Concealing and Extracting with Verifying Text in Frontal Face Image

Abstract

The transmission of data via public networks is often necessary, but such data may be easily compromised. One solution is to hide the true message in an innocent-looking image transmitted as the data. This paper describes an algorithm for concealing and extracting text in a frontal face image based around the hiding of the message. We can easily validate the hidden text by using the included hash. Our experiments use different facial images and the same hidden text to produce images that are exactly the same as the originals. This method can be easily applied to other areas, such as the protection of intellectual property.