TWM655760U - System for processing invoice data - Google Patents

Abstract

This new model is a system for processing invoice data, including a scanner, an OCR module, three files, a display interface and a conversion module. The above three files are the first file module, the second file module, and the third file module. The scanner is used to scan invoices, and the OCR module converts the scanned information into data and stores it in the first file module. The second file module contains a correction algorithm for data correction, and the corrected data is stored in the third file module. The display interface provides simultaneous display of invoice images and data for manual correction. Finally, the conversion module converts the corrected data into the accounting system format. It is worth noting that the third file module does not contain any formulas or programs.

Description

System for processing invoice data

This novel invention relates to the field of data processing and management, in particular to a method and system for processing invoice data.

Invoice processing involves handling multiple invoice types and formats. Traditional invoice processing methods often involve manual data entry, which is a time-consuming, error-prone and inefficient process. In addition, manually categorizing and sorting invoices can be a cumbersome task, especially when processing a large number of invoices.

Optical character recognition (OCR) technology has been used to automate the process of converting scanned invoice images into machine-readable text. However, OCR technology is not infallible and errors often occur. These errors can be caused by a variety of factors, such as poor scan quality, font variations, character spacing, and confusion between similar-looking characters. Therefore, the data obtained from OCR usually requires further correction and verification to ensure its accuracy.

Additionally, data extracted from invoices often requires further processing to be compatible with various accounting systems. This may involve converting the data into a specific format acceptable to the accounting system. Manually converting data into different formats can be a tedious and error-prone task.

Therefore, despite the integration of OCR technology, traditional invoice processing methods still face challenges in terms of efficiency, accuracy and compatibility with various accounting systems.

The present invention provides a system for processing invoice data, which includes a scanner, an OCR module, three files (a first file module, a second file module, and a third file module), a display interface, and a conversion module. The scanner is used to scan invoices, and the OCR module converts invoice information into data and stores it in the first file module. The second file module contains an algorithm for correcting data, and the corrected data is stored in the third file module. The display interface provides simultaneous display of invoice images and data for easy manual correction, while the conversion module converts the data into a format acceptable to the accounting system. This system is particularly suitable for file management in Excel format and supports running the OCR module on the Microsoft Power Automate platform.

In the above system, the algorithm in the second file module can be customized based on common error patterns observed in OCR technology. Invoices can be classified according to the invoice type. These categories can include sales hand-issued three-part invoice, sales hand-issued two-part invoice, purchase hand-issued three-part invoice, purchase machine-issued three-part invoice, purchase machine-issued two-part invoice and sales returns, purchase returns or allowance certificates.

In the above system, the display interface allows manual correction of the data corresponding to the invoice image. The algorithm uses data validation techniques to correct the data. These data validation techniques may include rule validation, pattern matching, range checking, list validation, multi-stage validation, manual review, and error logging and reporting.

In the above system, the folders and files used are named and managed with specific symbols and naming rules for easier identification and automatic processing. These symbols and naming rules can be customized based on invoice types.

In order to make the above features and advantages of the new model more obvious and easy to understand, the following is a detailed description of the preferred embodiment with the attached drawings.

100: Invoice data processing system

110: Scanner

120:OCR module

130: First file module

140: Second file module

142:Data correction algorithm

150: Third file module

160: Display interface

170:Conversion module

200:Accounting system

10: Invoice

S110~S170: Flowchart symbols

Figure 1 shows a flow chart of the new invoice data processing method.

Figure 2 shows the new invoice data processing system.

Figure 3 shows the invoice image and corresponding data displayed simultaneously on the display interface.

Please refer to FIG. 1, which is a flowchart of the novel invoice data processing method, which involves multiple steps. First, please refer to step S110, and multiple invoices are classified and scanned. In this embodiment, the classification of invoices 10 is based on the type of invoice, which depends on the specific business or accounting background of the country or region. Once the invoice is classified and scanned, step S120 is executed to convert the scanned information of the invoice into data. This conversion process uses optical character recognition technology (hereinafter referred to as OCR technology), which can convert the scanned image of the invoice into computer-readable text.

