Blind watermark extraction can be fulfilled without referring to the original host image. Therefore it has much prospective applications in the fields of information hiding and copyright protection. Based on blocking Discrete Cosine Transform (DCT) and blocking Discrete Wavelet Transform (DWT)

the proposed blind watermarking algorithm uses an embedding strategy of adaptively selecting either relationship embedding or quantization embedding. The relationship embedding comes from the observation that both the magnitudes of coefficients concerned in the relationship usually have the same change trend when the image is attacked. This strategy remarkably improves the robustness of watermark extraction. In addition

dither modulation is used in quantization embedding

by which the tradeoff between transparency and robustness of watermarking can be easily adjusted. Simulation results show that the algorithm is fairly robust against the attacks such as noise

filtering

compression

and contrast enhancement

and it maintains satisfying transparency.