DOX for Capturing

The reliable document recognition

DOX for Capturing automates the digitization of paper-based documents, while extracting data and information from them in a process supported by artificial intelligence. This way, it ensures a fast and state-of-the-art execution of mailroom workflows, laying the groundwork for further processing of incoming information in the best possible way.



  • Outstanding performance

    Due to its easy scalability, millions of documents can be quickly converted, while processing performance can be further increased with multiple servers and the CPUs.

  • Advanced recognition skills

    Due to machine learning and advanced language processing, texts and other contents (i.e. bar codes) can be extracted even from poor quality images or faxes, in nearly 200 languages.

  • Automatic correction and validation

    After recognition, the system automatically exchanges or modifies the fields based on the data validation and normalization rules.

  • Simple management

    Easy installation and configuration, requiring no special training. An unlimited number of users can utilise the services of a single server at a time.

How does DOX for Capturing work?

DOX for Capturing is an intelligent document recognition system to identify incoming contents and convert them into electronic files that can be archived for the long-run.

1 Read

All types of content from files or scanners, including office documents, image formats, emails, and attachments, are automatically processed in one single flow. The web scan and mobile scan modules enable the use of service also via web browser or mobile app.

2Character recognition

DOX for Capturing compiles multi-page documents or sets of documents from the images, the content and data of which are automatically extracted and validated.
Intelligent character recognition technology is capable of processing documents in almost 200 languages. Text content is extracted even in case of images and faxes of poor quality. In addition to structured or semi structured documents, handwritten texts as well as 1D and 2D barcodes can be recognized and processed.

3 Verification

The module enables various verification options available either installed or web-based from anywhere. The rate of matching under which visual control of the user is required can be defined, and if necessary data can be changed manually.
The group verification feature allows verifying predefined points or critical numbers within an entire document set by displaying the questionable versions in a single control window.

4 Export

According to the set of rules defined by the users, DOX for Capturing automatically forwards the extracted data and images to the place of further processing or archiving, in the format specified.

Simple monitoring and reporting


    • HTML5 administration and monitoring interface - accessible 24/7 from anywhere
    • Multi-level administration option
    • Monitoring of processing time, load and productivity on project level
    • Automatic notifications for critical failures or imminent database overflow
    • Various standard reports that can be customized as needed

The self-learning ability of DOX for Capturing allows users to "teach" the recognition of flexible, irregular documents, such as delivery notes or receipts, with leaving the possibility of editing, fine-tuning or discarding results if needed. Auto-learning technology not only ensures continuous improvement of the recognition capability, but also reduces support and maintenance needs.



  • Just like any other products built on DOX Enterprise platform, also DOX for Capturing can be smoothly and quickly integrated and connected to existing processes.
  • The applied integration layer supports all standard interface connections.
  • With standardized connectors that can be configured, it can be linked to the most common document archiving or content management systems, like: Saperion, OnBase, Opentext Livelink or IBM FileNet.