A Blind Watermarking Algorithm Based on Modular Arithmetic in the Frequency Domain

Abstract

Robustness is the important issue in watermarking, robustness at the same time with blind watermark recovering algorithm remains especially challenging. This paper presents a combined DWT and DCT still image blind watermarking algorithm. The two-level DWT are performed on the original image, the low-frequency sub-band is divided into blocks, the DCT is performed on the every block, the DCT coefficients of every block are sorted using Zig-Zag order, the DCT low-frequency coefficient is selected as embedding watermarking. The watermarking signals are embedded into the selected embedding points using the modular arithmetic. The watermark recovering is the inverse process of the watermark embedding, according to the answer of the modular arithmetic, we can estimate the value of embedded the watermark. The algorithm is compared with a pure DWT-based scheme Experiment results shown that the proposed algorithm is robust to many attacks such as JPEG compression, addition noise, cropping, JPEG compression, median filter, rotation, and resize etc. Proposed algorithm is shown to provide good results in term of image imperceptibility, too.