标题互相平行),以水平线正方向为参考,根据检测到的标题位置和下划线与该水平线的角度来确定单证影像的倾斜角度。其中,根据标题在图像中的大致位置可以定性判断图像倾斜的方向,比如,如果标题在下方,说明图像是颠倒的,如果标题在左边,说明图像是向左倾斜的,如果标题在右边,说明图像是向右倾斜的。但是具体倾斜了多少度,需要根据下划线与水平线的正方向的角度来确定。比如,检测到标题位置在图像上方,如果下划线与水平线的平行,那么说明该倾斜角度为0度。如果检测到标题位置在图像上方,但是下划线与水平线正方向的夹角为15度,那么说明单证影像的向左倾斜角度为15度。如果检测到标题位置在图像下方,且下划线与水平线正方向的倾斜角度为15度,那么可以计算得到单证影像的向左倾斜角度为195度(也就是向右偏移了165度)。
步骤206C,根据倾斜角度旋转单证影像,确定旋转后的单证影像四个角所对应的坐标位置。
本实施例中,当计算得到单证影像的倾斜角度后,根据该倾斜角度对单 证影像进行旋转,使其倾斜角度为0度。比如,如果检测到单证影像的倾斜角度为左偏90度,那么就需要将该图片顺时针旋转90度,使其处于正确的位置。然后计算旋转后单证影像对应的四个角的坐标位置。具体地,可以将单证影像加入到Matlab中,得到单证影像对应的图像矩阵,根据该图像矩阵便可确定图像边缘的位置,自动确定图像边缘四个角分别对应的坐标位置。
步骤206D,根据单证影像四个角所对应的坐标位置将单证影像与单证模板进行对齐。
在本实施例中,确定了单证影像四个角分别对应的坐标后,就可以计算该单证影像的坐标位置和单证模板所在的坐标位置之间的偏移量,由于两者都是矩形图像,且都处于正方向,所以只需要计算一个角的坐标与其相应的角的坐标之间的偏移量就可以确定单证影像和单证模板的偏移量。比如,分别计算单证影像和单证模板左上角的坐标之间的偏移量,然后根据该偏移量,将单证影像移动到与单证模板重合的位置,即将单证影像和单证模板进行严格对齐。
如图4所示,在一个实施例中,根据单证模板将对齐处理后的单证影像进行切割得到各个字段对应的目标数据区域的步骤208包括:
步骤208A,获取与单证模板对应的字段切割规则。
在本实施例中,由于相同类型的单证具有相同的字段布局,也就是说,相同类型的单证对应有相同的单证模板,而每个单证模板中都包含有若干个字段,为了提取单证影像中各个字段对应的数据信息,预先根据单证模板中字段的布局设置相应的字段切割规则,即设置每个字段对应的切割方式,根据该切割方式来提取每个字段对应的目标数据区域。比如,如果“姓名”字段对应的目标数据区域在“姓名”字段的正下方,且目标数据区域的形状规格大小为:矩形,1cmX3cm,那么相应的切割方式就是设置剪裁出该目标数据区域的矩形框。然后将单证模板的字段切割规则与单证模板进行一一对应存储。故,获取到单证模板后,就可以根据单证模板获取相应的字段切割规则。
步骤208B,根据字段切割规则对单证影像进行切割,得到各个字段对应的目标数据区域。
在本实施例中,根据单证模板获取该单证模板中各个字段对应的字段切割规则,然后根据字段切割规则对单证影像进行切割,得到各个字段对应的目标数据区域。目标数据区域是指包含与字段相应的数据信息的区域,比如,“姓名”字段对应的目标数据区域就是与该“姓名”字段对应的包含有数据信息的区域,比如,包含有具体姓名,比如“张三”的区域。
如图5所示,在一个实施例中,对目标数据区域中的信息进行识别,得到可编辑的数据信息的步骤210包括:
步骤210A,对目标数据区域中的数据信息进行定位。
在本实施例中,为了能够快速准确地识别得到目标数据区域中的数据信息,首先对目标数据区域图像中的数据信息进行定位,使被定位的数据信息包含在最小外接矩形中,即将目标数据区域中的数据信息通过一个最小外接矩形单独提取出来,得到一个目标区域图片,其中,最小外接矩形的四条边分别与数据信息的最上端、最下端、最左端以及最右端相切。后续将该包含有数据信息的目标区域图片作为识别对象。
步骤210B,采用图片文字识别技术对数据信息进行识别,得到可编辑的数据信息。
