CN109255300B - Bill information extraction method, bill information extraction device, computer equipment and storage medium - Google Patents
Bill information extraction method, bill information extraction device, computer equipment and storage medium Download PDFInfo
- Publication number
- CN109255300B CN109255300B CN201810921331.7A CN201810921331A CN109255300B CN 109255300 B CN109255300 B CN 109255300B CN 201810921331 A CN201810921331 A CN 201810921331A CN 109255300 B CN109255300 B CN 109255300B
- Authority
- CN
- China
- Prior art keywords
- information
- area
- target
- rectangular
- rectangular area
- 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.)
- Active
Links
- 238000000605 extraction Methods 0.000 title claims abstract description 30
- 238000000034 method Methods 0.000 claims abstract description 28
- 238000004590 computer program Methods 0.000 claims description 15
- 238000005516 engineering process Methods 0.000 abstract description 8
- 238000012015 optical character recognition Methods 0.000 abstract 2
- 238000010586 diagram Methods 0.000 description 12
- 238000010801 machine learning Methods 0.000 description 7
- 230000006870 function Effects 0.000 description 4
- 238000012545 processing Methods 0.000 description 4
- 238000004364 calculation method Methods 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 238000003491 array Methods 0.000 description 1
- 238000004422 calculation algorithm Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 235000019800 disodium phosphate Nutrition 0.000 description 1
- 230000010365 information processing Effects 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
- G06V30/40—Document-oriented image-based pattern recognition
- G06V30/41—Analysis of document content
- G06V30/414—Extracting the geometrical structure, e.g. layout tree; Block segmentation, e.g. bounding boxes for graphics or text
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Computer Graphics (AREA)
- Geometry (AREA)
- Artificial Intelligence (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Image Analysis (AREA)
Abstract
The embodiment of the invention provides a bill information extraction method, a bill information extraction device, computer equipment and a storage medium. According to the method, the information items in the bill are identified by utilizing technologies such as OCR (optical character recognition) or CTPN (computer-to-digital network), each information item is provided with a rectangular area, the identified information items comprise a first rectangular area of a master information item and a second rectangular area of a characteristic information item, master information is removed by covering the first rectangular area, a target area is found through preliminary positioning, the position of the target information is determined according to the relative position relation of the second rectangular areas in the target area, and finally the target information is extracted. The master information is removed, a large amount of unnecessary information is removed, rectangular areas needing to be positioned are reduced, the information extraction efficiency is improved, the target information is initially positioned, the approximate position is found first, then the target information is accurately positioned from the approximate position, and the problem of inaccurate positioning caused by direct positioning of the target information in the excessive rectangular areas is avoided.
Description
Technical Field
The present invention relates to the field of information processing technologies, and in particular, to a method and apparatus for extracting bill information, a computer device, and a storage medium.
Background
At present, two methods for extracting information in a bill are mainly adopted: 1. positioning the information frame of each item of information on the bill, positioning the information frame position needing to identify the information from all the information frames, and finally identifying the text information in the information frame by using a text identification technology; 2. and positioning the information frame of each item of information on the bill, firstly, finding the information frame of the information name, then finding the information frame of the corresponding characteristic value according to the information frame of the information name, and finally, identifying text information, namely the characteristic information, in the information frame of the characteristic value. The two methods are more suitable for bills with less information content, and the identification time is longer for bills with more information content, because the more information content is, the more information frames are, the more information frames needing to be identified are positioned from more information frames, more calculation resources and calculation time are needed to be consumed, and the information identification efficiency is very low under the condition of inaccurate positioning.
Disclosure of Invention
The embodiment of the invention provides a bill information extraction method, a bill information extraction device, computer equipment and a storage medium, so as to improve the identification efficiency of bill information.
In a first aspect, an embodiment of the present invention provides a method for extracting bill information, where the method includes: dividing the position of each item of information in the bill image into rectangular areas, wherein the information in the bill image comprises master information and characteristic information, and the rectangular areas comprise a first rectangular area corresponding to the master information and a second rectangular area corresponding to the characteristic information; identifying a first rectangular area in which the master information is located from all the rectangular areas; covering the first rectangular area to obtain a target image; dividing the target image into a plurality of second preset areas according to the dividing proportion of each first preset area in the preset bill template, wherein the information layout of the preset bill template is consistent with the information layout of the bill image, and each second preset area comprises a plurality of second rectangular areas; acquiring target areas in which target information to be extracted is located from the second preset areas according to preset positioning rules; determining a target rectangular area where the target information is located according to the relative position relation between the second rectangular areas in the target area; target information in the target rectangular area is identified.
In a second aspect, an embodiment of the present invention further provides a bill information extracting apparatus, where the bill information extracting apparatus includes a unit for implementing the bill information extracting method in the first aspect.
In a third aspect, an embodiment of the present invention further provides a computer device, including a memory, and a processor connected to the memory; the memory is used for storing a computer program for realizing the bill information extraction method; the processor is configured to run a computer program stored in the memory to perform the method as described in the first aspect above.
In a fourth aspect, embodiments of the present invention provide a storage medium storing one or more computer programs executable by one or more processors to implement the method of the first aspect.
The embodiment of the invention provides a bill information extraction method, a bill information extraction device, computer equipment and a storage medium. According to the embodiment of the invention, the target area is found by removing the master information and preliminary positioning of the information, the position of the target information is determined in the target area according to the relative position relation of each second rectangular area, and finally the target information is extracted. The master information is removed, a large amount of unnecessary information is removed, rectangular areas needing to be positioned are reduced, the information extraction efficiency is improved, the target information is initially positioned, the approximate position is found first, then the target information is accurately positioned from the approximate position, and the problem of inaccurate positioning caused by direct positioning of the target information in the excessive rectangular areas is avoided.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for the description of the embodiments will be briefly described below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a schematic flow chart of a bill information extraction method according to an embodiment of the present invention;
FIG. 2a is a schematic diagram of a ticket with single-sided master information provided in accordance with an embodiment of the present invention;
FIG. 2b is a schematic diagram of a ticket with double-sided master information according to one embodiment of the present invention;
FIG. 3 is a schematic sub-flowchart of a bill information extraction method according to an embodiment of the present invention;
FIG. 4 is a schematic sub-flowchart of a bill information extraction method according to another embodiment of the present invention;
FIG. 5a is a schematic diagram of a preset ticket template according to an embodiment of the present invention;
FIG. 5b is a schematic diagram of a target image according to an embodiment of the present invention;
FIG. 6 is a schematic sub-flowchart of a bill information extraction method according to another embodiment of the present invention;
FIG. 7 is a flowchart of a bill information extraction method according to another embodiment of the present invention;
FIG. 8 is a schematic block diagram of a bill information extracting apparatus according to an embodiment of the present invention;
FIG. 9 is a schematic block diagram of a subunit of a bill information extracting apparatus according to an embodiment of the present invention;
fig. 10 is a schematic block diagram of a subunit of a bill information extracting apparatus according to another embodiment of the present invention;
FIG. 11 is a schematic block diagram of a subunit of a bill information extracting apparatus according to another embodiment of the present invention;
FIG. 12 is a schematic block diagram of a bill information extracting apparatus according to another embodiment of the present invention;
fig. 13 is a schematic block diagram of a computer device according to an embodiment of the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and fully with reference to the accompanying drawings, in which it is evident that the embodiments described are some, but not all embodiments of the invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
It should be understood that the terms "comprises" and "comprising," when used in this specification and the appended claims, specify the presence of stated features, integers, steps, operations, elements, and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and/or groups thereof.
