A fragile zero watermarking schema to check integrity of relational databases based on discrete cosines transform


Creative Commons License

ŞAHİN Y. , ULUTAŞ G. , İMAMOĞLU M. B.

PAMUKKALE UNIVERSITY JOURNAL OF ENGINEERING SCIENCES-PAMUKKALE UNIVERSITESI MUHENDISLIK BILIMLERI DERGISI, vol.24, no.5, pp.887-897, 2018 (Journal Indexed in ESCI) identifier

  • Publication Type: Article / Article
  • Volume: 24 Issue: 5
  • Publication Date: 2018
  • Doi Number: 10.5505/pajes.2018.22120
  • Title of Journal : PAMUKKALE UNIVERSITY JOURNAL OF ENGINEERING SCIENCES-PAMUKKALE UNIVERSITESI MUHENDISLIK BILIMLERI DERGISI
  • Page Numbers: pp.887-897

Abstract

Nowadays rapid increase of the internet usage makes the data easily accessible. On the other hands, this situation arises another problem like protection of the database integrity. In recent years, database integrity protection becomes more popular and number of the works in this field is increasing in each day. Most works which are related with relational databases hides watermark information, which is generated from the current database, into the same database. This situation causes the data corruptions, which may be tolerable, on the database itself. Proposed method applies zero database watermarking scheme by transferring watermark information as a side information. The method divides the database into groups as the first step and then Discrete Cosine Transform (DCT) is applied by the method on each group which is represented by a matrix. Computed DCT coefficients are labeled as zero or one according to their signs. After computing coefficients and labeling operations for each groups, the watermark information for whole database will be obtained by concatenating groups' watermarks. Computed DCT coefficients will change if any modification has been occurred on the database and any attack could be detected at the group level. DCT is used by the method to obtain watermark information from the database at first time in the literature. Experimental results indicate that the proposed method can detect tampering on the database with finer accuracy compared to similar works.