Tamper Detection Technique for Text Images based on Vowels and Unicode Zero Length Characters

A novel watermarking technique for the tamper detection of English text images is proposed in this paper. The frequency of maximum occurring vowel in every sentence is counted and converted to Unicode Zero Width Characters (ZWCs) by using a lookup table. These ZWCs and total length of each sentence are added at the end of the sentence. ZWCs of the Hash value of the cover text is calculated and inserted at the end of the cover text. These values are extracted from the received data on the receiver side for the tamper detection. The hash value of the received text as well as frequency of maximum occurring vowel of each sentence are again calculated and compared to their extracted corresponding values to prove the authentication of the image. Comparison with existing state-of-the-art techniques shows the effectiveness of the proposed technique.