It should also be understood that the term "and/or" as used in the present specification and the appended claims refers to any and all possible combinations of one or more of the associated listed items, and includes such combinations.
It will also be understood that, although the terms first, second, etc. may be used herein to describe various elements, these elements should not be limited by these terms, these terms should be used merely to distinguish one element from another.
Fig. 1 is a flow chart of a bill information extraction method according to an embodiment of the present invention, where the bill information extraction method can be applied to a bill information extraction platform. The method includes steps S101 to S107.
S101, dividing the position of each item of information in the bill image into rectangular areas, wherein the information in the bill image comprises master information and characteristic information, and the rectangular areas comprise a first rectangular area corresponding to the master information and a second rectangular area corresponding to the characteristic information.
Before acquiring bill information, shooting the bill to obtain a bill image, and processing the bill image by utilizing an OCR technology or CTPN (Coloured and Time Based Petri Net, a natural scene text detection technology) to distinguish the position of each item of information of the bill, wherein each item of information is framed in a rectangular box to obtain a rectangular area corresponding to each item of information. The image bill of the embodiment of the invention is a forward-placed image, and the text in the image is in a transverse direction.
Any type of bill has a fixed preset bill template, and the content and layout of a certain type of bill are consistent with the content and layout of the corresponding preset bill template. The preset bill template/bill has fixed content, such as bill title, information name, etc. and the characteristic information corresponding to the information name is changed based on the actual condition. The information in the bill can be classified according to the fixity of the information, for example, the information is classified into the master information and the characteristic information, the master information refers to the fixed content (except the bill title) in the bill, and the characteristic information refers to a specific value corresponding to the master information.
For example, a ticket of a vehicle registration file, the ticket includes various information of the vehicle, such as information of a category, a factory name, a model, an engine number, etc., of the vehicle, the category, the factory name, the model, the engine number, etc., of the vehicle are fixed as information names, and specific information corresponding to the category, the factory name, the model, the engine number, etc., can be changed according to specific vehicles, so that the name "category, the factory name, the model, the engine number" is master information, and specific information corresponding to the category, the factory name, happiness, the engine number, etc., is characteristic information.
The position of each item of information of the bill can be limited by a rectangular area, and the rectangular area where all the information is located can be rapidly obtained by OCR technology or CTPN technology. The rectangular area corresponding to the master information is a first rectangular area, and the rectangular area corresponding to the characteristic information is a second rectangular area.
S102, identifying a first rectangular area where the master information is located from all the rectangular areas.
Common notes can be divided into two categories: notes with single sided master information and notes with double sided master information. A ticket with single-sided master information refers to master information arranged in a single-sided row as shown in fig. 2a, and a ticket with double-sided master information refers to master information arranged in a left-and right-sided row as shown in fig. 2 b.
It should be noted that the coordinate system in which the target image is located is a rectangular coordinate system in which the lower left corner vertex of the target image is taken as the origin, the transverse direction is taken as the abscissa, and the longitudinal direction is taken as the ordinate, and the position coordinates (including the abscissa and the ordinate) of each rectangular region mentioned throughout are represented by pixels, and the upper left corner vertex of the rectangular region is taken as the position coordinates of the rectangular region.
In one embodiment, the ticket image is a ticket image with single-sided master information.
As shown in fig. 3, step S102 includes substeps S121-S124.
S121, acquiring the abscissa corresponding to each rectangular area.
S122, acquiring the minimum abscissa from all the obtained abscissas.
For notes with single-sided master information, the master information is often a left-aligned column of information, and thus, acquiring a rectangular region for the leftmost column can determine substantially the first rectangular region for all master information. The more left the position is, the smaller the abscissa is, so that a rectangular area with the smallest abscissa can be directly determined as a rectangular area corresponding to the master information, i.e., the first rectangular area.
And S123, comparing the minimum abscissa with the abscissa to acquire the abscissa of which the difference value is within a first error range from the minimum abscissa as a target abscissa.
And S124, determining the rectangular areas corresponding to the minimum abscissa and the target abscissa as the first rectangular area.
The master information is a left-aligned layout, so that the difference between the abscissa of the rectangular area corresponding to any master information and the smallest abscissa is not too large, and the abscissas within the first error range are regarded as the same column of coordinates as the smallest abscissas, so that the first rectangular area corresponding to other master information can be found according to the difference between the abscissas of other rectangular areas and the smallest abscissas.
The first error range has the following two values:
(1) The pixels of the ticket image are the same as the pixels of the preset ticket template (which is in picture format), the first error range is [ -30P,30P ], e.g. in one embodiment the first error range is selected to be [ -10P,10P ] or [ -20P,20P ], where P refers to the pixel (pixel).
(2) The bill image is different from the pixel of the preset bill template, under the condition that the pixel of the bill image is the same as the pixel of the preset bill template, namely, the first error range corresponding to the condition (1) is defined as a reference error, and is set as x, the first error range y is calculated through a formula a/x=b/y, wherein a represents the pixel distance corresponding to the actual length of the preset bill template (picture), b represents the pixel distance corresponding to the actual length of the bill image, for example, the pixel of the preset bill template is 1024 x 1024, the pixel distance corresponding to the actual length is 1024 pixels, the pixel of the bill image is 800 x 800, and the pixel distance corresponding to the actual length is 800 pixels.
In another embodiment, the ticket image is a ticket image with double sided master information.
As shown in fig. 4, step S102 includes substeps S1201-S1211.
S1201, a rectangular area where the bill title is located in the bill image is identified as a title rectangular area.
The bill header is generally centered on the top of the bill, and the ordinate of the bill header is the smallest, so that the bill header is easily recognized, and the rectangular area in the middle of the top of the bill image is the rectangular area of the header.
S1202, acquiring a central line of the title rectangular area, wherein the central line is perpendicular to the text direction of the bill title.
The text direction refers to the direction in which the text is written, and is transverse when the ticket is placed in the forward direction.
The bill header is centrally located with the center line of the rectangular area of the header overlapping the center line of the entire bill image.
S1203 divides the ticket image into a left area and a right area according to the center line.
The bill image with double-sided master information is generally bounded by a center line, the master information is divided into left and right side arrangements, that is, the bill image can be divided into a left side area and a right side area by the center line, the master information of the left side area is arranged in left alignment, and the master information of the right side area is also arranged in left alignment by the center line.