在本实施例中,通过对目标数据区域中的数据信息进行定位提取到包含有数据信息的最小外接矩形图像后,将该提取到的最小外接矩形图像作为识别对象,采用图片文字识别技术对包含的数据信息进行识别,以便得到可编辑的数据信息。其中,图片文字识别技术可以采用现有的识别方法,比如,可以采用OCR(Optical Character Recognition,光学字符识别)方法进行识别,识别的过程就是将图片中的数据信息识别为可编辑的数据信息。在另一个实施例中,为了能够快速地对图片中的数据信息进行识别,对包含有数据信息的区域进行分割,形成若干个子图像,每个子图像中包含有部分数据信息, 通过对多个子图像并行进行识别,提升了文字识别的速度。
在一个实施例中,获取待录入信息的单证影像,确定单证影像的单证类型的步骤包括:获取待录入信息的单证影像,提取单证影像的影像编号,根据预设的影像编号和单证类型之间的对应关系,确定单证影像的单证类型。
在本实施例中,影像编号用于唯一标识一个单证影像,其中,影像编号可以直接使用单证号、也可以是单独为影像分配的编号。为了能够获取到待录入信息的单证影像的单证类型,预先将影像编号和单证类型进行关联,比如,可以使用影像编号的第二位数字与单证类型进行关联,比如,如果影像编号中的第二位数字为1,代表的险种为意外险,如果第二位数字为2,代表的是车险等,通过预先设置影像编号与单证类型之间的对应关系,根据获取到的单证影像的影像编号就可以确定单证影像的单证类型。
如图6所示,在一个实施例中,提出了一种单证信息录入的装置,该装置包括:
类型确定模块602,用于获取待录入信息的单证影像,确定单证影像的单证类型。
模板确定模块604,用于根据单证类型确定与单证影像对应的单证模板。
处理模块606,用于将单证影像与单证模板进行对齐处理。
切割模块608,用于根据单证模板将对齐处理后的单证影像进行切割得到各个字段对应的目标数据区域。
识别模块610,用于对目标数据区域中的信息进行识别,得到可编辑的数据信息。
存储模块612,用于将数据信息按照所对应的字段信息进行存储,完成信息录入。
如图7所示,在一个实施例中,处理模块606包括:
位置确定模块606A,用于确定单证影像中标题所在的位置。
角度计算模块606B,用于根据标题所在的位置计算单证影像的倾斜角 度。
坐标确定模块606C,用于根据倾斜角度旋转单证影像,确定旋转后的单证影像四个角所对应的坐标位置。
对齐模块606D,用于根据单证影像四个角所对应的坐标位置将单证影像与单证模板进行对齐。
在一个实施例中,切割模块还用于获取与单证模板对应的字段切割规则,根据字段切割规则对单证影像进行切割,得到各个字段对应的目标数据区域。
在一个实施例中,识别模块还用于对目标数据区域中的数据信息进行定位,采用图片文字识别技术对数据信息进行识别,得到可编辑的数据信息。
在一个实施例中,类型确定模块还用于获取待录入信息的单证影像,提取单证影像的影像编号,根据预设的影像编号和单证类型之间的对应关系,确定单证影像的单证类型。
上述单证信息录入的装置中的各个模块可全部或部分通过软件、硬件及其组合来实现。其中,网络接口可以是以太网卡或无线网卡等。上述各模块可以硬件形式内嵌于或独立于服务器中的处理器中,也可以以软件形式存储于服务器中的存储器中,以便于处理器调用执行以上各个模块对应的操作。该处理器可以为中央处理单元(CPU)、微处理器、单片机等。
上述单证信息录入的装置可以实现为一种计算机可读指令的形式,计算机可读指令可以在如图1所示的计算机设备上运行。
本申请实施例提出了一种计算机设备,计算机设备的内部结构可对应于如图1所示的结构,即该计算机设备既可以是服务器也可以是终端,其包括一系列存储于存储器上的计算机可读指令,当该计算机可读指令被处理器执行时,可以实现本申请各实施例提出的单证信息录入的方法。在一个实施例中,计算机设备包括存储器、处理器及存储在所述存储器上并可在所述处理器上运行的计算机可读指令,所述处理器执行所述计算机可读指令时实现以下步骤:获取待录入信息的单证影像,确定所述单证影像的单证类型;根据 所述单证类型确定与所述单证影像对应的单证模板;将所述单证影像与所述单证模板进行对齐处理;根据所述单证模板将对齐处理后的单证影像进行切割得到各个字段对应的目标数据区域;对所述目标数据区域中的信息进行识别,得到可编辑的数据信息;将所述数据信息按照所对应的字段信息进行存储,完成信息录入。
在一个实施例中,所述处理器所执行的所述将所述单证影像与所述单证模板进行对齐处理的步骤包括:确定所述单证影像中标题所在的位置,根据所述标题所在的位置计算单证影像的倾斜角度,根据所述倾斜角度旋转所述单证影像,确定旋转后的单证影像四个角所对应的坐标位置,根据所述单证影像四个角所对应的坐标位置将所述单证影像与所述单证模板进行对齐。
在一个实施例中,所述处理所执行的所述根据所述单证模板将对齐处理后的单证影像进行切割得到各个字段对应的目标数据区域的步骤包括:获取与所述单证模板对应的字段切割规则,根据所述字段切割规则对所述单证影像进行切割,得到各个字段对应的目标数据区域。
在一个实施例中,所述处理器所执行的所述对所述目标数据区域中的信息进行识别,得到可编辑的数据信息的步骤包括:对所述目标数据区域中的数据信息进行定位,采用图片文字识别技术对所述数据信息进行识别,得到可编辑的数据信息。
在一个实施例中,所述处理器所执行的所述获取待录入信息的单证影像,确定所述单证影像的单证类型的步骤包括:获取待录入信息的单证影像,提取所述单证影像的影像编号,根据预设的影像编号和单证类型之间的对应关系,确定所述单证影像的单证类型。
在一个实施例中,提出了一种计算机可读存储介质,其上存储有计算机指令,该指令被处理器执行时实现以下步骤:获取待录入信息的单证影像,确定所述单证影像的单证类型;根据所述单证类型确定与所述单证影像对应的 单证模板;将所述单证影像与所述单证模板进行对齐处理;根据所述单证模板将对齐处理后的单证影像进行切割得到各个字段对应的目标数据区域;对所述目标数据区域中的信息进行识别,得到可编辑的数据信息;将所述数据信息按照所对应的字段信息进行存储,完成信息录入。
在一个实施例中,所述处理器所执行的所述将所述单证影像与所述单证模板进行对齐处理的步骤包括:确定所述单证影像中标题所在的位置,根据所述标题所在的位置计算单证影像的倾斜角度,根据所述倾斜角度旋转所述单证影像,确定旋转后的单证影像四个角所对应的坐标位置,根据所述单证影像四个角所对应的坐标位置将所述单证影像与所述单证模板进行对齐。
在一个实施例中,所述处理所执行的所述根据所述单证模板将对齐处理后的单证影像进行切割得到各个字段对应的目标数据区域的步骤包括:获取与所述单证模板对应的字段切割规则,根据所述字段切割规则对所述单证影像进行切割,得到各个字段对应的目标数据区域。
在一个实施例中,所述处理器所执行的所述对所述目标数据区域中的信息进行识别,得到可编辑的数据信息的步骤包括:对所述目标数据区域中的数据信息进行定位,采用图片文字识别技术对所述数据信息进行识别,得到可编辑的数据信息。
在一个实施例中,所述处理器所执行的所述获取待录入信息的单证影像,确定所述单证影像的单证类型的步骤包括:获取待录入信息的单证影像,提取所述单证影像的影像编号,根据预设的影像编号和单证类型之间的对应关系,确定所述单证影像的单证类型。
本领域普通技术人员可以理解实现上述实施例方法中的全部或部分流程,是可以通过计算机可读指令来指令相关的硬件来完成,该计算机可读指令可存储于一计算机可读取存储介质中,该程序在执行时,可包括如上述各方法的实施例的流程。其中,前述的存储介质可为磁碟、光盘等非易失性存储介质。
以上所述实施例的各技术特征可以进行任意的组合,为使描述简洁,未对上述实施例中的各个技术特征所有可能的组合都进行描述,然而,只要这些技术特征的组合不存在矛盾,都应当认为是本说明书记载的范围。
以上所述实施例仅表达了本申请的几种实施方式,其描述较为具体和详细,但并不能因此而理解为对发明专利范围的限制。应当指出的是,对于本领域的普通技术人员来说,在不脱离本申请构思的前提下,还可以做出若干变形和改进,这些都属于本申请的保护范围。因此,本申请专利的保护范围应以所附权利要求为准。

