Docmarking: a system for marking text documents
Docmarking is a unique system for embedding digital watermarks into text documents. It allows creating a digital or physical document copy that is almost indistinguishable from the original yet exactly identifies the user or the device that was the intended recipient.
Features and advantages
Docmarking is based on research results in the areas of steganography, digital image processing, and machine learning. The marking system builds on the methods for text detection in images and image-based text classification (the digital watermark is embedded in the text areas that were found in the image).
Docmarking has a number of advantages compared to competing technologies. Watermark extraction does not require the original document. The system supports embedding a watermark in the same scanned document multiple times, and the previous watermark is erased when the new one is being embedded.
- Marking algorithms based on machine learning.
- Support for documents of all formats.
- Embedding a watermark when a document is displayed on a screen or printed.
- Standalone setup and work on the client side.
- Real time registering of the embedded watermarks on the server.
- Centralized 24/7 monitoring of the connected client devices.
Who is Docmarking target audience?
- Government entities and public offices.
- Companies that would like to enforce their guides for handling classified documents.
Supported operating systems
Windows (32-bit, 64-bit), Linux (64-bit), including Astra Linux 1.5/1.6 SE.