And S1204, acquiring the abscissa corresponding to each rectangular area in the left area.
S1205, obtaining the minimum abscissa from all the obtained abscissas.
And S1206, comparing the minimum abscissa with other abscissas to acquire an abscissa with a difference value within a first error range from the minimum abscissa as a first target abscissa.
And S1207, determining the rectangular area corresponding to the minimum abscissa and the first target abscissa as a first rectangular area corresponding to the master information of the left area.
The identification process of the first rectangular area corresponding to the master information in the left area is identical to the identification process of the first rectangular area of the bill image with the single-side master information, and is not described herein.
S1208, acquiring an abscissa corresponding to the central line.
And S1209, acquiring the abscissa corresponding to each rectangular area in the right area.
S1210, comparing the abscissa corresponding to each rectangular area in the right area with the abscissa corresponding to the central line, so as to obtain the abscissa with the difference value of the abscissas corresponding to the central line within a second error range as a second target abscissas.
And S1211, determining the rectangular area corresponding to the second target abscissa as a first rectangular area corresponding to the master information of the right area.
The left side area and the right side area are defined by the central line, the master information of the right side area is generally separated from the central line by a relatively close distance, and the master information of the right side area is relatively close to the central line, and the abscissa of the first rectangular area corresponding to the master information of the right side area is slightly larger than that of the central line, so that the first rectangular area corresponding to the master information in the right side area can be identified according to the central line, and particularly the rectangular area corresponding to the master information of the right side area is determined according to the rectangular area corresponding to the abscissa of the difference value between the abscissas of the central line within a certain distance range.
The second error range has the following two values:
(1) The pixels of the bill image are the same as those of the preset bill template, and the first error range is [10P,100P ], for example, in one embodiment, the first error range is selected to be [50P,100P ] or [60P,20P ];
(2) And (3) under the condition that the pixels of the bill image are different from the pixels of the preset bill template, namely under the condition that the pixels of the bill image are identical to the pixels of the preset bill template, defining a second error range corresponding to the condition (1) as a reference error, setting z as a second error range u, and calculating the second error range u through a formula a/z=b/u, wherein a represents a pixel distance corresponding to the actual length of the preset bill template (picture), and b represents a pixel distance corresponding to the actual length of the bill image.
And S103, covering the first rectangular area to obtain a target image.
The first rectangular area is covered, i.e. the master information in the first rectangular area is covered, leaving the second rectangular area. The master information is removed, the characteristic information is left, a large amount of unnecessary information is removed, the workload of subsequent machine learning recognition or text extraction can be reduced, and the information recognition or extraction efficiency is improved.
S104, dividing the target image into a plurality of second preset areas according to the dividing proportion of each first preset area in the preset bill template, wherein the information layout of the preset bill template is consistent with the information layout of the bill image, and each second preset area comprises a plurality of second rectangular areas.
Before identifying the information of a specific type of bill, machine learning is required to be carried out on the type of bill, the bill subjected to machine learning can be used as a preset bill template, the machine learning obtains the information distribution of the whole bill and the information (rectangular area) distribution condition around each item of information, and the second rectangular areas corresponding to the characteristic information of the bill used for machine learning are divided into different first preset areas according to a certain proportion. The division of the first preset areas can be preset manually, and each first preset area comprises a plurality of rectangular areas corresponding to the characteristic information. And taking the dividing proportion corresponding to the first preset region as a dividing reference for the second preset region of the target image, and correspondingly dividing the second preset region corresponding to the first preset region of the target image.
As shown in fig. 5a, after two first preset areas are respectively divided from top to bottom in the left area and the right area of the preset bill template, the position relationship (including a vertical coordinate difference and a horizontal coordinate difference) of the boundaries of the different first preset areas corresponding to the center point of the title is used as the dividing ratio of each boundary of the first preset areas, when the bill image needs to be divided into the second preset areas, the boundary of each second preset area is determined according to the dividing ratio of the boundary so as to realize the division of a plurality of second preset areas, and the second preset areas shown in fig. 5b are obtained, wherein the second preset areas comprise a second preset area 1, a second preset area 2, a second preset area 3 and a second preset area 4. As can be seen from fig. 5a and 5b, the second preset area in the target image is identical to the first preset area in the preset ticket template. In addition, the division ratio of each first preset area is preset to be set.
And obtaining the boundary position of each second preset area in the target image according to the dividing ratio of each first preset area in the preset bill template. Obtaining the transverse boundary position of the second preset area according to the dividing proportion of the transverse boundary of the first preset area has the following two conditions:
(1) The pixels of the bill image are the same as those of the preset bill template, and the position relation of the boundary of the first preset area corresponding to the title center point is the position of the boundary of the second preset area corresponding to the title center point.
(2) The bill image is different from the pixels of the preset bill template, and the pixel distance from the abscissa of the transverse boundary of the second preset area to the abscissa of the title center point is calculated through a formula of B/b=a/a, wherein a represents the actual length of the preset bill template, a represents the pixel distance corresponding to the actual length of the preset bill template, B represents the actual distance between a transverse boundary of the second preset area and the abscissa of the title center point, B represents the pixel distance between the transverse boundary and the abscissa of the title center point, and the transverse boundary of the second preset area can be determined through the pixel distance B.
The target image does not have a first rectangular region therein, and for the longitudinal boundaries of each second preset region, the longitudinal boundaries may be determined using the target image boundaries (even in combination with the center line of the target image). If the target image is a bill image with double-sided master information, the left side boundary and the center line of the target image can be used as the left and right longitudinal boundaries of each second preset area of the left side area range, and the right side boundary and the center line of the target image can be used as the left and right longitudinal boundaries of each second preset area of the right side area range; if the target image is a bill image with single-sided master information, the left and right boundaries of the target image can be used as the left and right longitudinal boundaries of each second preset area, so that the longitudinal boundaries of the first preset area in fig. 5a and the longitudinal boundaries of the second preset area in fig. 5b can overlap with the boundaries or the central lines of the images, thereby reducing the process of determining the longitudinal boundaries of the second preset area, simplifying the procedure, and improving the recognition efficiency.
S105, acquiring target areas where target information to be extracted is located from the second preset areas according to preset positioning rules.
The bill information extraction method of the embodiment of the invention is used for extracting one or more target information in the bill instead of all information, so that machine learning is needed to be carried out on the preset bill template in the early stage to obtain various information and information distribution conditions in the bill, and specific information is obtained in which first preset area, and as the layout of the target image is consistent with the layout of the preset bill template, the layout of the second preset area is consistent with the layout of the first preset area, and the second preset area is converted into the target image, so that specific target information can be preset in which second preset area.
The preset positioning rule defines that a specific target information item is in a specific second preset area, for example, the target information needed by us is the model of the vehicle, as can be seen from fig. 5a, the model is a first preset area on the left, correspondingly, for the target image, the target area of the model is a second preset area 1 on the left, and the information that the target area of the model is the second preset area 1 is summarized into the preset positioning rule, so that after the target image is divided into four first preset areas as shown in fig. 5b, the target preset area where the model is located can be determined as the first preset area 1 on the left of the target image according to the preset positioning rule.