After the scanned information is converted into data, step S130 is executed to transfer the converted data into the first file module. This first file module serves as a storage for the converted data, allowing subsequent processing and manipulation of the data. The first file module is usually an Excel file, which provides a convenient and widely used format for storing and processing data.

After the converted data is transferred to the first file module, step S140 is executed to transfer the data from the first file module to the second file module for data correction. The second file module includes algorithms for correcting the data in the first file module. These algorithms are customized based on common error patterns observed in OCR technology. By correcting these common errors, the accuracy of the data can be significantly improved.

After the data in the second file module is corrected, step S150 is executed to transfer the corrected data from the second file module to the third file module. This transfer process is intended to maintain a file that does not contain any formulas or programs, thereby improving performance optimization. In this embodiment, the first file module, the second file module, and the third file module are all Excel files.

In addition to the data correction and transfer process, the method also involves simultaneously displaying the invoice image and the corresponding data on a display interface (as shown in step S160). This display interface allows manual correction of the data corresponding to the invoice image. By providing a visual representation of the invoice image and the corresponding data, the user can easily identify and correct any differences or errors in the data.

Afterwards, step S170 is executed to convert the corrected data into a format acceptable to the accounting system. This conversion process ensures that the processed invoice data can be seamlessly integrated into the existing accounting system, thereby promoting efficient and accurate accounting practices.

Below, each step of the above method will be described in more detail. Please refer to Figure 1 and Figure 2 at the same time. Figure 2 shows the new invoice data processing system of the present invention. The invoice data processing system 100 includes several components, each of which is designed to perform a specific function in the whole process. These components include a scanner 110, an optical character recognition module (hereinafter referred to as OCR module 120), a first file module 130, a second file module 140, a third file module 150, a display interface 160, and a conversion module 170.

The display interface serves as a visual platform for manually correcting the data corresponding to the invoice image. The interface is designed to present the user with a side-by-side view of the invoice image and the corresponding data, thereby providing a clear and intuitive way to compare and verify the accuracy of the data.

A conversion module is a software component designed to perform data conversion tasks. It includes a set of predefined rules and algorithms that indicate how to convert data in a third-party file module into a format acceptable to an accounting system. The conversion module can be customized to suit different accounting systems, thus ensuring that the method is flexible and versatile.

The initial step of the method, i.e., step S110, is to classify and scan a plurality of invoices 10. This step involves classifying the invoices 10 into different categories according to the type of invoice 10. The classification of the invoices 10 is not arbitrary, but is determined by specific characteristics inherent in the invoices 10 themselves. These characteristics may include, but are not limited to, the nature of the transaction represented by the invoice, the parties involved in the transaction, and the format or structure of the invoice.

In this embodiment, the invoice 10 is divided into six categories. These categories include sales hand-issued triple invoice, sales hand-issued double invoice, purchase hand-issued triple invoice, purchase machine-issued triple invoice, purchase machine-issued double invoice, and sales return, purchase return or allowance certificate. Each category represents a different type of invoice commonly seen in business and accounting practice.

Once the invoice is sorted, it is scanned. The scanning is performed by a scanner 110, which is responsible for scanning the invoice and converting it into a digital image. The scanner 110 can be any device capable of capturing a high-quality image of the invoice, such as a flatbed scanner or a document feeder scanner. The scanner is designed to capture all relevant information contained in the invoice, including text information, numerical information, and any other symbols or markings that may be present on the invoice.

The scanning process involves capturing an image of each invoice, which can then be processed and analyzed by the OCR module 120 in step S120. The scanning process is intended to capture all relevant information contained in the invoice, including text information, numeric information, and any other symbols or marks that may be present on the invoice. The result of the scanning process is a digital image of the invoice, which serves as input to the subsequent OCR process.

That is, after scanning and classifying the invoice, the method continues to convert the scanned invoice information into data. In the present embodiment, this conversion process is achieved by using the OCR module 120. The OCR module 120 performs OCR technology, which is a technology that can convert text images in different types of documents (such as scanned paper documents, PDF files, or images captured by a digital camera) into editable and searchable data. The OCR module 120 is responsible for converting the scanned invoice image into text that can be read by computer software. This involves analyzing the structure of the scanned image and identifying and extracting the text information contained inside the image. The OCR module 120 can recognize a variety of characters and symbols, including letters, numbers, and special characters. This allows comprehensive extraction of all relevant information from the invoice, not only the main text of the invoice, but also any other information or annotations that may be present. That is, OCR technology is used to convert the scanned invoice image into machine-readable text, and then convert the extracted text into data that can be stored, manipulated and analyzed in a digital format. In addition, in step S120, the OCR module 120 allows comprehensive extraction of all relevant information from the invoice 10, including not only the main text of the invoice, but also any other information or annotations that may be present.

It is worth noting that although OCR technology is generally accurate and reliable, it is not infallible and sometimes errors occur. These errors may be caused by a variety of factors, such as poor scan quality, font variations, character spacing, and confusion between similar-looking characters. Therefore, the data obtained from the OCR module 120 generally requires further correction and verification to ensure its accuracy, which is facilitated by the subsequent steps of the novel method.

In this embodiment, the OCR technology used is executed on the Microsoft Power Automate platform. Microsoft Power Automate is a cloud-based service provided by Microsoft that allows users to create and manage automated workflows between multiple applications and services. In the context of this novel method, Microsoft Power Automate is used to automate the OCR process, thereby improving the efficiency and accuracy of converting scanned invoice information into data.

Once the scanned invoice information is converted into data using OCR technology, the next step of the method, i.e., step S130, is to transfer these converted data into the first file module 130. This transcription process involves converting the data from its original format (generated by the OCR process) into a format suitable for further processing and manipulation. The first file module 130, as a storage for the converted data, provides a structured and organized format to store the data, which facilitates subsequent operations such as data correction, data verification and data transfer. The first file module 130 is usually an Excel file, which is a format widely used for data storage and manipulation. Excel files offer a range of features that are useful for working with invoice data, such as the ability to organize data in rows and columns, perform calculations on data, and apply various data formatting and data validation rules.

It is worth noting that transferring the converted data to the first file module 130 is not just a simple copy process. Instead, it involves converting the data from its initial format (resulting from the OCR process) to a format suitable for further processing and manipulation. This conversion process may involve various operations such as data formatting, data cleaning, and data validation, all of which are to ensure that the data is in a suitable state for subsequent steps.

In this embodiment, the first file module 130 serves as the initial storage for the converted data, which provides a starting point for subsequent data correction and data transfer processes and serves as a reference point for the manual correction process. By storing the converted data in the first file module 130, the method ensures that the data is available for these subsequent processes at any time, thereby facilitating efficient and accurate processing of invoice data.

After the converted data is transferred to the first file module 130, the method proceeds to the next step, namely: step S140, transferring the data from the first file module 130 to the second file module 140 for data correction. This transfer process is intended to facilitate the correction of any errors or inaccuracies that may have been introduced during the OCR process. The second file module 140 serves as a platform for applying various data correction algorithms 142 that are intended to identify and correct common error patterns in the data.

The data correction algorithm 142 in the second file module 140 is customized based on common error patterns observed in OCR technology. These error patterns may include, but are not limited to, character confusion, noise interference, alignment and spacing problems, font and size variations, and difficulties with handwritten text. By customizing the algorithm to address these specific error patterns, the method can significantly improve the accuracy of the data obtained from the OCR process.

The data correction algorithm 142 operates by comparing the data in the first file module 130 to a set of predefined rules or patterns. If the data matches a known error pattern, the algorithm corrects the data. This correction process may involve various operations, such as replacing incorrect characters, adjusting character spacing, correcting alignment problems, and standardizing font and size variations. The corrected data is then stored in the second file module 140 for further processing.

It is important to note that the data correction process of step S140 is not a one-size-fits-all solution. Rather, it is a flexible and adaptable process that can be customized based on specific error patterns prevalent in the data. This customization is facilitated by the use of algorithms in the second file module 140 that can be adjusted and fine-tuned based on observed error patterns. This ensures that the data correction process is as effective and accurate as possible, thereby improving the overall quality and reliability of the invoice data.

After the data correction process is completed in the second file module 140, step S150 is performed and the corrected data is then transferred to the third file module 150. This transfer process is an integral part of the overall method and has a specific purpose, namely to optimize performance in the data processing workflow. The third file module 150 is intended to be a clean and optimized file that does not contain any formulas or programs. This is in contrast to the second file module 140, which contains various algorithms and formulas for data correction.

The reason for this migration process is to maintain a performance optimized file. Files containing formulas or programs (such as the second file module) may be slow to load and process. This can be a problem when processing large amounts of data, which is often the case in invoice processing. By migrating the corrected data to the third file module 150, which does not contain any formulas or programs, this approach ensures that the data can be accessed and processed quickly and efficiently.

The third file module 150, like the first file module 130, is also typically an Excel file. However, unlike the first file module 130 and the second file module 140, the third file module 150 is specifically formatted to not contain any formulas or programs (algorithms). This is achieved by using a specific file format, such as XLSX, which is a file format used by Microsoft Excel that does not support macros. By using this file format, this step ensures that the third file module 150 is a clean and optimized file suitable for further processing and manipulation.

In summary, the transfer of the corrected data from the second file module 140 to the third file module 150 is a strategic step in the method to optimize the performance of the data processing workflow by maintaining a clean and optimized third file module 150. This step ensures that the invoice data can be processed quickly and efficiently, thereby improving the overall efficiency and effectiveness of the invoice processing method.

After the corrected data is transferred to the third file module 150, step S160 is executed to display the invoice image and the corresponding data simultaneously on the display interface 160 (as shown in FIG. 3). This display interface 160 serves as a visual platform for manually correcting the data corresponding to the invoice image. The display interface 160 is intended to provide the user with a side-by-side view of the invoice image and the corresponding data, thereby providing a clear and intuitive way to compare and verify the accuracy of the data.

Display interface 160 is intended to facilitate the manual correction process by providing a clear and intuitive way to compare and verify the accuracy of the data. By displaying the invoice image and the corresponding data side by side, display interface 160 enables the user to easily identify any inconsistencies or errors in the data. This visual comparison process is particularly useful in situations where the OCR process results in errors or inaccuracies in the data. By visually comparing the invoice image to the corresponding data, the user can easily identify and correct these errors, thereby improving the overall accuracy of the data.

The manual correction process facilitated by the display interface 160 is an integral part of the overall approach. Despite the use of advanced OCR technology and sophisticated data correction algorithms, there may still be instances where errors or inaccuracies in the data are not detected or corrected by these automated processes. In such cases, the manual correction process provides a final layer of quality control, ensuring that the data is as accurate and reliable as possible before being converted into a format acceptable to the accounting system 200.

In summary, the display interface 160 serves as a tool for manually correcting data corresponding to an invoice image. By providing a clear and intuitive way to compare and verify the accuracy of data, the display interface 160 improves the overall efficiency and effectiveness of the invoice processing method.

After the manual correction process facilitated by the display interface 160, the method proceeds to step S170, which converts the corrected data into a format acceptable to the accounting system 200. This conversion process is a key step in the overall method because it ensures that the processed invoice data can be seamlessly integrated into the existing accounting system, thereby promoting efficient and accurate accounting practices.

In this step S170, the conversion process involves converting the corrected data stored in the third file module 150 into a format compatible with the accounting system 200. This involves mapping the data fields in the third file module to the corresponding fields in the accounting system 200 and converting the data values into a format recognizable by the accounting system. This may involve various operations such as data formatting, data type conversion, and data validation, depending on the specific requirements of the accounting system 200.

To facilitate this conversion process, the method uses a conversion module 170. The conversion module 170 is a software component designed to perform data conversion tasks. It includes a set of predefined rules and algorithms that guide how to convert the data in the third file module 150 into a format acceptable to the accounting system 200. The conversion module 170 can be customized to adapt to different accounting systems, thereby ensuring that the method is flexible and versatile.

The conversion module 170 converts the data into the required format by reading the data in the third file module 150, applying the conversion rules and algorithms, and then outputting the converted data. The output of the conversion module 170 is a set of data presented in a format acceptable to the accounting system. These data can then be imported into the accounting system 200, thereby completing the invoice processing workflow.

In summary, converting the corrected data into a format acceptable to the accounting system 200 is a strategic step in the disclosed method, ensuring that the processed invoice data can be seamlessly integrated into the existing accounting system. By using the conversion module to facilitate this process, the method ensures that the invoice data can be processed accurately and efficiently, thereby improving the overall efficiency and effectiveness of the invoice processing method.

In this embodiment, a specific process is employed to name and manage folders and files. This process involves the use of specific symbols and naming conventions designed to facilitate easier identification and automated handling of folders and files. These symbols and naming conventions are based on the type of invoice, thereby providing a logical and intuitive system to organize and manage invoice data.

Among them, the symbols used in the naming and management process represent different types of invoices. For example, the symbol '▲' represents a sales invoice, while the symbol '▼' represents a purchase invoice. The number of 'ʃ' symbols indicates whether the invoice is a triple or double invoice. For example, three 'ʃ' symbols, that is: ʃʃʃ, represent a triple invoice, while two 'ʃ' symbols, that is: ʃʃ, represent a double invoice. The presence of the 'III' or 'II' symbol indicates that the invoice is issued by a cash register, where 'III' represents a triple cash register invoice and 'II' represents a double cash register invoice. The combination of the '▲' and '▼' symbols represents a sales return, a purchase withdrawal, or a discount certificate.

Additionally, naming conventions involve appending a sequence number or other symbol to a symbol representing an invoice type. For example, a file or folder name might be '▲ʃʃ-1-xxx-load', where '▲ʃʃ' represents a manually issued invoice, '-1-' is a sequence number, 'xxx' is a placeholder for additional information, and 'load' indicates that the processing workflow is in the downloading data phase. These naming conventions provide a systematic and consistent way to name and manage folders and files, thereby facilitating efficient and accurate processing of invoice data.

By using these specific symbols and naming conventions, the present embodiment provides an efficient and effective way to manage folders and files involved in the invoice processing workflow. This not only improves the overall efficiency of the method, but also improves the accuracy and reliability of invoice data processing, thereby helping to improve the overall effectiveness of the invoice processing method.

In the above-mentioned embodiment, it is mentioned that the data correction algorithm 142 included in the second file module 140 uses various data verification techniques to identify and correct common error patterns in the data. These techniques are intended to improve the accuracy and reliability of the data obtained from the OCR process. Below, these data correction algorithms will be introduced in more detail with examples.

One of these techniques is rule validation. This involves comparing the data to a set of predefined rules or patterns. If the data matches a known error pattern, the algorithm makes corrections to the data. This can involve various actions, such as replacing incorrect characters, adjusting character spacing, correcting alignment issues, and regulating font and size variations. Another technique used by the algorithm is pattern matching. This involves identifying specific patterns or structures in the data, such as dates, phone numbers, or invoice numbers. By identifying these patterns, the algorithm can accurately extract and verify the corresponding data.

Range checking is another technique used by the algorithm. This involves comparing numeric data to a predefined range or limit. If the data falls outside this range, it is flagged as potentially erroneous. This technique is particularly useful for validating numeric data, such as invoice amounts or tax rates. List validation is another technique employed by the algorithm. This involves comparing the data to a predefined list of valid values. If the data does not match any of the values in the list, it is flagged as potentially erroneous. This technique is particularly useful for validating categorical data, such as product codes or supplier names.

The algorithm also employs a multi-stage validation process. This involves applying multiple validation techniques in sequence. Each stage of the validation process is designed to catch and correct different types of errors, thereby improving the overall accuracy of the data.

In addition to these automated validation techniques, the approach also involves a manual review process. This involves a human operator reviewing the data and manually correcting any errors or inconsistencies. This manual review process acts as a final layer of quality control, ensuring that the data is as accurate and reliable as possible.

Additionally, the algorithm includes error logging and reporting capabilities. This involves recording all identified errors and corresponding corrections in an error log. This error log can be reviewed and analyzed to identify common error patterns and improve the performance of the algorithm. The error log also serves as a record of the data correction process, providing transparency and accountability.

In summary, the data correction algorithm 142 in the second file module 140 uses various data validation techniques to improve the accuracy and reliability of the data obtained from the OCR process. These techniques, combined with the manual review process and error logging capabilities, ensure that the data is as accurate and reliable as possible, thereby improving the overall efficiency and effectiveness of the invoice processing method.

In summary, this patent discloses a method and system for processing invoice data efficiently and accurately. The method and system utilize advanced OCR technology and sophisticated data correction algorithms, as well as manual review processes and error recording functions to ensure that the data is as accurate and reliable as possible. In addition, the method and system also include a specific process to name and manage folders and files, and convert the corrected data into a format acceptable to the accounting system, thereby improving the overall efficiency and effectiveness of the invoice processing method. These features make the method and system of great practical value to enterprises that process large amounts of invoice data.

Although the present invention has been disclosed as above with the preferred embodiment, it is not intended to limit the present invention. Anyone with common knowledge in the relevant technical field can make some changes and modifications without departing from the spirit and scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the scope defined in the attached patent application.

100: Invoice data processing system

110: Scanner

120:OCR module

130: First file module

140: Second file module

142:Data correction algorithm

150: Third file module

160: Display interface

170:Conversion module

200:Accounting system

10: Invoice

Claims (9)

A system for processing invoice data, the system comprising: a scanner for scanning a plurality of invoices; an OCR module electrically connected to the scanner, the OCR module for converting the scanned invoice information into data; a first file module electrically connected to the OCR module, the first file module for receiving and storing the converted data; a second file module electrically connected to the first file module, the second file module for receiving the data from the first file module and performing data modification, wherein the second file module comprises a module for modifying the invoice information; An algorithm for calculating positive data; a third file module electrically connected to the second file module, the third file module is used to receive and store the corrected data from the second file module; a display interface electrically connected to the third file module, the display interface is used to simultaneously display the image of the invoice and the corresponding data for manual correction; and a conversion module electrically connected to the third file module, the conversion module is used to convert the corrected data into an acceptable format for an accounting system; wherein the third file module does not contain any formula or program. A system as described in claim 1, wherein the first file module, the second file module and the third file module are all Excel files. A system as claimed in claim 1 or claim 2, wherein the algorithm in the second file module is customized based on common error patterns observed in the OCR module. A system as claimed in claim 1, wherein the invoices are classified according to invoice categories. A system as described in claim 1, wherein the algorithm uses a data validation technique to modify the data, and the data validation technique includes rule validation, pattern matching, range checking, list validation, or multi-stage validation. A system as described in claim 1, wherein the folders and files used in the system are named and managed using specific symbols and naming rules for easier identification and automatic processing. A system as claimed in claim 6, wherein the symbols and naming conventions are customized based on the invoice type. A system as described in claim 2, wherein the first file module and the third file module are XLSX files, and the second file module is an XLSM file. A system as described in claim 1, wherein the OCR module is executed on the Microsoft Power Automate platform.

Images