Abstract:

Concealing information in an image is increasingly used not only to convey information unnoticed, but also to mark the image itself. Marking an image with a visually visible mark is a simple and effective way of securing rights to the image in question. However, currently available watermark removal methods capable of detecting and removing such an embedded watermark invisible to the human eye are insufficient for business applications.

The aim of this paper is to provide a brief description of a method for embedding concealed information in an image with controllable properties of the underlying information. The method consists of a system for embedding and retrieving information using Artificial Neural Networks (ANN). Despite its simplicity, the method presented has satisfactory results for use in the protection of intellectual property or the covert transmission of information for business purposes.