| Summary: | 碩士 === 玄奘大學 === 資訊科學學系碩士班 === 95 === Due to the rapid growth on computer science and Internet technology, it becomes easier to access or to disseminate digital multimedia through Internet and World Wide Web. Therefore, the protecting of intellectual property right is also getting more important. There are many researches in digital watermarking based on spatial domain and frequency domain. However, most of them embedded the watermark in the whole image, there is no way to extract the watermark from part of the image cropped from original. Therefore, we propose a new blind DWT-based image watermark and use the location tag to solve the cropped problem. We also use the error correcting code to improve the robustness of against lossy data compression. The experiment results show that the image protected via our method can extract the watermark if the location tag can be detected even suffer from cropped and image compression attack. For those images of containing subject, like portrait, our method provides more robustness for protecting intellectual property right.
|
|---|
