Self-adaptive Watermarking Algorithm Based On DWT And HVS

The rapid development and wide application of multimedia and network technology have made the storage, processing and distribution of digital multimedia products much easier and faster, and the digital copyright has become more important at the same time. Encryption techniques can be used to protect digital data. However, after the decryption of the data, the data is in the clear and no longer protected. Digital watermarking is another effective solution that is regarded as the last line of defense. It has become a hotspot in the field of information security recently.This thesis introduces digital watermarking technology and wavelet transform theory, and analyzes the image scrambling and gray-scale image decomposed into bit-planes technology; it makes a serious research on the application of the image iterative blending technology and human visual system (HVS) characteristics in digital watermarking technology.An algorithm based on DWT and iterative blending was proposed. The application of digital image iterative blending method to watermarking technology, we choose LL2 and HH1 bands as the embedded position after second-level wavelet decomposition, and both carry out three times iterative embedding, finally achieve the double-watermarking iterative embedding. The watermarking was scrambled before it was embedded. The usefulness of private keys and iterative parameter enhance the security of watermarking information. We use iterative degree to compromise the watermarking robustness and imperceptibility.An algorithm based on DWT and human visual system was proposed. It uses the application of the HVS masking factors of frequency, luminance and texture in the watermarking to give the new noise sensitivity mode. We judge the single visual influence factor after calculate the product of the three noise sensitivity mode, and select the maximum value of the product and two single factors as the final embedded intensity coefficient of the noise sensitivity mode. Associate with wavelet transform and gray-scale image decomposed into bit-planes technology, accomplishing self-adaptive embedding of gray-scale image.The result of the emulation experiment indicates that the two proposed watermarking algorithms could bring the better watermarking robustness and imperceptibility. The algorithms are robust to JPEG lossy compression, cropping, noising, median filtering and so on, and it is useful in protecting copyright.