S106, determining a target rectangular area where the target information is located according to the relative position relation between the second rectangular areas in the target area.
For the same type of bill, the relative positional relationship of each item of feature information with respect to the other item of feature information is different, and therefore, it is possible to determine which second matrix area the target information is specific in, by the relative positional relationship between the respective second rectangular areas.
In one embodiment, as shown in FIG. 6, step S106 includes sub-steps S161-S163.
And S161, acquiring the position coordinates of each second rectangular area in the target area.
S162, one of the second rectangular areas is used as a reference rectangular area one by one, and rectangular distribution conditions around the reference rectangular area are analyzed according to the relation between the position coordinates of other second rectangular areas and the position coordinates of the reference rectangular area.
And (3) taking a second rectangular area as a reference rectangular area, comparing the abscissa of the reference rectangular area with the abscissas of other second rectangular areas, comparing the ordinate of the reference rectangular area with the abscissas of other second rectangular areas, and analyzing the rectangular distribution condition of the second rectangular areas around the reference rectangular area according to the size relation between the abscissa of the reference rectangular area and the abscissas of other second rectangular areas and the size relation between the ordinate of the reference rectangular area and the ordinates of other second rectangular areas.
And S163, if the rectangular distribution situation around the reference rectangular area accords with a preset layout record, determining the reference rectangular area as the target rectangular area.
The machine learning preset bill template obtains the distribution situation around each item of information, the distribution situation (namely the preset layout record) of the second rectangular area around the preset recorded target information is obtained, one second rectangular area is sequentially taken as a reference rectangular area from top to bottom, the rectangular distribution situation is analyzed until the second rectangular area conforming to the preset layout record is found, and the second rectangular area conforming to the preset layout record is the target rectangular area, namely the position where the target information to be extracted is located.
S107, identifying target information in the target rectangular area.
The characteristic information in the target rectangular area is the target information which needs to be extracted, and text information in the target rectangular area can be rapidly identified by using a text identification technology.
According to the bill information extraction method provided by the embodiment of the invention, the target area is found by removing the master information and preliminary positioning of the information, the position of the target information is determined in the target area according to the relative position relation of each second rectangular area, and finally the target information is extracted. The master information is removed, a large amount of unnecessary information is removed, rectangular areas needing to be positioned are reduced, the information extraction efficiency is improved, the target information is initially positioned, the approximate position is found first, then the target information is accurately positioned from the approximate position, and the problem of inaccurate positioning caused by direct positioning of the target information in the excessive rectangular areas is avoided.
Fig. 7 is a flow chart of a bill information extracting method according to another embodiment of the present invention, the method includes the following steps S201 to S210:
s201, comparing the bill image with a preset bill template.
S202, if the text direction of the bill image is inconsistent with the text direction of the preset bill template, acquiring an included angle between the text direction of the bill image and the text direction of the preset bill template.
The preset bill template is a forward placed image, the text direction is transverse, and if the bill image is inclined at a certain angle, the text direction forms a certain included angle with the text direction of the preset bill template.
S203, rotating the bill image by the included angle so that the text direction of the bill image is consistent with the text direction of the preset bill template.
And aiming at the bill image with the inclined shooting angle, carrying out rotation correction on the included angle of the bill image, and then carrying out information identification.
The embodiment can still recognize the information of the bill image with the inclined shooting angle.
S204, dividing the position of each item of information in the bill image into rectangular areas, wherein the information in the bill image comprises master information and characteristic information, and the rectangular areas comprise a first rectangular area corresponding to the master information and a second rectangular area corresponding to the characteristic information.
S205, identifying a first rectangular area where the master information is located from all the rectangular areas.
S206, covering the first rectangular area to obtain a target image.
S207, dividing the target image into a plurality of second preset areas according to the dividing ratio of each first preset area in the preset bill template, wherein the information layout of the preset bill template is consistent with the information layout of the bill image, and each second preset area comprises a plurality of second rectangular areas.
S208, acquiring target areas where target information to be extracted is located from the second preset areas according to preset positioning rules.
S209, determining a target rectangular area where the target information is located according to the relative position relation between the second rectangular areas in the target area.
S210, identifying target information in the target rectangular area.
Steps S204 to S210 are the same as steps S101 to S107 in the foregoing embodiments, and reference may be made to the foregoing embodiments for specific description of steps S204 to S210, which are not repeated in the embodiments of the present invention.
Fig. 8 is a schematic block diagram of a bill information extracting apparatus 100 according to an embodiment of the present invention. The bill information extracting apparatus 100 includes a unit for performing the bill information extracting method described above, and may be configured in a desktop computer, a tablet computer, a portable computer, or the like. The bill information extracting apparatus 100 includes a first dividing unit 101, a first identifying unit 102, a covering unit 103, a second dividing unit 104, a first acquiring unit 105, a determining unit 106, and a second identifying unit 107.
The first dividing unit 101 is configured to divide a location of each item of information in a ticket image into rectangular areas, where the information in the ticket image includes master information and feature information, and the rectangular areas include a first rectangular area corresponding to the master information and a second rectangular area corresponding to the feature information;
the first identifying unit 102 is configured to identify a first rectangular area in which the master information is located from all the rectangular areas;
the covering unit 103 is used for covering the first rectangular area to obtain a target image;
the second dividing unit 104 is configured to divide the target image into a plurality of second preset areas according to a dividing ratio of each first preset area in a preset bill template, where an information layout of the preset bill template is consistent with an information layout of the bill image, and each second preset area includes a plurality of second rectangular areas;
the first obtaining unit 105 is configured to obtain, according to a preset positioning rule, a target area in which target information to be extracted is located from the plurality of second preset areas;
the determining unit 106 is configured to determine a target rectangular area in which the target information is located according to a relative positional relationship between each second rectangular area in the target area;
The second identifying unit 107 is configured to identify target information in the target rectangular area.
In one embodiment, the ticket image is a ticket image with single-sided master information. As shown in fig. 9, the first identifying unit 102 includes the following sub-units:
the first obtaining subunit 121 is configured to obtain an abscissa corresponding to each rectangular area.
A second acquiring subunit 122 is configured to acquire the minimum abscissa from all the obtained abscissas.
A first comparing subunit 123, configured to compare the minimum abscissa with other abscissas to obtain, as a target abscissa, an abscissa whose difference from the minimum abscissa is within a first error range.
A first determining subunit 124, configured to determine, as the first rectangular area, a rectangular area corresponding to the minimum abscissa and the target abscissa, respectively.
In another embodiment, the ticket image is a ticket image with double sided master information. As shown in fig. 10, the first identifying unit 102 includes the following sub-units:
a first identifying subunit 1201 is configured to identify, as a header rectangular area, a rectangular area where a header of the ticket is located in the ticket image.
