TWI737168B - Detecting system for electronic invoice certificate and method thereof - Google Patents

Abstract

The present invention mainly provides a detection system for an electronic invoice proof sheet, which obtains an invoice graphic by a graphic acquisition device, reads an invoice content in the invoice graphic through a graphic reading unit, and then uses the detection unit according to the invoice content Or the size and specifications of the invoice graphics are tested to generate plural test information, a test result is generated based on each test information, and finally the test result is output through an output module to provide a complete and automatic inspection electronic invoice certificate The method of linking further improves the problem that the prior art must use manual detection or cannot be effectively detected through automatic identification.

Description

Detection system and method of electronic invoice certificate

The present invention provides a detection system and method for the electronic invoice certification sheet, and particularly refers to a detection system for the electronic invoice certification sheet for detecting whether the content and size of the electronic invoice conform to the standard specifications.

In response to the reduced use of paper and to meet the environmental requirements of energy saving and carbon reduction, government agencies have fully implemented electronic invoices since 2010, and have gradually phased out the issuance of paper invoices, so that they can gradually move towards the goal of paperless invoices.

In recent years, in order to implement the completeness of the electronic consumption mechanism, to avoid missing or misinvoicing, or reuse of invoices by the industry, one-dimensional barcodes and two-dimensional barcodes have been added to the new electronic invoice certification link. To facilitate subsequent testing and auditing.

However, nowadays, when checking the content of the electronic invoice certificate, it usually only uses the background to obtain the information of the electronic invoice, and then uses the application on the mobile phone to scan the one-dimensional barcode or the two-dimensional barcode to perform related detection actions. Provides the function of automatic detection, and cannot detect the version of the two-dimensional barcode, the error tolerance standard, or the size and text content of the electronic invoice. Therefore, how to provide a complete system that can automatically detect the electronic invoice certificate is a problem that must be solved today.

In order to solve the aforementioned problems, the purpose of the present invention is to provide an electronic invoice certificate Minglian’s detection system uses a graphics reading unit to read the invoice content in the invoice graphics, and then uses the detection unit to detect the invoice content or invoice graphics to solve the problems of the prior art.

To achieve the above-mentioned purpose, the present invention provides an electronic invoice verification page detection system, which includes: a graphic acquisition device configured to obtain an invoice graphic from an electronic invoice; a graphic reading unit that is connected to the The graphic acquisition device is connected, the graphic reading unit reads the invoice graphic, and generates an invoice content according to the invoice graphic; a detection unit is connected to the graphic reading device, the detection unit is based on a set one The invoice specification detects the invoice content or the invoice graphic, and generates a detection result based on the plural detection information generated after detecting the invoice content or the invoice graphic; an output module, which is connected with the detection module, and the output module The group outputs the test result to the outside.

Preferably, the graphic reading unit includes: a one-dimensional bar code reading module, which is configured to read the one-dimensional bar code in the invoice graphic and parse the one-dimensional bar code into a one-dimensional string; A two-dimensional barcode reading module, which is set to read a plurality of two-dimensional barcodes in the invoice graph, and to parse each two-dimensional barcode into a two-dimensional character string; a text recognition module, which is set To read a text graphic in the invoice graphic, and the text recognition module textualizes the text graphic to obtain a text content, wherein the invoice content includes the one-dimensional string, the two-dimensional string, and the text content .

Preferably, when the text recognition module obtains the text content, the text recognition module determines whether the text content meets the plural specification text set in the text recognition module, and if the text content lacks any of the specifications When writing text, the text recognition module re-reads the text graphic and textualizes the text graphic by block reading. When the text graphic is completely textualized, the text recognition module is stacked in order The text of each block obtained after textualization The text content of each block is compared with the text content of each block to determine whether the text content of each block meets the specified text; wherein, if the text recognition module determines the text content of each block When the text content meets the standard text, the text recognition module outputs the text content composed of each block; wherein, if the text recognition module determines that the text content of each block does not meet the standard text, The text recognition module reads the text graphics again, and textizes the text graphics in a block reading manner.

Preferably, the detection unit includes: a specification analysis module connected to the graph reading module, and the specification analysis module detects the text content and an item in the invoice specification according to a text specification in the invoice specification. The one-dimensional specification detects the one-dimensional character string and the one-dimensional specification in the invoice specification detects each of the two-dimensional character strings to generate a plurality of the detection information; a size detection module, which is similar to the graphics reading module Link to obtain the invoice graphic, the size detection module detects the invoice graphic according to a size specification in the invoice specification to generate the detection information; a specification setting module, which is associated with the specification analysis module and the size detection The modules are connected, and the specification setting module is set to set the invoice specification so as to provide the invoice specification to the specification analysis module and the size detection module.

Preferably, when the specification analysis module detects the text content according to the text specification, the specification analysis module is based on the figure or text of the business person’s identification stamp, the text of the electronic invoice certificate, the text of the year and The subsequent character track number belongs to the Republic of China year and period, the unified invoice character track number, the recording method of the transaction date and time, the text of the random code and the number of subsequent numbers, the total text and subsequent numbers, the seller or the seller number The number of text and subsequent digits to detect the text content.

Preferably, when the specification analysis module detects the text content according to the text specification At the time, the specification analysis module further detects the content of the text based on the text compiled by the buyer or the buyer and the number of subsequent digits, or the record items of the transaction details.

Preferably, when the specification analysis module detects the one-dimensional character string according to the one-dimensional specification, the specification analysis module is based on whether the one-dimensional character string conforms to the recording method of a code type, and the one-dimensional character string Is the first 5 codes of the period, the last 4 codes a random code, whether the middle 10 codes are the uniform invoice character track number, and whether the period, the random code and the uniform invoice character track number conform to the period of the text content Do not, the random code and the uniform invoice character track number to detect the one-dimensional character string; the barcode type is Code 39 (Code 39).

Preferably, when the specification analysis module detects the two-dimensional character string according to the two-dimensional specification, the specification analysis module captures a piece of version information in the neighboring locations of the plural positioning patterns in the two-dimensional barcode And an error-proof standard information, the specification analysis module then detects the version information and the error-proof standard information according to the two-dimensional specification.

Preferably, when the specification analysis module detects the two-dimensional character string according to the two-dimensional specification, the specification analysis module detects one of the two-dimensional character string according to the two-dimensional specification with an error-proof standard information The number of sales digits before the 77th code, encrypted verification information, and the unified invoice track number that is the same as the text, transaction date and time, random code numbers, total numbers, seller number numbers, and buyers Integrated digital inspection.

Preferably, the encrypted verification information is a string combination of the uniform invoice track number and the random code, and is converted by an encryption process to generate the encrypted verification information.

Preferably, the specification analysis module further detects the transaction details after the 78th code of the two-dimensional string or/and the record items of the transaction details of another two-dimensional string according to the two-dimensional specification Whether it conforms to the transaction details recorded in the text content.

Preferably, when the size detection module detects the invoice graphic according to the size specification, the size detection module captures a width or a length of the invoice graphic to measure according to the width or the length Whether it meets the size specification; if it does, the size detection module will then detect the unmeasured width or length of the invoice graphic according to the ratio recorded in the size specification, as well as the one-dimensional barcode, The position, width and length of the two-dimensional bar code and the text graphic; if not, the size detection module determines that the invoice graphic does not meet the size specification.

Preferably, the detection result includes at least one qualified item that meets the invoice specification, at least one error item that does not meet the invoice specification, or at least one suggestion item that provides correction suggestions for the error item.

Preferably, the invoice graphic includes a graphic on the front of the invoice, a graphic on the back of the invoice, a detailed invoice graphic, or a combination of any two or more thereof.

Preferably, the invoice content includes at least a one-dimensional bar code, a two-dimensional bar code, a text content, or a combination of two or more thereof.

In order to solve the aforementioned problems, another object of the present invention is to provide a detection method for the electronic invoice verification sheet, which discloses a step method in which the graphics reading unit reads the invoice graphics, and the detection unit detects the invoice graphics or the content of the invoice. Advance to solve the problems of the prior art.

In order to achieve the above-mentioned other object, the present invention provides a method for detecting an electronic invoice certificate, which includes: using a graphic obtaining device to obtain an invoice graphic from an electronic invoice; using a graphic reading module to read the invoice graphic. Invoice graphics, and generate an invoice content based on the invoice graphics; use a detection unit to detect the invoice content or the invoice according to a set invoice specification Graphics, and generate a detection result based on the plurality of detection information generated after detecting the invoice content or the invoice graphics; an output module is used to output the detection result to the outside.

Preferably, when the graphic reading unit reads the invoice graphic, it includes the following steps: using a one-dimensional barcode reading module to read the one-dimensional barcode in the invoice graphic, and the one-dimensional barcode Parse into a one-dimensional character string; use a two-dimensional barcode reading module to read the plural two-dimensional barcodes in the invoice graphics, and parse each two-dimensional barcode into a two-dimensional character string; use a text recognition module The group reads a text graphic in the invoice graphic, and the text recognition module textualizes the text graphic to obtain a text content; wherein the invoice content includes the one-dimensional string, the two-dimensional string, and the text content .

Preferably, when the text recognition module obtains the text content, the text recognition module determines whether the text content meets the plural specification text set in the text recognition module, and if the text content lacks any of the specifications When writing text, the text recognition module re-reads the text graphic and textualizes the text graphic by block reading. When the text graphic is completely textualized, the text recognition module is stacked in order After textualization, the text content of each block is obtained, and the text content of each block is compared with the text content of each block according to the comparison of the text content of each block to determine whether the text content of each block meets the specification text; wherein, If the text recognition module determines that the text content of each block conforms to the standard text, the text recognition module outputs the text content composed of each block; wherein, if the text recognition module determines each block When the text content of does not meet the standard text, the text recognition module re-executes the steps of re-reading the text graphic and texturing the text graphic in a block-based reading manner.

Preferably, when the detection unit detects the invoice content or the invoice graph according to the invoice specification, it includes the following steps: using a specification analysis module according to the invoice specification A text specification detects the text content, a one-dimensional specification in the invoice specification detects the one-dimensional character string, and a two-dimensional specification in the invoice specification detects each of the two-dimensional character strings to generate a plurality of the detection information; The size detection module detects the invoice graphic according to a size specification in the invoice specification to generate the detection information; uses a specification setting module to set the invoice specification to provide the invoice specification to the specification analysis module and the size Detection module.

Preferably, when the size detection module detects the invoice graphic according to the size specification, the size detection module captures a width or a length of the invoice graphic to measure according to the width or the length Whether it meets the size specification; if it does, the size detection module will then detect the unmeasured width or length of the invoice graphic according to the ratio recorded in the size specification, as well as the one-dimensional barcode, The position, width and length of the two-dimensional bar code and the text graphic; if not, the size detection module determines that the invoice graphic does not meet the size specification.

100: detection system

10: Graphic acquisition device

11: Invoice graphics

20: Graphic reading unit

201: Invoice content

21: One-dimensional barcode reading module

211: One-dimensional string

22: Two-dimensional barcode reading module

221: two-dimensional string

23: text recognition module

231: text content

232: Specification text

30: detection unit

301: Detection Information

302: test result

31: Specification Analysis Module

32: size detection module

33: Specification setting module

331: invoice specifications

3311: text specifications

3312: One-dimensional specifications

3313: Two-dimensional specifications

3314: size specifications

40: output module

S11~S14, S121~S123, S1231~S1234, S131~S133: step flow

Figure 1 is a block diagram of the detection system of the present invention;

Figure 2 is a block diagram of the data transfer of the detection system of the present invention;

Figure 3 is a flow chart of the detection steps of the electronic invoice certificate of the present invention;

FIG. 4 is a flowchart of the steps of reading the invoice graphics by the graphics reading unit of the present invention;

FIG. 5 is a flowchart of the steps for detecting text content by the text recognition module of the present invention;

Fig. 6 is a flow chart of the steps of the detection unit of the present invention to detect the content of the invoice according to the invoice specification.

Specific embodiments will be described below to illustrate the implementation of the present invention, but they are not It is not intended to limit the scope of protection of the present invention.

Please refer to FIG. 1 and FIG. 2, which are the block diagram of the detection system and the block diagram of the data transmission of the detection system of the present invention. As shown in the figure, the detection system 100 of the electronic invoice verification sheet of the present invention is mainly composed of a graphic acquisition device 10, a graphic reading unit 20, a detection unit 30, and an output module 40.

The graphic obtaining device 10 is configured to obtain an invoice graphic 11 from an electronic invoice (which can be an electronic invoice in paper or an electronic graphic file). The graphic obtaining device 10 may specifically be a camera, a scanner, or other capable of obtaining the invoice graphic. The device of 11, wherein the invoice graphic 11 includes a graphic on the front of the invoice, a graphic on the back of the invoice, a detailed invoice graphic, or a combination of any two or more thereof.

The graphic reading unit 20 is mainly used to read information in a graphic, and the read information is analyzed or textualized by the modules provided. When the graphic reading unit 20 and When the graphic obtaining device 10 is connected, the graphic reading unit 20 will read the invoice graphic 11 and generate an invoice content 201 based on the invoice graphic 11. Specifically, when the graphic reading unit reads 20 to the invoice graphic 11, since the electronic invoice usually contains a one-dimensional bar code, a two-dimensional bar code and text, the graphic reading unit 20 will use one by one. The one-dimensional barcode reading module 21 reads the one-dimensional barcode in the invoice graphic 11, and parses the one-dimensional barcode into a one-dimensional character string 211, and the one-dimensional barcode reading module 22 reads the invoice graphic 11 A plurality of two-dimensional barcodes in the two-dimensional barcode, and each two-dimensional barcode is parsed into a two-dimensional character string 221, a text recognition module 23 reads a text graphic in the invoice graphic 11, and the text recognition module 23 is a text The text graphics are transformed to obtain a text content 231, where the one-dimensional character string 211 and the two-dimensional character string 221 may specifically be character strings formed by combining English and numbers. It can be seen that the invoice content 201 includes the one-dimensional character string 211, the two-dimensional character string 221, the text content 231, or any combination of two or more thereof.

When the text recognition module 23 textualizes the text graphic, the text recognition module 23 can, for example, use Optical Character Recognition (OCR) the text graphic to obtain the text content 231 in the text graphic. However, if only the optical character recognition is used to obtain the text content 231, it is difficult to recognize the text in the text graphics 100%. Therefore, after the text recognition module 23 obtains the text content 231, It will determine whether the text content 231 meets the plural specification text 232 set in the text recognition module 23. Each specification text 232 is used to determine the content that should be recorded on the electronic invoice. If the text content 231 is missing When any of the standard characters 232, the character recognition module 23 re-reads the character graphics, and characterizes the character graphics in a block reading manner. When the character graphics are completely characterized, the character recognition The module 23 sequentially stacks the text content 231 of each block obtained after literalization, and compares each standard text 232 with the text content 231 of each block to determine the text of each block Whether the content 231 conforms to each standard text 232.

Since the specifications or dimensions of the electronic invoice are often changed based on the revision of the key points of the electronic invoice implementation, the detection unit 30 is mainly used to detect whether the electronic invoice meets the invoice specifications and the size of the invoice formulated by the key points of the current electronic invoice implementation operation. , Version information and fault tolerance standards. When the detection unit 30 is connected to the graphic reading device 10, the detection unit 30 detects the invoice content 201 or the invoice graphic 11 according to a set invoice specification 331, and detects the invoice content 201 or the invoice according to The plural detection information 301 generated after the graph 11 generates a detection result 302, where the detection result 302 may include at least one qualified item that meets the invoice specification 331, at least one error item that does not meet the invoice specification, or The erroneous item provides at least one suggested item for correcting suggestions, where the qualified item refers to an item that meets the invoice specification 331, the erroneous item is an item that does not meet the invoice specification 331, and the suggested item is to provide Suggestions for correction of specific items (regardless of whether it is the qualified item or the wrong item).

Specifically, the detection unit 30 uses a specification analysis module 31, a size detection module 32, and a specification setting module 33 included in the detection unit 30 to perform detection. When the detection unit 30 detects the invoice For content 201, the specification analysis module 31 detects the text content 231 according to a text specification 3311 in the invoice specification 331, a one-dimensional specification 3312 in the invoice specification 331 to detect the one-dimensional string 211, and the invoice A two-dimensional specification 3313 in the specification 331 detects each of the two-dimensional character strings 221 to generate a plurality of the detection information 301.

When the specification analysis module 31 detects the one-dimensional character string 211, the specification analysis module 31 mainly detects whether the one-dimensional character string 211 conforms to the recording method of a code type, and the first 5 codes of the one-dimensional character string 211 Whether it’s a period, whether the last 4 codes are random codes, whether the middle 10 codes are unified invoice character track numbers, and whether the period, the random code, and the unified invoice character track number conform to the period and the text content 231 Random code and the uniform invoice character track number; the barcode types include Code 39 (Code 39), Code 93 (Code 93), Code 11 (Code 11), MSI barcode, China Post barcode (Toshiba code) ), three-two code (Code 32), or interleaved twenty-five bar codes, etc. In this embodiment, the bar code type is specifically code 39 (Code 39).

When the specification analysis module 31 detects the two-dimensional character string 221, the specification analysis module 31 captures a plurality of positioning patterns in the two-dimensional bar code (for example, the three corners of the two-dimensional bar code are provided with "back" The positioning pattern) is a version information and an error-proof standard information in the surrounding area, the specification analysis module then detects the version information and the error-proof standard information according to the two-dimensional specification, because the two-dimensional bar code has a variety of Versions and various error-proofing standards. In this way, the version information and the error-proofing standard information retrieved by the specification analysis module can determine whether the version and error-proofing standards used by the two-dimensional barcode are consistent with the implementation of the current electronic invoice Operation points or what the user wants to use Version and error-proofing standards. In this embodiment, the two-dimensional specification detects the version information and the error-proof standard information according to the version of the two-dimensional barcode V6 (41x41) or higher, and the error-proof standard of Level L (tolerance rate of 7%) or higher.

When the specification analysis module 31 detects the two-dimensional character string 221, the specification analysis module 31 then detects one of the two-dimensional characters according to the two-dimensional specification 3313 with the error-proof standard information (tolerance rate of 7%). The number of sales digits before the 77th code of the string, the encrypted verification information, and the unified invoice track number that is the same as the text, the transaction date and time, the number of the random code, the total number, and the number of the seller number And the number compiled by the buyer. In this embodiment, the number of sales figures is the 8-digit untaxed amount, and the amount is converted into hexadecimal digits; the encrypted verification information (24 digits) is the unified invoice track number After the random code is combined in a string, it is encrypted according to an encryption procedure (for example, Advanced Encryption Standard (AES)), and is converted into the encrypted verification information by an encoding procedure (for example, Base 64); The invoice track number has 10 digits; the first three digits of the transaction date and time are the year of the Republic of China, the middle two digits are the month, and the last two digits are the date; the total number of digits is the 8-digit tax-included amount, and Convert the amount to hexadecimal; the number (8 digits) of the seller’s serial number is the uniform number of the seller; the number (8 digits) compiled by the buyer is the uniform serial number of the buyer, if the buyer is general consumption When it is, it is recorded as 8 zeros.

When the specification analysis module 31 detects the two-dimensional character string 221, the specification analysis module 31 further detects the transaction details after the 78th code of the two-dimensional character string 221 according to the two-dimensional specification 3313 or/and the other two Whether the record items in the transaction details of the dimension string 221 match the record items in the transaction details in the text content. Specifically, when there is too much information about the transaction, if only one of the two-dimensional barcodes is used to record the transaction details after the 78th digit, it may not be possible to record all the transaction information. In this case, two two-dimensional barcodes will be provided on the general electronic invoice. One of the two-dimensional string 221 records the information before the 77th code and the transaction details after the 78th code, and the other The two-dimensional character string 221 will record one of the transaction details that cannot be recorded after the 78th code of the two-dimensional barcode. In this way, the 31 sets of the specification analysis module can detect one of the two-dimensional character string 221 according to the two-dimensional specification 3313 After the 78th code of or/and whether the transaction details of the other two-dimensional string 221 match the transaction details in the text content 231.

The size detection module 32 detects the invoice graphic 11 according to a size specification 3314 in the invoice specification 331 to generate the detection information 301, and when the size detection module 32 performs detection, the size detection module 32 First capture a width or a length of the invoice graphic 11 (the following is to capture the width as an example), to measure whether the width meets the width limit recorded in the size specification 3314 (which can be a range value) Or a numerical value. In this embodiment, the width limit is 5.7 cm). If it meets the requirements, the size detection module 32 will follow the scale described in the size specification 3314 (for example, 3:4, 9:16) Detect the unmeasured length of the invoice graphic 11 (for example, 9 cm, the error value is ±3%), and the one-dimensional barcode (for example, the height must be 0.5 cm or more) and the two-dimensional barcode in the invoice graphic 11 And the position, width, and length of the text and graphics (for example, the width and length of the two-dimensional barcode are 1.5 cm each), and the width is used as the measurement reference; but if it does not match, the size detection module 32 determines The invoice graphic 11 does not conform to the size specification 3314, so when one of the inspection information 301 is output, the inspection information 301 will contain the error item that records the size inspection as an error, and can be further generated for size inspection Suggested items.

The specification setting module 33 is mainly set up to allow the user to set the invoice specification 331, or to receive the invoice specification 331 from outside, so as to provide the invoice specification 331 to the specification analysis module 31 and the size inspection module. Group 32.

The output module 40 is used to transmit information to the outside, so that an external computing device, processing device, analysis device, or receiving device can receive the output information from the output module 40, so that when the output When the module 40 is connected to the detection unit 30, the output module 40 will output the detection result 302 received from the detection unit 30 to the outside, so as to provide the electronic invoice developer to confirm and print the electronic Whether the invoice complies with the regulations, or provide an auxiliary tool for the staff of the IRS to review the electronic invoice application.

Please refer to Figures 3 to 6, which are the flow chart of the detection steps of the electronic invoice certificate sheet of the present invention, the flow chart of the steps for the graphics reading unit to read the graphics of the invoice, the flow chart of the steps for the text recognition module to detect the text content, and the detection A flow chart of the steps for the unit to check the contents of the invoice according to the invoice specifications. As shown in the figure, the present invention further discloses the detection method of the electronic invoice verification sheet. The detection method is applied to the above-mentioned electronic invoice verification sheet detection system, which mainly includes:

S11: Use a graphic obtaining device to obtain an invoice graphic from an electronic invoice;

S12: Use a graphic reading module to read the invoice graphic, and generate an invoice content based on the invoice graphic;

S13: Use a detection unit to detect the invoice content or the invoice graphic according to a set invoice specification, and generate a detection result based on the plural detection information generated after detecting the invoice content or the invoice graphic;

S14: Use an output module to output the detection result to the outside.

When the graphic reading unit reads the invoice graphic, it will include the following steps:

S121: Use the one-dimensional barcode reading module to read the one-dimensional barcode in the invoice graph, and parse the one-dimensional barcode into a one-dimensional string;

S122: Use a one- and two-dimensional barcode reading module to read the plural two-dimensional barcodes in the invoice graph, and parse each of the two-dimensional barcodes into a two-dimensional character string respectively;

S123: Use a text recognition module to read a text graphic in the invoice graphic, and the text recognition module textualizes the text graphic to obtain a text content;

Wherein, the invoice content includes the one-dimensional character string, the two-dimensional character string, and the text content.

When the text recognition module obtains the text content, it will perform the following steps:

S1231: The text recognition module judges whether the text content meets the plural specification text set in the text recognition module;

S1232: If the text content lacks any of the standard text, the text recognition module re-reads the text graphic, and textualizes the text graphic in a block reading manner;

S1233: When the text graphics are completely textualized, the text recognition module sequentially stacks the text content of each block obtained after textualization;

S1234: Determine whether the text content of each block meets the text content of each standard according to the comparison of the text of each specification with the text content of each block;

Wherein, if the text recognition module determines that the text content of each block conforms to the standard text, the text recognition module outputs the text content composed of each block;

Wherein, if the text recognition module determines that the text content of each block does not meet the specifications of the text, the text recognition module re-reads the text graphics and textualizes it in a block-based reading method. The steps of the text graphics.

When the detection unit detects the content of the invoice according to the invoice specifications, it includes: Next steps:

S131: Use a specification analysis module to detect the text content according to a text specification in the invoice specification, detect the one-dimensional character string in the one-dimensional specification in the invoice specification, and detect each of the two-dimensional specifications in the invoice specification. A two-dimensional character string to generate a plurality of the detection information;

S132: Use a size detection module to detect the invoice graphics according to a size specification in the invoice specification to generate the detection information;

S133: Use a specification setting module to set the invoice specification, so as to provide the invoice specification to the specification analysis module and the size detection module.

Wherein, when the size detection module detects the invoice graphic according to the size specification, the size detection module captures a width or a length of the invoice graphic to measure whether the width or length conforms to The size specification; if it meets the specifications, the size detection module will then detect the unmeasured width or length of the invoice graphic according to the ratio recorded in the size specification, and the one-dimensional barcode and the two-dimensional barcode in the invoice graphic. Dimension bar code and the position, width and length of the text graphic; if not, the size detection module will determine that the invoice graphic does not meet the size specification.

With the technical content contained in the above-mentioned embodiments, the present invention can provide a complete and automatic way of detecting the electronic invoice certificate, thereby improving the problem that the prior art must use manual detection or cannot effectively detect through automatic identification.

The above detailed description is a specific description of a feasible embodiment of the present invention, but this embodiment is not intended to limit the scope of the present invention. Any equivalent implementation or modification that does not deviate from the technical spirit of the present invention shall include In the scope of the patent in this case.

100: detection system

10: Graphic acquisition device

11: Invoice graphics

20: Graphic reading unit

201: Invoice content

21: One-dimensional barcode reading module

211: One-dimensional string

22: Two-dimensional barcode reading module

221: two-dimensional string

23: text recognition module

231: text content

232: Specification text

30: detection unit

301: Detection Information

302: test result

31: Specification Analysis Module

32: size detection module

33: Specification setting module

331: invoice specifications

3311: text specifications

3312: One-dimensional specifications

3313: Two-dimensional specifications

3314: size specifications

40: output module

Claims (16)

A detection system for an electronic invoice certificate, which includes: a graphic acquisition device configured to obtain an invoice graphic from an electronic invoice; a graphic reading unit connected to the graphic acquisition device, and the graphic reading unit Reading the invoice graphic, and generating an invoice content based on the invoice graphic; a detection unit connected to the graphic reading device, the detection unit detecting the invoice content or the invoice graphic according to a set invoice specification, And generate a detection result according to the plural detection information generated after detecting the invoice content or the invoice graph; and an output module connected to the detection module, and the output module outputs the detection result to the outside; wherein , The graphic reading unit includes: a one-dimensional bar code reading module, which is configured to read the one-dimensional bar code in the invoice graphic, and parse the one-dimensional bar code into a one-dimensional string; Barcode reading module, which is set to read a plurality of two-dimensional barcodes in the invoice graph, and to parse each two-dimensional barcode into a two-dimensional character string; a text recognition module, which is set to read A text graphic in the invoice graphic, and the text recognition module textualizes the text graphic to obtain a text content; wherein, the invoice content includes the one-dimensional character string, the two-dimensional character string, and the text content; wherein, When the text recognition module obtains the text content, the text recognition module determines whether the text content meets the plural specification text set in the text recognition module. If the text content lacks any of the specification text, the text recognition module The text recognition module re-reads the text graphics and textualizes the text graphics by block reading. When the text graphics are completely textualized, the text recognition module sequentially stacks the textualized text. The text content of each block is obtained, and the text content of each specification is compared with the text content of each block to determine Determine whether the text content of each block meets the standard text; among them, if the text recognition module determines that the text content of each block meets the standard text, the text recognition module outputs the composition of each block The text content; wherein, if the text recognition module determines that the text content of each block does not meet the standard text, the text recognition module re-reads the text and graphics, and reads the text in a block format Ways to textualize the text graphics. For example, the detection system of the electronic invoice verification sheet described in the first item of the scope of patent application, wherein the detection unit includes: a specification analysis module connected with the graph reading module, and the specification analysis module is based on the invoice A text specification in the specification detects the text content, a one-dimensional specification in the invoice specification detects the one-dimensional character string, and a two-dimensional specification in the invoice specification detects each of the two-dimensional character string to generate a plurality of the detection information ; A size detection module, which is connected with the graphics reading module to obtain the invoice graphics, the size detection module detects the invoice graphics according to a size specification in the invoice specifications to generate the detection information; The specification setting module is connected with the specification analysis module and the size detection module, and the specification setting module is set to set the invoice specification to provide the invoice specification to the specification analysis module and the size detection Module. For example, the detection system of the electronic invoice certification sheet described in the second item of the scope of patent application, wherein, when the specification analysis module detects the text content according to the text specification, the specification analysis module is based on the graphic of the business person’s identification mark Or text, the text of the electronic invoice certificate, the text of the year and the subsequent character track number, the year of the Republic of China and the period to which it belongs, the unified invoice character track number, the recording method of the transaction date and time, the text of the random code and the subsequent numbers The number, total text and subsequent numbers, the text of the seller or seller number, and the number of subsequent numbers to detect the text content. For example, the detection system of the electronic invoice verification sheet described in item 3 of the scope of patent application, wherein, when the specification analysis module detects the text content according to the text specification, the specification analysis module further bases on the text compiled by the buyer or the buyer The number of digits in the subsequent digits, or the record items of the transaction details, can be used to detect the text content. For example, the electronic invoice verification system described in the second item of the scope of patent application, wherein, when the specification analysis module detects the one-dimensional character string according to the one-dimensional specification, the specification analysis module is based on the one-dimensional character Whether the string conforms to the recording method of a code type, and whether the first 5 codes of the one-dimensional string are period types, the last 4 codes are random codes, the middle 10 codes are unified invoice track numbers, and whether the period type and the random code are random codes. Whether the code and the uniform invoice character track number match the period of the text content, the random code and the uniform invoice character track number to detect the one-dimensional character string; where the barcode type is a three-nine code (Code 39) . For example, the detection system of the electronic invoice verification sheet described in the second item of the scope of patent application, wherein, when the specification analysis module detects the two-dimensional character string according to the two-dimensional specification, the specification analysis module captures the two-dimensional character string A version information and an error-proof standard information in the neighboring places of the plural positioning patterns in the barcode, the specification analysis module then detects the version information and the error-proof standard information according to the two-dimensional specification. For example, the detection system of the electronic invoice verification sheet described in item 2 of the scope of patent application, wherein, when the specification analysis module detects the two-dimensional character string according to the two-dimensional specification, the specification analysis module is based on an error-proof The two-dimensional specification of the standard information detects the number of sales figures before the 77th code of the two-dimensional string, the encrypted verification information, and the unified invoice track number and transaction date and time that are the same as the text content. , The number of random code, the total number, the number of the seller number and the number compiled by the buyer. For example, the detection system for the electronic invoice verification sheet described in item 7 of the scope of patent application, wherein the encrypted verification information is a string combination of the uniform invoice track number and the random code, and An encryption process is used to convert to generate the encryption verification information. For example, the detection system of the electronic invoice certification sheet described in item 7 of the scope of patent application, wherein the specification analysis module further detects the transaction details after the 78th code of the two-dimensional string according to the two-dimensional specification or/and another Whether the transaction details of the two-dimensional string match the transaction details in the text content. For example, the inspection system of the electronic invoice verification sheet described in the second item of the scope of patent application, wherein, when the size inspection module detects the invoice pattern according to the size specification, the size inspection module captures one of the invoice patterns A width or a length to measure whether it meets the size specification based on the width or the length; if it does, the size detection module will then detect the unmeasured portion of the invoice graph according to the ratio recorded in the size specification The width or the length, and the position, width and length of the one-dimensional bar code, the two-dimensional bar code and the text graphic in the invoice graphic; if not, the size detection module determines that the invoice graphic does not meet the size Specification. For example, the detection system of the electronic invoice certification sheet described in the first item of the scope of patent application, wherein the detection result includes at least one qualified item that meets the invoice specifications, at least one error item that does not meet the invoice specifications, or for the error The project provides at least one suggestion item for revision suggestions. For example, the inspection system for the electronic invoice verification sheet described in the first item of the scope of patent application, wherein the invoice graphic includes an invoice front graphic, an invoice back graphic, an invoice detail graphic or a combination of any two or more thereof. For example, the inspection system of the electronic invoice verification sheet described in the first item of the scope of patent application, wherein the invoice content includes at least one one-dimensional bar code, one two-dimensional bar code, one text content, or a combination of any two or more thereof. A method for detecting an electronic invoice certificate, which includes: using a graphic obtaining device to obtain an invoice graphic from an electronic invoice; using a graphic reading module to read the invoice graphic, and generating an invoice based on the invoice graphic Content; use a detection unit to detect the invoice content or the invoice graphic according to a set invoice specification, and generate a detection result based on the plural detection information generated after detecting the invoice content or the invoice graphic; use an output module The detection result is output to the outside; wherein, when the graphic reading unit reads the invoice graphic, it includes the following steps: using a one-dimensional barcode reading module to read the one-dimensional barcode in the invoice graphic, And parse the one-dimensional barcode into a one-dimensional character string; use a two-dimensional barcode reading module to read the plural two-dimensional barcodes in the invoice graph, and parse each two-dimensional barcode into a two-dimensional character respectively String; use a text recognition module to read a text graphic in the invoice graphic, and the text recognition module textualizes the text graphic to obtain a text content; wherein, the invoice content includes the one-dimensional string, the two-dimensional The character string and the text content; wherein, when the text recognition module obtains the text content, the following steps are performed: the text recognition module determines whether the text content meets the plural specification text set in the text recognition module ; If the text content lacks any of the standard text, the text recognition module re-reads the text graphic and textualizes the text graphic by block reading; when the text graphic is fully textualized , The text recognition module sequentially stacks the text content of each block obtained after textualization; according to the comparison of each standard text with the text content of each block, the text content of each block is determined Whether it conforms to each standard text; wherein, if the text recognition module determines that the text content of each block conforms to each standard text, the text recognition module outputs the text content composed of each block; Wherein, if the text recognition module determines that the text content of each block does not meet the specifications of the text, the text recognition module re-reads the text graphics and textualizes it in a block-based reading method. The steps of the text graphics. For example, the detection method of the electronic invoice verification sheet described in item 14 of the scope of patent application, wherein, when the detection unit detects the content of the invoice according to the invoice specification, it includes the following steps: using a specification analysis module according to the invoice specification Use a text specification to detect the text content, a one-dimensional specification in the invoice specification to detect the one-dimensional character string, and a two-dimensional specification in the invoice specification to detect each of the two-dimensional character strings to generate a plurality of the detection information; A size detection module detects the invoice graphics according to a size specification in the invoice specifications to generate the detection information; uses a specification setting module to set the invoice specifications to provide the invoice specifications to the specification analysis module and the Size detection module. For example, the detection method of the electronic invoice certification sheet described in item 15 of the scope of patent application, wherein, when the size detection module detects the invoice graphics according to the size specifications, the size detection module captures one of the invoice graphics A width or a length to measure whether it meets the size specification based on the width or the length; if it does, the size detection module will then detect the unmeasured portion of the invoice graph according to the ratio recorded in the size specification The width or the length, and the position, width and length of the one-dimensional bar code, the two-dimensional bar code and the text graphic in the invoice graphic; if not, the size detection module determines that the invoice graphic does not meet the size Specification.

Images