Claims (20)

  1. 一种单证信息录入的方法,包括:
    获取待录入信息的单证影像,确定所述单证影像的单证类型;
    根据所述单证类型确定与所述单证影像对应的单证模板;
    将所述单证影像与所述单证模板进行对齐处理;
    根据所述单证模板将对齐处理后的单证影像进行切割得到各个字段对应的目标数据区域;
    对所述目标数据区域中的信息进行识别,得到可编辑的数据信息;及
    将所述数据信息按照所对应的字段信息进行存储,完成信息录入。
  2. 根据权利要求1所述的方法,其特征在于,所述将所述单证影像与所述单证模板进行对齐处理,包括:
    确定所述单证影像中标题所在的位置;
    根据所述标题所在的位置计算单证影像的倾斜角度;
    根据所述倾斜角度旋转所述单证影像,确定旋转后的单证影像四个角所对应的坐标位置;
    根据所述单证影像四个角所对应的坐标位置将所述单证影像与所述单证模板进行对齐。
  3. 根据权利要求1所述的方法,其特征在于,所述根据所述单证模板将对齐处理后的单证影像进行切割得到各个字段对应的目标数据区域,包括:
    获取与所述单证模板对应的字段切割规则;
    根据所述字段切割规则对所述单证影像进行切割,得到各个字段对应的目标数据区域。
  4. 根据权利要求1所述的方法,其特征在于,所述对所述目标数据区域中的信息进行识别,得到可编辑的数据信息,包括:
    对所述目标数据区域中的数据信息进行定位;
    采用图片文字识别技术对所述数据信息进行识别,得到可编辑的数据信息。
  5. 根据权利要求1所述的方法,其特征在于,所述获取待录入信息的单证影像,确定所述单证影像的单证类型,包括:
    获取待录入信息的单证影像,提取所述单证影像的影像编号;
    根据预设的影像编号和单证类型之间的对应关系,确定所述单证影像的单证类型。
  6. 一种单证信息录入的装置,包括:
    类型确定模块,用于获取待录入信息的单证影像,确定所述单证影像的单证类型;
    模板确定模块,用于根据所述单证类型确定与所述单证影像对应的单证模板;
    处理模块,用于将所述单证影像与所述单证模板进行对齐处理;
    切割模块,用于根据所述单证模板将对齐处理后的单证影像进行切割得到各个字段对应的目标数据区域;
    识别模块,用于对所述目标数据区域中的信息进行识别,得到可编辑的数据信息;及
    存储模块,用于将所述数据信息按照所对应的字段信息进行存储,完成信息录入。
  7. 根据权利要求6所述的装置,其特征在于,所述处理模块包括:
    位置确定模块,用于确定所述单证影像中标题所在的位置;
    角度计算模块,用于根据所述标题所在的位置计算单证影像的倾斜角度;
    坐标确定模块,用于根据所述倾斜角度旋转所述单证影像,确定旋转后的单证影像四个角所对应的坐标位置;
    对齐模块,用于根据所述单证影像四个角所对应的坐标位置将所述单证影像与所述单证模板进行对齐。
  8. 根据权利要求6所述的装置,其特征在于,所述切割模块还用于获取与所述单证模板对应的字段切割规则,根据所述字段切割规则对所述单证影 像进行切割,得到各个字段对应的目标数据区域。
  9. 根据权利要求6所述的装置,其特征在于,所述识别模块还用于对目标数据区域中的数据信息进行定位,采用图片文字识别技术对数据信息进行识别,得到可编辑的数据信息。
  10. 根据权利要求6所述的装置,其特征在于,类型确定模块还用于获取待录入信息的单证影像,提取单证影像的影像编号,根据预设的影像编号和单证类型之间的对应关系,确定单证影像的单证类型。
  11. 一种计算机设备,包括存储器和处理器,所述存储器中存储有计算机可读指令,所述计算机可读指令被所述处理器执行时,使得所述处理器执行以下步骤:
    获取待录入信息的单证影像,确定所述单证影像的单证类型;
    根据所述单证类型确定与所述单证影像对应的单证模板;
    将所述单证影像与所述单证模板进行对齐处理;
    根据所述单证模板将对齐处理后的单证影像进行切割得到各个字段对应的目标数据区域;
    对所述目标数据区域中的信息进行识别,得到可编辑的数据信息;及
    将所述数据信息按照所对应的字段信息进行存储,完成信息录入。
  12. 根据权利要求11所述的计算机设备,其特征在于,所述将所述单证影像与所述单证模板进行对齐处理,包括:
    确定所述单证影像中标题所在的位置;
    根据所述标题所在的位置计算单证影像的倾斜角度;
    根据所述倾斜角度旋转所述单证影像,确定旋转后的单证影像四个角所对应的坐标位置;
    根据所述单证影像四个角所对应的坐标位置将所述单证影像与所述单证模板进行对齐。
  13. 根据权利要求11所述的计算机设备,其特征在于,所述根据所述单 证模板将对齐处理后的单证影像进行切割得到各个字段对应的目标数据区域,包括:
    获取与所述单证模板对应的字段切割规则;
    根据所述字段切割规则对所述单证影像进行切割,得到各个字段对应的目标数据区域。
  14. 根据权利要求11所述的计算机设备,其特征在于,所述对所述目标数据区域中的信息进行识别,得到可编辑的数据信息,包括:
    对所述目标数据区域中的数据信息进行定位;
    采用图片文字识别技术对所述数据信息进行识别,得到可编辑的数据信息。
  15. 根据权利要求11所述的计算机设备,其特征在于,所述获取待录入信息的单证影像,确定所述单证影像的单证类型,包括:
    获取待录入信息的单证影像,提取所述单证影像的影像编号;
    根据预设的影像编号和单证类型之间的对应关系,确定所述单证影像的单证类型。
  16. 一个或多个存储有计算机可读指令的非易失性可读存储介质,所述计算机可读指令被一个或多个处理器执行时,使得所述一个或多个处理器执行以下步骤:
    获取待录入信息的单证影像,确定所述单证影像的单证类型;
    根据所述单证类型确定与所述单证影像对应的单证模板;
    将所述单证影像与所述单证模板进行对齐处理;
    根据所述单证模板将对齐处理后的单证影像进行切割得到各个字段对应的目标数据区域;
    对所述目标数据区域中的信息进行识别,得到可编辑的数据信息;及
    将所述数据信息按照所对应的字段信息进行存储,完成信息录入。
  17. 根据权利要求16所述的存储介质,其特征在于,所述将所述单证影 像与所述单证模板进行对齐处理,包括:
    确定所述单证影像中标题所在的位置;
    根据所述标题所在的位置计算单证影像的倾斜角度;
    根据所述倾斜角度旋转所述单证影像,确定旋转后的单证影像四个角所对应的坐标位置;
    根据所述单证影像四个角所对应的坐标位置将所述单证影像与所述单证模板进行对齐。
  18. 根据权利要求16所述的存储介质,其特征在于,所述根据所述单证模板将对齐处理后的单证影像进行切割得到各个字段对应的目标数据区域,包括:
    获取与所述单证模板对应的字段切割规则;
    根据所述字段切割规则对所述单证影像进行切割,得到各个字段对应的目标数据区域。
  19. 根据权利要求16所述的存储介质,其特征在于,所述对所述目标数据区域中的信息进行识别,得到可编辑的数据信息,包括:
    对所述目标数据区域中的数据信息进行定位;
    采用图片文字识别技术对所述数据信息进行识别,得到可编辑的数据信息。
  20. 根据权利要求16所述的存储介质,其特征在于,所述获取待录入信息的单证影像,确定所述单证影像的单证类型,包括:
    获取待录入信息的单证影像,提取所述单证影像的影像编号;
    根据预设的影像编号和单证类型之间的对应关系,确定所述单证影像的单证类型。
PCT/CN2017/108079 2017-06-23 2017-10-27 单证信息录入的方法、装置、计算机设备及存储介质 Ceased WO2018233171A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201710487870.X 2017-06-23
CN201710487870.XA CN107679442A (zh) 2017-06-23 2017-06-23 单证信息录入的方法、装置、计算机设备及存储介质