A third obtaining subunit 1202 is configured to obtain a center line of the rectangular area of the title, where the center line is perpendicular to a text direction of the ticket title.
A dividing subunit 1203 is configured to divide the ticket image into a left area and a right area according to the center line.
A fourth obtaining subunit 1204, configured to obtain an abscissa corresponding to each rectangular area in the left area.
A fifth acquisition subunit 1205 is configured to acquire the minimum abscissa from all the obtained abscissas.
A second comparing subunit 1206 is configured to compare the minimum abscissa with other abscissas to obtain, as the first target abscissa, an abscissa having a difference from the minimum abscissa within the first error range.
The second determining subunit 1207 is configured to determine, as the first rectangular area corresponding to the master information of the left area, the rectangular area corresponding to the minimum abscissa and the first target abscissa.
A sixth obtaining subunit 1208 is configured to obtain an abscissa corresponding to the center line.
A seventh acquiring subunit 1209 is configured to acquire an abscissa corresponding to each rectangular area in the right area.
And a third comparing subunit 1210, configured to compare the abscissa corresponding to each rectangular area in the right area with the abscissa corresponding to the central line, so as to obtain, as a second target abscissa, an abscissa of the difference value of the abscissas corresponding to the central line within a second error range.
A third determining subunit 1211, configured to determine a rectangular area corresponding to the second target abscissa as a first rectangular area corresponding to the master information of the right area.
In an embodiment, as shown in fig. 11, the determining unit 106 includes the following subunits:
an eighth acquisition subunit 161 configured to acquire position coordinates of each second rectangular area in the target area.
And the analysis subunit 162 is configured to analyze the rectangular distribution situation around the reference rectangular area according to the relationship between the position coordinates of the other second rectangular areas and the position coordinates of the reference rectangular area by taking one of the second rectangular areas as the reference rectangular area one by one.
And a fourth determining subunit 163, configured to determine the reference rectangular area as the target rectangular area if the rectangular distribution situation around the reference rectangular area conforms to a preset layout record.
In another embodiment, as shown in fig. 12, the bill information extracting apparatus 100 further includes a comparing unit 108, a second acquiring unit 109, and a rotating unit 110.
The comparing unit 108 is configured to compare the bill image with a preset bill template before the first dividing unit divides the position of each item of information in the bill image into rectangular areas.
The second obtaining unit 109 is configured to obtain an included angle between the text direction of the ticket image and the text direction of the preset ticket template if the text direction of the ticket image is inconsistent with the text direction of the preset ticket template.
The rotation unit 110 is configured to rotate the bill image by the included angle so that the text direction of the bill image is consistent with the text direction of the preset bill template.
The bill information extracting apparatus 100 corresponds to the bill information extracting method, and in this embodiment, reference may be made to the method embodiment for the portion of the bill information extracting apparatus 100 that is not detailed, and the details are not described here.
The bill information extracting apparatus 100 described above may be implemented in the form of a computer program that can be run on a computer device as shown in fig. 13.
Fig. 13 is a schematic block diagram of a computer device 200 according to an embodiment of the present invention. The computer device 200 may be a terminal or a server, where the terminal may be an electronic device with a communication function, such as a smart phone, a tablet computer, a notebook computer, a desktop computer, a personal digital assistant, and a wearable device. The server may be an independent server or a server cluster formed by a plurality of servers.
The computer device 200 includes a processor 202, memory, and a network interface 205 connected via a system bus 201, wherein the memory may include a non-volatile storage medium 203 and an internal memory 204.
The non-volatile storage medium 203 of the computer device 200 may store an operating system 2031 and a computer program 2032, the computer program 2032, when executed, may cause the processor 202 to perform a ticket information extraction method. The internal memory 204 provides an environment for the execution of the computer program 2032 in the nonvolatile storage medium 203. The processor 202 of the computer device 200 is used to provide computing and control capabilities to support the operation of the entire computer device 200. The network interface 205 of the computer device 200 is used for network communication such as transmitting assigned tasks, receiving data, and the like.
Those skilled in the art will appreciate that the embodiment of the computer device shown in fig. 13 is not limiting of the specific construction of the computer device, and in other embodiments, the computer device may include more or less components than those shown, or certain components may be combined, or a different arrangement of components. For example, in some embodiments, the computer device may include only a memory and a processor, and in such embodiments, the structure and function of the memory and the processor are consistent with the embodiment shown in fig. 13, and will not be described again.
When the processor 202 runs the computer program 2032 in the nonvolatile storage medium 203, the processor 202 performs the steps of: dividing the position of each item of information in the bill image into rectangular areas, wherein the information in the bill image comprises master information and characteristic information, and the rectangular areas comprise a first rectangular area corresponding to the master information and a second rectangular area corresponding to the characteristic information; identifying a first rectangular area in which the master information is located from all the rectangular areas; covering the first rectangular area to obtain a target image; dividing the target image into a plurality of second preset areas according to the dividing proportion of each first preset area in the preset bill template, wherein the information layout of the preset bill template is consistent with the information layout of the bill image, and each second preset area comprises a plurality of second rectangular areas; acquiring target areas in which target information to be extracted is located from the second preset areas according to preset positioning rules; determining a target rectangular area where the target information is located according to the relative position relation between the second rectangular areas in the target area; feature information in the target rectangular region is identified.
In one embodiment, the ticket image is a ticket image with single-sided master information. The processor 202, when executing the step of identifying the first rectangular area where the master information is located from all the rectangular areas, specifically executes the following steps: acquiring the abscissa of each rectangular area; acquiring a minimum abscissa; comparing other abscissas with the minimum abscissas to obtain abscissas, the difference value of which is within a first error range from the minimum abscissas, as target abscissas; and determining the rectangular areas corresponding to the minimum abscissa and the target abscissa as the first rectangular areas.
In one embodiment, the ticket image is a ticket image with double sided master information. The processor 202, when executing the step of identifying the first rectangular area where the master information is located from all the rectangular areas, specifically executes the following steps: identifying a title rectangular area where a title of the ticket is located in the ticket image; acquiring a central line of the title rectangular area, wherein the central line is perpendicular to the text direction of the bill title; dividing the bill image into a left area and a right area according to the central line; acquiring the abscissa of each rectangular area in the left area; acquiring a minimum abscissa; comparing other abscissas with the minimum abscissas to obtain abscissas, the difference value of which is within a first error range from the minimum abscissas, as first target abscissas; determining the rectangular area corresponding to the minimum abscissa and the first target abscissa as a first rectangular area corresponding to the master information of the left area; acquiring the abscissa of the central line; acquiring the abscissa of each rectangular area in the right area; comparing the abscissa of each rectangular area in the right area with the abscissa of the central line to obtain an abscissa of which the difference value with the abscissa of the central line is in a second error range as a second target abscissa; and determining the rectangular area corresponding to the second target abscissa as a first rectangular area corresponding to the master information of the right area.
