TITLE:
Spatial Image Watermarking by Error-Correction Coding in Gray Codes
AUTHORS:
Tadahiko Kimoto
KEYWORDS:
Gray Code; Error-Correcting Code; Digital Watermark; Spatial Domain; JPEG DCT-Based Compression
JOURNAL NAME:
Journal of Signal and Information Processing,
Vol.4 No.3,
August
8,
2013
ABSTRACT:
In this
paper, error-correction coding (ECC) in Gray codes is considered and its
performance in the protecting of spatial image watermarks against lossy data
compression is demonstrated. For this purpose, the differences between bit patterns
of two Gray codewords are analyzed in detail. On the basis of the properties, a
method for encoding watermark bits in the Gray codewords that represent signal
levels by a single-error-correcting (SEC) code is developed, which is referred
to as the Gray-ECC method in this paper. The two codewords of the SEC code
corresponding to respective watermark bits are determined so as to minimize the
expected amount of distortion caused by the watermark embedding. The stochastic
analyses show that an error-correcting capacity of the Gray-ECC method is
superior to that of the ECC in natural binary codes for changes in signal
codewords. Experiments of the Gray-ECC method were conducted on 8-bit
monochrome images to evaluate both the features of watermarked images and the
performance of robustness for image distortion resulting from the JPEG
DCT-baseline coding scheme. The results demonstrate that, compared with a conventional
averaging-based method, the Gray-ECC method yields watermarked images with less
amount of signal distortion and also makes the watermark comparably robust for
lossy data compression.