Skip to content

Latest commit

 

History

History
89 lines (64 loc) · 2.55 KB

README.md

File metadata and controls

89 lines (64 loc) · 2.55 KB
Raw

PDF invoice data extractor

Features

  • INPUT: PDFs with OCR text layer
  • OUTPUT: Key-value pairs of invoice data
  • Extracts invoice data by evaluating multiple regular expressions against the plain text of the pdf document
  • Provides a GUI for data validation
  • Electron app with node.js backend
  • Vendor matching based on vat number (CHE-xxx.xxx.xxx), IBAN from separate CSV-list
  • Export format options: JSON, XML
  • Focus on scanned and OCR'd documents with a scan resolution of 300 dpi.
  • NO OCR! This software does not perform any OCR on the PDF documents. It merely extracts the existing text content.

Screenshots

GUI

Motivation

  • I needed a way to extract textual data from PDF documents specifically invoices.
  • All available solutions required the upload of the PDF documents to a webservice which was not an option.
  • Furthermore I required a solution to easily validate the extracted data with a GUI.

Data extraction

Regular expressions are used to extract the field values from a text only version of the PDF document.

PDF extraction pipeline

                extract text from
                text layer in PDF           regex
1 PDF document -------------------> 1 text -------> 1-N key-value pairs
                                                     (extracted_data)

Validation pipeline

                     user validation
1-N key-value pairs ----------------> 1-N key-value pairs
 (extracted_data)                      (validated_data)

JS-File invokation

  • main.js
    • public/web/index_frontend_controller.html
      • ../build/pdf.js
      • viewer.js
        • ../build/pdf.js
      • ../extractor/FrontendController.js
        • ../extractor/SuppliersLoader.js
        • ../extractor/Sidebar.js
          • ../extractor/SidebarField.js
            • Validator{..}.js (ValidatorClass)
        • ../extractor/Queue.js
    • ./public/extractor/BackendController.js
      • ./PdfExtractJob.js
        • Extractor{..}.js (ExtractorClass)
          • ./Extractor.js (Base class)
      • ./SuppliersLoader.js
      • ./Queue.js

Installation and Usage

  1. Clone git archive
  2. cd into the directory
  3. on the command line: npm install
  4. start with: npm start

npm-dependencies

These dependencies are installed by running npm install

  • express: 4.17.1
  • js2xmlparser: 4.0.0
  • pdf.js-extract: 0.1.3
  • electron: 6.0.12

Packaged libraries

A pre-built version of the PDF.js viewer is included in this project.