In one embodiment, when the step of determining the target rectangular area in which the target information is located according to the relative positional relationship between the second rectangular areas in the target area is performed, the processor 202 specifically performs the following steps: acquiring position coordinates of each second rectangular area in the target area; taking a second rectangular area as a reference rectangular area one by one, and analyzing rectangular distribution conditions around the reference rectangular area according to the relation between the position coordinates of other second rectangular areas and the position coordinates of the reference rectangular area; and if the rectangular distribution condition around the reference rectangular area accords with a preset layout record, determining the reference rectangular area as the target rectangular area.
In one embodiment, before executing the step of dividing the location of each item of information in the bill image into rectangular areas, the processor 202 specifically executes the following steps: comparing the bill image with a preset bill template; if the text direction of the bill image is inconsistent with the text direction of the preset bill template, acquiring an included angle between the text direction of the bill image and the text direction of the preset bill template; and rotating the bill image by the included angle so that the text direction of the bill image is consistent with the text direction of the preset bill template.
It should be appreciated that in embodiments of the present application, the processor 202 may be a central processing unit (Central Processing Unit, CPU), the processor 202 may also be other general purpose processors, digital signal processors (Digital Signal Processor, DSPs), application specific integrated circuits (Application Specific Integrated Circuit, ASICs), off-the-shelf programmable gate arrays (Field-Programmable Gate Array, FPGAs) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, or the like. Wherein the general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.
Those skilled in the art will appreciate that all or part of the flow in a method embodying the above described embodiments may be accomplished by computer programs instructing the relevant hardware. The computer program comprises program instructions, and the computer program can be stored in a storage medium, which is a computer readable storage medium. The program instructions are executed by at least one processor in the computer system to implement the flow steps of the embodiments of the method described above.
Accordingly, the present application also provides a storage medium storing one or more computer programs executable by one or more processors, the steps of: dividing the position of each item of information in the bill image into rectangular areas, wherein the information in the bill image comprises master information and characteristic information, and the rectangular areas comprise a first rectangular area corresponding to the master information and a second rectangular area corresponding to the characteristic information; identifying a first rectangular area in which the master information is located from all the rectangular areas; covering the first rectangular area to obtain a target image; dividing the target image into a plurality of second preset areas according to the dividing proportion of each first preset area in the preset bill template, wherein the information layout of the preset bill template is consistent with the information layout of the bill image, and each second preset area comprises a plurality of second rectangular areas; acquiring target areas in which target information to be extracted is located from the second preset areas according to preset positioning rules; determining a target rectangular area where the target information is located according to the relative position relation between the second rectangular areas in the target area; feature information in the target rectangular region is identified.
In one embodiment, the ticket image is a ticket image with single-sided master information. When the step of identifying the first rectangular area where the master information is located from all the rectangular areas is realized, the following steps are specifically realized: acquiring the abscissa of each rectangular area; acquiring a minimum abscissa; comparing other abscissas with the minimum abscissas to obtain abscissas, the difference value of which is within a first error range from the minimum abscissas, as target abscissas; and determining the rectangular areas corresponding to the minimum abscissa and the target abscissa as the first rectangular areas.
In one embodiment, the ticket image is a ticket image with double sided master information. When the step of identifying the first rectangular area where the master information is located from all the rectangular areas is realized, the following steps are specifically realized: identifying a title rectangular area where a title of the ticket is located in the ticket image; acquiring a central line of the title rectangular area, wherein the central line is perpendicular to the text direction of the bill title; dividing the bill image into a left area and a right area according to the central line; acquiring the abscissa of each rectangular area in the left area; acquiring a minimum abscissa; comparing other abscissas with the minimum abscissas to obtain abscissas, the difference value of which is within a first error range from the minimum abscissas, as first target abscissas; determining the rectangular area corresponding to the minimum abscissa and the first target abscissa as a first rectangular area corresponding to the master information of the left area; acquiring the abscissa of the central line; acquiring the abscissa of each rectangular area in the right area; comparing the abscissa of each rectangular area in the right area with the abscissa of the central line to obtain an abscissa of which the difference value with the abscissa of the central line is in a second error range as a second target abscissa; and determining the rectangular area corresponding to the second target abscissa as a first rectangular area corresponding to the master information of the right area.
In an embodiment, when the step of determining the target rectangular area where the target information is located according to the relative positional relationship between the second rectangular areas in the target area is implemented, the following steps are specifically implemented: acquiring position coordinates of each second rectangular area in the target area; taking a second rectangular area as a reference rectangular area one by one, and analyzing rectangular distribution conditions around the reference rectangular area according to the relation between the position coordinates of other second rectangular areas and the position coordinates of the reference rectangular area; and if the rectangular distribution condition around the reference rectangular area accords with a preset layout record, determining the reference rectangular area as the target rectangular area.
In an embodiment, before the step of dividing the position of each item of information in the bill image into rectangular areas, the specific implementation further includes the following steps: comparing the bill image with a preset bill template; if the text direction of the bill image is inconsistent with the text direction of the preset bill template, acquiring an included angle between the text direction of the bill image and the text direction of the preset bill template; and rotating the bill image by the included angle so that the text direction of the bill image is consistent with the text direction of the preset bill template.
The storage medium may be a U-disk, a removable hard disk, a Read-Only Memory (ROM), a magnetic disk, or an optical disk, or other various computer-readable storage media that can store program codes.
Those of ordinary skill in the art will appreciate that the elements and algorithm steps described in connection with the embodiments disclosed herein may be embodied in electronic hardware, in computer software, or in a combination of the two, and that the elements and steps of the examples have been generally described in terms of function in the foregoing description to clearly illustrate the interchangeability of hardware and software. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the solution. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present invention.
In the several embodiments provided by the present invention, it should be understood that the disclosed apparatus and method may be implemented in other manners. For example, the device embodiments described above are merely illustrative. For example, the division of each unit is only one logic function division, and there may be another division manner in actual implementation. For example, multiple units or components may be combined or may be integrated into another system, or some features may be omitted, or not performed.
The steps in the method of the embodiment of the invention can be sequentially adjusted, combined and deleted according to actual needs. The units in the device of the embodiment of the invention can be combined, divided and deleted according to actual needs. In addition, each functional unit in the embodiments of the present invention may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit.
The integrated unit may be stored in a storage medium if implemented in the form of a software functional unit and sold or used as a stand-alone product. Based on such understanding, the technical solution of the present invention is essentially or a part contributing to the prior art, or all or part of the technical solution may be embodied in the form of a software product stored in a storage medium, comprising several instructions for causing a computer device (which may be a personal computer, a terminal, a network device, etc.) to perform all or part of the steps of the method according to the embodiments of the present invention.
While the invention has been described with reference to certain preferred embodiments, it will be understood by those skilled in the art that various changes and substitutions of equivalents may be made and equivalents will be apparent to those skilled in the art without departing from the scope of the invention. Therefore, the protection scope of the invention is subject to the protection scope of the claims.
