Exploitation Of The Digital Watermarking System Based On Image Security

The robust DCT-based watermarking algorithm is studied in this paper. The grayscale image and RGB color image are researched respectively. The program is processed in the software-MATLAB. Then, the digital watermarking system is developed. In this paper, the binary BMP image is chosen as the embedded watermark. For the grayscale image, it only has one channel, in which the watermark is embedded and extracted. For the RGB color image, it has three channels. In this paper, the RGB scale is changed into YcbCr scale. Then, the Y channel is extracted, in which the watermark is embedded and extracted. The digital watermark embedding and extraction system is carried out before the following experiment parts:First is the research on the parameters for embedding and extracting variables, including the embedded watermark strength K, the number of image blocks, different watermark sizes. The second is on verification of robustness of the algorithm, namely, the ability to resist attacks, including shearing, altered, JPEG compression, Gaussian noise, printing-scanning. Finally, by the MATLAB software GUI design, the digital watermark embedding and extraction systems are developed.The results of the influence of parameters of embedding and extraction show that the algorithm can embed a certain size of the watermark image into the original image and the watermark can be extracted out. The greater of the embedded strength K, the quality of the watermarked image is poorer. And the K has nothing to do with the similarity of the extracted watermark and the original watermark. The greater number of image blocks, the quality of the image is poorer. In addition, the similarity is greater. The relationship of the watermark and the original image is as follows:the original image size is M*M, the watermark size n is less than M/n, n is the number of image blocks. If the watermark is larger than this range, the embedding and extraction can’t be completed.The results of robust validation study show that this algorithm is highly robust. After shearing, altered, JPEG compression, Gaussian noise attacking, the watermark can also be extracted. After the printing-scanning attacking, the value of gamma of the scanner is changed to get a good result. The similarity of the extracted watermark and the original watermark is improved.According to the results of the research of the algorithm, the embedding and extraction software menu bar, visual interface design and signal to noise ratio, the calculation of the similarity functional are designed in the digital watermark embedding and extraction software development process. The digital watermark embedding and extracting software for the grayscale and RGB color image is developed.