Patent Application
20170220530
This invention relates to extracting renderable text from documents.
In an entity, processes may receive Adobe™ portable document format (hereinafter, “PDF”) documents. Some examples of such processes may be financial statement analytics, payroll payments processing and non-tabular data conversions for account settings. The processes may require manipulating data included on the PDF documents.
For the purposes of this application, PDF documents are typically substantially unable to be manipulated. Conventionally, the data was manually read from the PDF documents and re-entered into a computer application. The re-entry process is cumbersome as well as error-prone.
Therefore, a generic renderable text extraction tool may be desirable. Preferably, the tool may enable extraction of text from an editable PDF document. It may also be desirable for the tool to export the extracted text into a format specified by a user.
An apparatus for extracting renderable text from an Adobe PDF™ document is provided. The apparatus may include a user interface. The user interface may be configured to receive the Adobe PDF™ document. The user interface may also be configured to render the Adobe PDF™ document into a dots per inch (“DPI”) image. For the purposes of this application, rendering may be understood to mean utilizing a method that converts a PDF document into a DPI image. The original PDF document may be stored for later use. The DPI image may function easily in a Java™ region, as opposed to a PDF document, which may not function properly in a Java™ region. This may be at least because the DPI image is in a different format than is acceptable by the Java™ region.
The region may be a Java™ abstract window toolkit (“AWT”) region. The region may be a Java™ Swing region.
Java™ AWT may be a Java™ interface to a computer system's native operating system's graphical user interface (“GUI”) code. Therefore, AWT connects with the operating system layer. Because AWT utilizes operating system GUI code, an application created in AWT may work inconsistently across platforms.
Java™ Swing, on the other hand, may be a library of Java™ code. Java™ Swing may include its own functionality and logic to produce GUIs. Java™ Swing generally utilizes very minimal basic operating system functions. Therefore, Java™ Swing is more portable than Java™ AWT. But, because of the additional overhead produced by the layer of Java logic, Java™ Swing is sometimes slower than Java™ AWT.
The user interface may add the DPI image to a panel. The panel may be a Java™ JPanel or any other suitable panel or software structure. A JPanel may be a lightweight GUI container. The JPanel may hold various GUI elements. A JPanel may inherit its qualities from Java™ JComponent, which may inherit its qualities from Java™ Container, which may inherit its qualities from Java™ Component, which may inherit its qualities from Java™ Object.
Therefore, it may be understood that a JPanel may reside in a component. A component may be a Java™ JComponent or any other suitable component or software structure. A JComponent may be a base class for many Java™ Swing components.
In some instances, a top-level Swing container may be used as a component. A top-level Swing container may include a list of components. The components may include a root pane. The root pane may include a layered pane, a content pane and a glass pane.
The layered pane may be utilized to position the contents of the root pane. The content pane may include the root pane's visible components. The glass pane may be hidden initially. If made visible, the glass pane may act like a sheet of glass over the other parts of the root pane. The glass pane may be used to catch events or paint over an area of the root pane that already contains components. For example, one can display an image over multiple components using the glass pane.
The user interface may insert a pane into the component. The pane may be positioned on top of the DPI image. The pane may have a transparent quality. The pane may be a glass pane or any other suitable pane or software structure. The insertion may occur following rendering the PDF document and displaying the DPI image. The insertion may occur prior to rendering the PDF document or displaying the DPI image. The insertion may occur prior to displaying the DPI image, but after rendering the DPI image.
The user interface may support the use of a line insertion tool. The line insertion tool may enable a user to place one or more horizontal lines on the pane. The line insertion tool may also enable a user to place one or more vertical lines on the pane. The line insertion tool may include a toggle feature that enables a user to switch between horizontal line creation and vertical line creation.
The user interface may receive horizontal and vertical lines from the user. In some embodiments, upon receipt of at least two horizontal lines and at least two vertical lines from the user, the user interface may calculate a plurality of intersection points using a line intersection algorithm. The intersection points may outline a plurality of rectangular areas. The user interface may create a plurality of templates based on the rectangular areas.
The apparatus may also include a text extraction parser. The text extraction parser may be configured to extract text from a plurality of portions of the editable PDF document corresponding to the templates.
The text extraction parser may be configured to transform the extracted text into renderable text. The text extraction parser may also be configured to export the renderable text, utilizing the templates for text structure, into a manipulate-able document.
The objects and advantages of the invention will be apparent upon consideration of the following detailed description, taken in conjunction with the accompanying drawings, in which like reference characters refer to like parts throughout, and in which:
A method for extracting renderable text from an Adobe PDF™ document is provided. The method may include receiving the Adobe PDF™ document. The method may include rendering the Adobe PDF™ document to a dots per inch (“DPI”) image. The rendering may utilize an image, a font a glyph or any other suitable software structures. The DPI image may be used in a Java™ abstract window toolkit (“AWT”) region. The DPI image may be used in a Java™ swing region. The DPI image may be used in any other suitable region.
The method may also include adding the DPI image to a panel. The panel may be a Java™ JPanel. The panel may be any suitable panel or software structure. The panel may reside in a component. The component may be a Java™ JComponent. The component may be any suitable component or software structure.
The method may also include displaying the DPI image within the component on a screen. The screen may be the screen of a personal computer, work computer, tablet, smartphone, and any other suitable computing device.
The method may also include inserting a pane into the component on top of the DPI image. The pane may have transparent quality. The insertion may occur upon rendering the Adobe™ PDF document. The insertion may occur upon displaying the DPI image on the screen. The insertion may occur upon both rendering the Adobe™ PDF document and displaying the DPI image.
The method may also include using a line insertion tool. The line insertion tool may enable a user to place one or more horizontal lines on the pane. The line insertion tool may also enable a user to place one or more vertical lines on the pane. The line insertion tool may be associated with a mouse. In certain embodiments, a user may toggle between creation of horizontal lines and creation of vertical lines by right clicking on the mouse.
In some embodiments, right clicking the mouse may open a mini menu. The menu may include two options: vertical and horizontal. The user may choose the vertical option, create vertical lines, and then choose the horizontal option, and then create horizontal lines, or vice versa.
The method may also include receiving horizontal lines and/or vertical lines from the user. In some embodiments, upon receipt of at least two horizontal lines and at least two vertical lines from the user, the method may include calculating, using a line intersection algorithm, a plurality of intersection points of the horizontal and vertical lines. The plurality of intersection points may outline a plurality of rectangular areas.
In some embodiments, the edges of a page may be utilized as vertical and/or horizontal lines. In these embodiments, if a user initially defines a horizontal line, he may create a row utilizing the one horizontal line. If a user initially defines a vertical line, he may be required to define at least one additional vertical line (two total vertical lines) and at least one horizontal line to create one row and one column.
The method may also include creating a plurality of templates based on the plurality of rectangular areas. A template may be a Java™ object. The template object may be based on the rectangular areas. A template may include data that enables a system to properly invoke the rectangular areas. In some embodiments, the rectangular areas themselves may be defined as Java™ objects, and therefore used in the text extraction.
Templates may also include extrapolating a portion of the defined rectangular areas from one portion of a page and utilizing the rectangular areas for a different portion of the page. The method may also include extracting text from a plurality of portions of the editable PDF document corresponding to the templates.
The method may also include transforming the extracted text into renderable text. The method may also include exporting, utilizing the templates for text structure, the renderable text into a manipulate-able document. The text structure may include columns, rows, paragraphs or any other suitable text structure. The manipulate-able document may be a Microsoft Excel™ spreadsheet or any other suitable document format.
The method may also include creating, utilizing multiple threads, compressed metadata comprising the renderable text, for each page of the Adobe PDF™ document. The metadata may be the rectangular area data or the template data. The rectangular area data/template data may be metadata about data included on the PDF page. The method may include saving the compressed metadata in a cache memory. The cache memory may an offline cache memory. Saving the compressed metadata in an offline cache memory may ensure optimal memory usage of the underlying computer system. This may be because the text extraction process may be CPU-intensive. Accessing the cache memory, as opposed to accessing the hard drive, during the text extraction process may increase speed and performance of the system.
The method may also include generating, utilizing a second set of multiple threads, a preview corresponding to the renderable text. The preview may include rows and columns which may be defined by the horizontal and vertical lines.
Illustrative embodiments of apparatus and methods in accordance with the principles of the invention will now be described with reference to the accompanying drawings, which form a part hereof. It is to be understood that other embodiments may be utilized and structural, functional and procedural modifications may be made without departing from the scope and spirit of the present invention.
Step 2 shows a user may define coordinates and/or templates on top of glass panel 204. The coordinates and/or templates may define sections of text to be extracted from PDF page 202.
User interface 201 may transmit the user-defined coordinates in addition to the PDF document to itext engine 206, as shown at step 3. Itext Engine 206 may include parser 208. Parser 208 may extract data from the transmitted PDF document 202 based on the received, previously, user-defined coordinates and/or templates. Step 4 shows itext engine 206, utilizing multiple threads, may cache the zipped metadata for each page of the PDF, together with the text content, offline, during processing of the document. Caching the metadata offline may ensure optimal memory usage of itext engine 206 during the text extraction process.
Step 5 shows itext engine 206, utilizing multiple threads, generating a preview of a renderable document to be generated, for e.g., a Microsoft™ Excel spreadsheet, on user interface 201.
Upon clicking an export button on preview 210, as shown at step 6, Apache™ POI (a Java™ application programming interface (“API”) for Microsoft™ documents) 212 may generate Microsoft Excel™ spreadsheet 214, as shown at step 7.
In some embodiments, the process may not require human intervention. In these embodiments, the system may define the vertical and horizontal lines based on whitespace or any other suitable indicator.
