Research On Watermarking Algorithm Of Color Image

In recent years, the spread of the Internet and the advance in digital techniques for transmitting and storing information have made it possible to copy and edit images, audio, video and other types of multimedia data. Multimedia data by digital format can be duplicated without loss of quality and changed easily at low cost. These situations cause the illicit actions such as interception, duplication, misuse and unauthorized modification of information. One approach to address the problem is watermarking technique. The concept of digital watermarking is associated with the data-hiding technique. Digital watermarking has been proposed as a way to identify the source, creator, owner, distributor, or authorized consumer of multimedia data.In this thesis, the current watermark developing status, basic characters and problems of digital watermark technology were analyzed, and the general frame of digital watermark processing system was discussed and three watermarking schemes are proposed.In the first scheme, presents a color image watermarking method with blind detection based on amplitude modulation. On the basis of the visual masking of human visual system, the embedded positions were chosen adaptively. Basically, the watermark was embedded by modifying the pixel values in the blue channel of an image, while the watermark retrieval is achieved by using a prediction technique based on linear combination of nearby pixel values around the embedded pixels. We use a technique of gaussian weighting marks by averaging the luminance of its own pixel and its neighboring pixels, and use it as an additional scaling factor to tune the strength of embedding watermark. Experimental results show that the presented scheme is invisible and robust against various signal processing such as cropping, compression, filtering, noise adding, image enhancement and geometrical attacks.In the second scheme, a new algorithm is proposed; the watermark was embedded by modifying the pixel values in the blue channel of an image, while the watermark retrieval is achieved by using a prediction technique based on support vector regression model. The model is trained by learning the relationship between the selected pixel and statistical feature of neighboring area. The technique of Gaussian weighting marks by averaging the luminance of its own pixel and its neighboring pixels is used as an additional scaling factor to tune the strength of embedding watermark. Experimental results show that the presented scheme is invisible and robust against various processing such as cropping, compression, filtering, noise adding, image enhancement and geometrical attacks.In the third scheme, a blind watermarking algorithm for digital image based on DWT and DCT is proposed. The experimental result indicates that the algorithm is visually imperceptible and robust to general image processing.