Claims (7)
1. The bill information extraction method is characterized by comprising the following steps:
dividing the position of each item of information in the bill image into rectangular areas, wherein the information in the bill image comprises master information and characteristic information, and the rectangular areas comprise a first rectangular area corresponding to the master information and a second rectangular area corresponding to the characteristic information;
identifying a first rectangular area in which the master information is located from all the rectangular areas;
covering the first rectangular area to obtain a target image;
dividing the target image into a plurality of second preset areas according to the dividing proportion of each first preset area in the preset bill template, wherein the information layout of the preset bill template is consistent with the information layout of the bill image, and each second preset area comprises a plurality of second rectangular areas;
acquiring target areas in which target information to be extracted is located from the second preset areas according to preset positioning rules;
determining a target rectangular area where the target information is located according to the relative position relation between the second rectangular areas in the target area;
identifying target information in the target rectangular area;
The bill image is a bill image with single-side master information;
the identifying the first rectangular area where the master information is located from all the rectangular areas comprises the following steps:
acquiring an abscissa corresponding to each rectangular area;
acquiring a minimum abscissa from all the obtained abscissas;
comparing the minimum abscissa with other abscissas to obtain an abscissa with a difference value within a first error range from the minimum abscissa as a target abscissa;
determining rectangular areas corresponding to the minimum abscissa and the target abscissa as the first rectangular areas;
the determining the target rectangular area where the target information is located according to the relative positional relationship between the second rectangular areas in the target area includes:
acquiring position coordinates of each second rectangular area in the target area;
one second rectangular area is taken as a reference rectangular area one by one, and rectangular distribution conditions around the reference rectangular area are analyzed according to the relation between the position coordinates of other second rectangular areas and the position coordinates of the reference rectangular area;
and if the rectangular distribution condition around the reference rectangular area accords with a preset layout record, determining the reference rectangular area as the target rectangular area.
2. The bill information extracting method according to claim 1, wherein the bill image is a bill image having double-sided master information;
the identifying the first rectangular area where the master information is located from all the rectangular areas comprises the following steps:
identifying a rectangular area where a bill title is located in the bill image as a title rectangular area;
acquiring a central line of the title rectangular area, wherein the central line is perpendicular to the text direction of the bill title;
dividing the bill image into a left area and a right area according to the central line;
acquiring an abscissa corresponding to each rectangular area in the left area;
acquiring a minimum abscissa from all the obtained abscissas;
comparing the minimum abscissa with other abscissas to obtain an abscissa with a difference value within a first error range from the minimum abscissa as a first target abscissa;
determining the rectangular area corresponding to the minimum abscissa and the first target abscissa as a first rectangular area corresponding to the master information of the left area;
acquiring an abscissa corresponding to the central line;
acquiring an abscissa corresponding to each rectangular area in the right area;
Comparing the abscissa corresponding to each rectangular area in the right area with the abscissa corresponding to the central line to obtain an abscissa of which the difference value of the abscissas corresponding to the central line is in a second error range as a second target abscissas;
and determining the rectangular area corresponding to the second target abscissa as a first rectangular area corresponding to the master information of the right area.
3. The method for extracting bill information according to claim 1, wherein before dividing the position of each item of information in the bill image into rectangular areas, further comprising:
comparing the bill image with a preset bill template;
if the text direction of the bill image is inconsistent with the text direction of the preset bill template, acquiring an included angle between the text direction of the bill image and the text direction of the preset bill template;
and rotating the bill image by the included angle so that the text direction of the bill image is consistent with the text direction of the preset bill template.
4. A bill information extracting apparatus, comprising:
the first dividing unit is used for dividing the position of each item of information in the bill image into rectangular areas, wherein the information in the bill image comprises master information and characteristic information, and the rectangular areas comprise a first rectangular area corresponding to the master information and a second rectangular area corresponding to the characteristic information;
The first identification unit is used for identifying a first rectangular area where the master information is located from all the rectangular areas;
a covering unit for covering the first rectangular region to obtain a target image;
the second dividing unit is used for dividing the target image into a plurality of second preset areas according to the dividing proportion of each first preset area in the preset bill template, wherein the information layout of the preset bill template is consistent with the information layout of the bill image, and each second preset area comprises a plurality of second rectangular areas;
the first acquisition unit is used for acquiring target areas where target information to be extracted is located from the second preset areas according to preset positioning rules;
a determining unit, configured to obtain position coordinates of each second rectangular area in the target area; one second rectangular area is taken as a reference rectangular area one by one, and rectangular distribution conditions around the reference rectangular area are analyzed according to the relation between the position coordinates of other second rectangular areas and the position coordinates of the reference rectangular area; if the rectangular distribution situation around the reference rectangular area accords with a preset layout record, determining the reference rectangular area as a target rectangular area;
A second identifying unit for identifying target information in the target rectangular area;
the bill image is a bill image with single-side master information;
the first recognition unit includes:
the first acquisition subunit is used for acquiring the abscissa corresponding to each rectangular area;
a second acquiring subunit configured to acquire a minimum abscissa from all the obtained abscissas;
a first comparing subunit, configured to compare the minimum abscissa with other abscissas to obtain an abscissa with a difference value from the minimum abscissa within a first error range as a target abscissa;
and the first determination subunit is used for determining the rectangular areas corresponding to the minimum abscissa and the target abscissa as the first rectangular area.
5. The ticket information extraction apparatus of claim 4 wherein the ticket image is a ticket image with double sided master information;
the first recognition unit includes:
the first identification subunit is used for identifying a rectangular area where the bill title is located in the bill image as a title rectangular area;
the third acquisition subunit is used for acquiring the central line of the rectangular area of the title, and the central line is perpendicular to the text direction of the bill title;
A dividing subunit for dividing the bill image into a left side area and a right side area according to the center line;
a fourth obtaining subunit, configured to obtain an abscissa corresponding to each rectangular area in the left area;
a fifth acquisition subunit configured to acquire a minimum abscissa from all the obtained abscissas;
a second comparing subunit, configured to compare the minimum abscissa with other abscissas to obtain an abscissa with a difference value from the minimum abscissa within a first error range as a first target abscissa;
a second determining subunit, configured to determine a rectangular area corresponding to the minimum abscissa and the first target abscissa as a first rectangular area corresponding to the master information of the left area;
a sixth obtaining subunit, configured to obtain an abscissa corresponding to the center line;
a seventh obtaining subunit, configured to obtain an abscissa corresponding to each rectangular area in the right area;
a third comparing subunit, configured to compare an abscissa corresponding to each rectangular area in the right area with an abscissa corresponding to the central line, so as to obtain an abscissa of a difference value of the abscissas corresponding to the central line within a second error range as a second target abscissas;
And the third determination subunit is used for determining the rectangular area corresponding to the second target abscissa as a first rectangular area corresponding to the master information of the right area.