It should be appreciated that UI 302, PDF reader UI 304, PDF itext tool reader 306, itext parser 308 and file store render listener 310 may be objects in an OOP programming language.
The user may define templates at UI 302 (method 328). The user may also click export at UI 302 (method 328). PDF reader UI 304 may receive the document and the defined templates. PDF reader UI 304 may transmit the PDF document, utilizing PDF Extractor (method 330), to PDF itext tool reader 306. PDF itext tool reader 306 may extract PDF columns from the PDF document (method 332). Steps 334 and 336 may show a return message informing the applicable parties that the method steps have been completed successfully or unsuccessfully. In some embodiments, steps 334 and 336 may be a return message call in an OOP language or any other suitable programming language.
The user may define templates at UI 302 (method 338). The user may also apply the templates to other pages (method 338). The user may click apply (method 338). PDF reader UI 304 may receive the document and the defined templates. PDF reader UI 304 may transmit the document and the defined templates, utilizing PDF Extractor method (method 340), the PDF itext tool reader 306. PDF itext tool reader 306 may extract PDF columns from the document corresponding to the pages chosen at 338 (method 342).
Steps 344 and 346 may be a return message informing the applicable parties that the method steps have been completed successfully or unsuccessfully. In some embodiments, steps 344 and 346 may be a return message call in an OOP language or any other suitable programming language.
Thread pool executor 404 may also execute the data extraction (method 410). Step 412 may be a return to PDF itext tool reader 402. Method 414 may show PDF itext tool reader 402 testing the PDF columns. Testing of the PDF columns may ensure the quality of the text extraction. Method 416 may show pre-processing or pre-conditioning of the data to enable Microsoft Excel™ processing.
UI 502 may also direct a user to open a recently opened file. Utilizing cursor 512, a user may choose open recent files option 508. Upon clicking open recent files option 508, submenu 510 may open and display to the user recently opened files.
Exit option 514 may enable a user to close UI 502.
A user may right click, or utilize any other suitable mouse or keyboard operation, to open menu 610. Menu 610 may include vertical and horizontal options. Utilizing cursor 604, a user may select either vertical or horizontal. Upon selecting, for example, vertical, the user may click on PDF page 602. At the location that the mouse click was received the system may draw a vertical line across PDF page 602.
In some embodiments, the colors of both the vertical lines and the horizontal lines may be the same. In some embodiments, the horizontal lines may be one color and the vertical lines may be a different color.
In
In
Actions menu 722 may include action buttons. Selecting export button 734 may export the text from PDF page 704 into a manipulate-able document, for example, Microsoft Excel™. Selecting preview button 736 may enable a user to view a preview of the manipulate-able document. Selecting clear button 738 may delete all of the vertical and horizontal lines on a specific page. Selecting clear all button 740 may delete all of the vertical and horizontal lines in an entire document. Selecting exit button 742 may close the program.
Page navigation menu 724 may enable a user to change pages with a PDF document. A user may type a page number into text entry field 728. Upon entering the page number or upon typing an “Enter” key, the system may “go to” the selected page, as shown at 726. A user may also utilize forward arrow 734 and back arrow 730 to change the page. Field 732 shows a user what page he or she is currently viewing.
Templating menu 726 may enable using the template, created for the current page, for other pages in the document. A user may select option 728, to apply the template to the entire document. A user may select option 730 to apply the template to a selected set of other pages in the document. Options 728 and 730 may be mutually exclusive of each other, when one option is selected, the other option may be disabled. The user may enter the page numbers into text entry field 732. A user may select apply button 734 to apply the changes selected in templating menu 726.
Upon selecting the export button discussed in
In some embodiments, the system may utilize NBID and hostname-based digital signature authentication. The authentication may allow finer control of use of the PDF extraction application.
PDF extraction tool may also include format to export menu 1110. Format to export menu 1110 may enable a user to export the PDF document to a variety of formats, for example, XLS (Excel) (1108), CSV (comma separated values), XML (eXtensible markup language), XMI (eXtensible markup language metadata interchange), JSON (JavaScript object notation) and text. Format to export menu 1110 may include remove button 1112 and apply button 1114. Remove button 1112 may enable a user to remove one of the options to export. Apply button 1114 may enable a user to apply the selected format to export. A user may select more than one format to export.
PDF extraction tool may also include use/save template menu 1116. Use/save template menu 1116 may enable a user to create a template in a specific document, save the created template and utilize the template for other documents. TemplateDetails0 1118 may be an example of a saved template. A user may utilize more than one template for a specific document. Use/save template menu 1116 may include remove button 1120 and apply button 1122. Remove button 1120 may enable a user to remove a selected template. Apply button 1122 may enable a user to apply a selected template to the document.
Thus, methods and apparatus for a PDF extraction tool have been provided. Persons skilled in the art will appreciate that the present invention can be practiced by other than the described embodiments, which are presented for purposes of illustration rather than of limitation, and that the present invention is limited only by the claims that follow.