Publications (1)

Publication Number Publication Date
WO2018233171A1 true WO2018233171A1 (zh) 2018-12-27

Family

ID=61133491

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2017/108079 Ceased WO2018233171A1 (zh) 2017-06-23 2017-10-27 单证信息录入的方法、装置、计算机设备及存储介质

Country Status (2)

Country Link
CN (1) CN107679442A (zh)
WO (1) WO2018233171A1 (zh)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111488399A (zh) * 2020-05-06 2020-08-04 北京俩撇科技有限公司 一种区块链系统、交易处理方法及装置
CN111767818A (zh) * 2020-06-23 2020-10-13 北京思特奇信息技术股份有限公司 一种业务智能受理的方法和装置
CN112016561A (zh) * 2020-09-01 2020-12-01 中国银行股份有限公司 一种文本识别方法及相关设备

Families Citing this family (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108335199B (zh) * 2018-03-13 2021-01-26 平安普惠企业管理有限公司 贷款请求处理方法、装置、计算机设备和存储介质
CN109815954A (zh) * 2019-01-31 2019-05-28 科大讯飞股份有限公司 增值税发票图像的方向校正方法、装置、设备及存储介质
CN109934185B (zh) * 2019-03-18 2021-05-04 网易有道信息技术(北京)有限公司 数据处理方法及装置、介质和计算设备
CN109977949B (zh) * 2019-03-20 2024-01-26 深圳华付技术股份有限公司 边框微调的文本定位方法、装置、计算机设备及存储介质
CN111652229B (zh) * 2020-05-25 2023-09-12 泰康保险集团股份有限公司 一种信息输入方法、装置、电子设备及存储介质
CN112668041B (zh) * 2020-12-17 2023-07-04 平安消费金融有限公司 单证文件的生成方法、装置、计算机设备和存储介质
CN114998924A (zh) * 2022-05-18 2022-09-02 共道网络科技有限公司 信件信息的数字化管理方法和装置

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103034848A (zh) * 2012-12-19 2013-04-10 方正国际软件有限公司 一种表单类型的识别方法
CN103208004A (zh) * 2013-03-15 2013-07-17 北京英迈杰科技有限公司 票据信息区域自动识别和提取方法及设备
CN103617415A (zh) * 2013-11-19 2014-03-05 北京京东尚科信息技术有限公司 一种自动识别发票的装置和方法
WO2015062275A1 (en) * 2013-10-29 2015-05-07 Tencent Technology (Shenzhen) Company Limited Method, apparatus and system for information identification
CN105005793A (zh) * 2015-07-15 2015-10-28 广州敦和信息技术有限公司 一种发票字条自动识别录入的方法及装置
CN105678612A (zh) * 2015-12-30 2016-06-15 远光软件股份有限公司 移动端原始凭证电子化智能填单系统及方法

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103488965B (zh) * 2013-08-20 2016-06-15 成都我来啦网格信息科技有限公司 运单信息录入和彩色色块编码解码系统
CN105095842B (zh) * 2014-05-22 2018-12-11 口碑控股有限公司 一种单据的信息识别的方法和装置
CN105528604B (zh) * 2016-01-31 2018-12-11 华南理工大学 一种基于ocr的票据自动识别与处理系统