6. A computer device comprising a memory, and a processor coupled to the memory; the memory is used for storing a computer program for realizing the bill information extraction method; the processor being adapted to run a computer program stored in the memory to perform the method of any one of claims 1 to 3.
7. A computer readable storage medium storing one or more computer programs executable by one or more processors to implement the method of any one of claims 1 to 3.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810921331.7A CN109255300B (en) | 2018-08-14 | 2018-08-14 | Bill information extraction method, bill information extraction device, computer equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810921331.7A CN109255300B (en) | 2018-08-14 | 2018-08-14 | Bill information extraction method, bill information extraction device, computer equipment and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109255300A CN109255300A (en) | 2019-01-22 |
| CN109255300B true CN109255300B (en) | 2023-12-01 |
Family
ID=65049185
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810921331.7A Active CN109255300B (en) | 2018-08-14 | 2018-08-14 | Bill information extraction method, bill information extraction device, computer equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109255300B (en) |
Families Citing this family (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110689010B (en) * | 2019-09-27 | 2021-05-11 | 支付宝(杭州)信息技术有限公司 | Certificate identification method and device |
| CN110781877B (en) * | 2019-10-28 | 2024-01-23 | 京东方科技集团股份有限公司 | Image recognition method, device and storage medium |
| CN110942054B (en) * | 2019-12-30 | 2023-06-30 | 福建天晴数码有限公司 | Page content identification method |
| CN111461100B (en) * | 2020-03-31 | 2023-04-28 | 重庆农村商业银行股份有限公司 | Bill identification method and device, electronic equipment and storage medium |
| CN111652141B (en) * | 2020-06-03 | 2023-05-05 | 广东小天才科技有限公司 | Question segmentation method, device, equipment and medium based on question numbers and text lines |
| CN111680686B (en) * | 2020-06-08 | 2023-05-12 | Oppo（重庆）智能科技有限公司 | Signboard information identification method, device, terminal and storage medium |
| CN112732955A (en) * | 2021-03-31 | 2021-04-30 | 国网浙江省电力有限公司 | Financial certificate storage and recording method in standard cost accounting |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH10124610A (en) * | 1996-10-23 | 1998-05-15 | Nec Corp | Optical character reading device |
| CN101887413A (en) * | 2009-05-14 | 2010-11-17 | 北大方正集团有限公司 | Structure processing method and system of plate type table |
| JP2016167159A (en) * | 2015-03-09 | 2016-09-15 | キヤノン株式会社 | Information processing apparatus |
| CN107798299A (en) * | 2017-10-09 | 2018-03-13 | 平安科技（深圳）有限公司 | Billing information recognition methods, electronic installation and readable storage medium storing program for executing |
| CN107808154A (en) * | 2017-12-08 | 2018-03-16 | 上海慧银信息科技有限公司 | The method and apparatus for extracting cash register billing information |
| CN108052662A (en) * | 2017-12-29 | 2018-05-18 | 上海器魂智能科技有限公司 | Web-based assembly layout method, apparatus, equipment and storage medium |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP4603658B2 (en) * | 2000-07-07 | 2010-12-22 | キヤノン株式会社 | Image processing apparatus, image processing method, and storage medium |
| US8300928B2 (en) * | 2008-01-25 | 2012-10-30 | Intermec Ip Corp. | System and method for locating a target region in an image |
| US20120278154A1 (en) * | 2011-04-28 | 2012-11-01 | Microsoft Corporation | Marketing inventory based on spoilage |
| US10235585B2 (en) * | 2016-04-11 | 2019-03-19 | The Nielsen Company (US) | Methods and apparatus to determine the dimensions of a region of interest of a target object from an image using target object landmarks |
-
2018
- 2018-08-14 CN CN201810921331.7A patent/CN109255300B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH10124610A (en) * | 1996-10-23 | 1998-05-15 | Nec Corp | Optical character reading device |
| CN101887413A (en) * | 2009-05-14 | 2010-11-17 | 北大方正集团有限公司 | Structure processing method and system of plate type table |
| JP2016167159A (en) * | 2015-03-09 | 2016-09-15 | キヤノン株式会社 | Information processing apparatus |
| CN107798299A (en) * | 2017-10-09 | 2018-03-13 | 平安科技（深圳）有限公司 | Billing information recognition methods, electronic installation and readable storage medium storing program for executing |
| CN107808154A (en) * | 2017-12-08 | 2018-03-16 | 上海慧银信息科技有限公司 | The method and apparatus for extracting cash register billing information |
| CN108052662A (en) * | 2017-12-29 | 2018-05-18 | 上海器魂智能科技有限公司 | Web-based assembly layout method, apparatus, equipment and storage medium |
Non-Patent Citations (2)
| Title |
|---|
| 曹玉东 等.银行票据图像中的文本区域定位算法.《辽宁工业大学学报(自然科学版)》.2017,第37卷(第5期),第281-301页. * |
| 谢文彬 等.一种基于结构特征的票据印刷号码识别方法.《包装工程》.2018,第39卷(第1期),第202-206页. * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109255300A (en) | 2019-01-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109255300B (en) | Bill information extraction method, bill information extraction device, computer equipment and storage medium | |
| US20210192202A1 (en) | Recognizing text in image data | |
| CN109492643B (en) | Certificate identification method and device based on OCR, computer equipment and storage medium | |
| US10303968B2 (en) | Method and apparatus for image recognition | |
| CN110705405A (en) | Target labeling method and device | |
| US9679354B2 (en) | Duplicate check image resolution | |
| US20180253852A1 (en) | Method and device for locating image edge in natural background | |
| CN111325104A (en) | Text recognition method, device and storage medium | |
| CN110751149A (en) | Target object labeling method and device, computer equipment and storage medium | |
| CN112101386B (en) | Text detection method, device, computer equipment and storage medium | |
| CN114038004A (en) | Certificate information extraction method, device, equipment and storage medium | |
| CN111160395A (en) | Image recognition method and device, electronic equipment and storage medium | |
| CN111275051A (en) | Character recognition method, character recognition device, computer equipment and computer-readable storage medium | |
| WO2018121414A1 (en) | Electronic device, and target image recognition method and apparatus | |
| US20170287138A1 (en) | Apparatus and method for dividing image into regions | |
| RU2633182C1 (en) | Determination of text line orientation | |
| CN113557520A (en) | Character processing and character recognition method, storage medium and terminal device | |
| CN112396047B (en) | Training sample generation method and device, computer equipment and storage medium | |
| CN112597940B (en) | Certificate image recognition method and device and storage medium | |
| CN113780116A (en) | Invoice classification method and device, computer equipment and storage medium | |
| JP4967045B2 (en) | Background discriminating apparatus, method and program | |
| US11335108B2 (en) | System and method to recognise characters from an image | |
| CN111241974B (en) | Bill information acquisition method, device, computer equipment and storage medium | |
| CN114494751A (en) | License information identification method, device, equipment and medium | |
| JP6175904B2 (en) | Verification target extraction system, verification target extraction method, verification target extraction program |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |