pdf2Data

Many PDF documents businesses need to process, such as registration forms, invoices etc. follow a common structure. If we take the example of an invoice document, addresses, purchase order numbers and similar document elements tend to be located in one place, and only the content such as item descriptions, quantities and cost of items change from invoice to invoice.

iText pdf2Data offers an easy way to extract data from such PDF documents by defining areas and rules in a template which correspond to the content you want to extract. The template can then be visually validated with other documents to confirm data is recognized correctly, before being parsed by the pdf2Data SDK to process all subsequent documents matching that template.

Unlike AI-based alternatives, you don’t need hundreds of samples and intensive supervision to train the recognition process. The content recognition is controlled by the template you configure, meaning no training is required before you can begin extracting data. You only need one example document to enable data extraction from all subsequent documents.

AI recognition has other disadvantages too. Any changes to the required output (such as adding a new field) will require models to be retrained, and multiple language support is minimal at best. Documents using the same layout but containing content in different languages can give wildly inconsistent results.

iText pdf2Data on the other hand suffers from none of these drawbacks. Making modifications to templates is quick and easy, and it offers excellent language support.

By using the intuitive browser-based pdf2Data template creator, it’s easy to create a template for data extraction. Simply create a template PDF based on a sample document, by defining selectors for areas of interest. Selectors are configurable rules to detect different types of content for extraction.

Many selectors are available to define, including Date, Time, Image, Barcode, Table, Price etc. enabling pdf2Data to intelligently recognize and extract data and other content. The selectors can be configured to detect:

• page range and the position on the page
• specific font styles, font color, and text patterns
• fixed keywords next to the data
• automatic recognition of table structures

The pdf2Data template creator was designed to allow non-developers such as business users or functional analysts to define and modify templates as required, enabling more collaborative workflows.

Want to try it out? We have an online demo of pdf2Data to test with an example document, or one you upload yourself.

The recognition process is based on the following steps:

Step 1. Upload a sample PDF document (this will become our template).
Step 2. Select data in the document you would like to extract and define relevant extraction rules (selectors) for the correct data extraction.
Step 3. Upload any other PDF document based on the same template and confirm your data was recognized correctly.
Step 4. Start using the template in the pdf2Data server-side component. You can integrate it into your document workflow as a Java or .NET library, or as a command-line application, enabling you to process potentially millions of documents with ease.

Below you can see an example of using the pdf2Data SDK to parse a pre-defined template. After loading the license file required to enable pdf2Data to work, you can parse a document against your template and extract the data with just a couple of lines of code.

1
2
3
4
5
6
7
8
9
10
11
12
13
14
// Make sure to load license file before invoking any code
LicenseKey.loadLicenseFile(pathToLicenseFile);
 
// Parse template into an object that will be used later on
Template template = Pdf2DataExtractor.parseTemplateFromPDF(pathToPdfTemplate);
 
// Create an instance of Pdf2DataExtractor for the parsed template
Pdf2DataExtractor extractor = new Pdf2DataExtractor(template);
 
// Feed file to be parsed against the template. Can be called multiple times for different files
ParsingResult result = extractor.recognize(pathToFileToParse);
 
// Save result to XML or explore the ParsingResult object to fetch information programmatically
result.saveToXML(pathToOutXmlFile);

1
2
3
4
5
6
7
8
9
10
11
12
13
14
// Make sure to load license file before invoking any code
LicenseKey.LoadLicenseFile(pathToLicenseFile);
 
// Parse template into an object that will be used later on
Template template = Pdf2DataExtractor.ParseTemplateFromPDF(pathToPdfTemplate);
 
// Create an instance of Pdf2DataExtractor for the parsed template
Pdf2DataExtractor extractor = new Pdf2DataExtractor(template);
 
// Feed file to be parsed against the template. Can be called multiple times for different files
ParsingResult result = extractor.Recognize(pathToFileToParse);
 
// Save result to XML or explore the ParsingResult object to fetch information programmatically
result.SaveToXML(pathToOutXmlFile);

Data is extracted in XML format, such as the example below:

<?xml version="1.0" encoding="UTF-8"?>
<elements>
  <data name="DATE">
    <text x="61.4" y="519.83" width="38.56" height="8.0" page="1">08/12/2016</text>
    <text x="96.28" y="477.57" width="38.56" height="8.0" page="2">16/01/2017</text>
  </data>
  <data name="END_USER_ADDRESS">
    <text x="102.25" y="612.39" width="98.84" height="40.0" page="1">Angela Merkel
To the att. of Angela Merkel
059-X025 KucheMacherStraße
71060 Sandelfängen Germany</text>
  </data>
  <data name="FAX">
    <text x="486.04" y="727.53" width="58.16" height="9.0" page="1">+32 92 70 33 75</text>
  </data>
</elements>