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103034848A (zh) * 2012-12-19 2013-04-10 方正国际软件有限公司 一种表单类型的识别方法
CN103208004A (zh) * 2013-03-15 2013-07-17 北京英迈杰科技有限公司 票据信息区域自动识别和提取方法及设备
WO2015062275A1 (en) * 2013-10-29 2015-05-07 Tencent Technology (Shenzhen) Company Limited Method, apparatus and system for information identification
CN103617415A (zh) * 2013-11-19 2014-03-05 北京京东尚科信息技术有限公司 一种自动识别发票的装置和方法
CN105005793A (zh) * 2015-07-15 2015-10-28 广州敦和信息技术有限公司 一种发票字条自动识别录入的方法及装置
CN105678612A (zh) * 2015-12-30 2016-06-15 远光软件股份有限公司 移动端原始凭证电子化智能填单系统及方法

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111488399A (zh) * 2020-05-06 2020-08-04 北京俩撇科技有限公司 一种区块链系统、交易处理方法及装置
CN111767818A (zh) * 2020-06-23 2020-10-13 北京思特奇信息技术股份有限公司 一种业务智能受理的方法和装置
CN111767818B (zh) * 2020-06-23 2024-04-26 北京思特奇信息技术股份有限公司 一种业务智能受理的方法和装置
CN112016561A (zh) * 2020-09-01 2020-12-01 中国银行股份有限公司 一种文本识别方法及相关设备
CN112016561B (zh) * 2020-09-01 2023-08-04 中国银行股份有限公司 一种文本识别方法及相关设备

Also Published As

Publication number Publication date
CN107679442A (zh) 2018-02-09

Similar Documents

Publication Publication Date Title
WO2018233171A1 (zh) 单证信息录入的方法、装置、计算机设备及存储介质
US11341605B1 (en) Document rectification via homography recovery using machine learning
WO2018233055A1 (zh) 保单信息录入的方法、装置、计算机设备及存储介质
US10699146B2 (en) Mobile document detection and orientation based on reference object characteristics
US10303968B2 (en) Method and apparatus for image recognition
CN111860489A (zh) 一种证件图像校正方法、装置、设备及存储介质
US8885972B2 (en) Straightening out distorted perspective on images
CN111444908A (zh) 图像识别方法、装置、终端和存储介质
CN111507251A (zh) 试题图像中作答区域的定位方法、装置和电子设备
CN111595850A (zh) 切片缺陷检测方法、电子装置及可读存储介质
CN111860527A (zh) 图像校正方法、图像校正装置、计算机设备及存储介质
WO2018233038A1 (zh) 基于深度学习的车牌识别方法、装置、设备及存储介质
US10410053B2 (en) Method, apparatus, system, and storage medium for detecting information card in image
US12141938B2 (en) Image processing system, image processing method, and program
WO2014086287A1 (zh) 文本图像自动切分方法及装置,自动切分手写条目的方法
WO2023098045A1 (zh) 图像对齐方法、装置、计算机设备和存储介质
CN110473249A (zh) 一种网页用户界面与设计稿的对比方法、装置及终端设备
WO2018205789A1 (zh) 卡片放置方向的识别方法、装置以及图像处理装置
CN103366165B (zh) 图像处理装置、图像处理方法以及设备
WO2021051527A1 (zh) 基于图像分割的文字定位方法、装置、设备及存储介质
US11881043B2 (en) Image processing system, image processing method, and program
CN115984859A (zh) 一种图像文字识别的方法、装置及存储介质
US9483834B1 (en) Object boundary detection in an image
WO2020232866A1 (zh) 扫描文本分段方法、装置、计算机设备和存储介质
WO2015021737A1 (zh) 一种将纸质文件转换为电子文件的方法

Legal Events

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

Ref document number: 17914912

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 14/05/2020)

122 Ep: pct application non-entry in european phase

Ref document number: 17914912

Country of ref document: EP

Kind code of ref document: A1