Markov matrix and entropy based tamper detection technique for text images

Abstract EN

In this paper, a novel watermarking technique for the tamper detection of text images is pro- posed.

Entropy of every sentence is computed and Markov matrix using the occurrences of the characters is used to generate a character pattern.

Entropy and character patterns are converted to Unicode Zero Width Characters (ZWCs) by using a lookup table.

The ZWCs of entropy of each sentence is embedded at the end of every sentence after terminator.

ZWCs of the character patterns are embedded in the end of the text of the image.

On receiver side, ZWCs are extracted and converted to numerical form using the same lookup table.

Entropy of every sentence and character patterns are recalculated and compared with extracted values for tamper detection.

Comparison of technique with existing state-of-art techniques shows the effectiveness of the proposed technique.