الفهرس 
IntroductionOnly 14 pages are availabe for public view
 |  | from | 125 |  |  |
| | | from | 125 | | |
Abstract
With the rapid development of computer and communication networks, digital multimedia reproduction and distribution are becoming extremely easier and faster. However, these advances also afford unprecedented opportunities to pirate copyrighted digital multimedia products. Digital watermarking has been proposed as a means to claim and prove ownership of the source image and to protect digital media against theft. This thesis is concerned with the design of blind robust image watermarking for copyright protection and owership verification information is in the form of a logo image, generally smaller than the host image and maybe embedded by either linear addition or quantization systems. There are two important advantages of embedding a logo image as watermark data, first, the extracted image can be correlated with the originally embedded image by human observer. Second, the existence of a visual logo in the image may be a better proof of ownership than a high statistical correlation value.
Since we propose blind watermarking techniques, two different approaches are considered in this thesis. The first is a blind embedding approach, where the host image acts as interference at the decoder. it is based on the principle of spread spectrum communication, in which we propose an adaptive direct sequence spread spectrum DSSS embedding using a perceptual model of the human visual system HVS in the DCT domain called watson`s perceptual model . The issue of blind detection is addressed. Three different decoder are considered and adjusted to be used with our proposed embedding process. The performance of the proposed embedding and decoding processes is evaluated under image mainpulations and under JPEG compression and compared to other watermarking schemes. The experimental results show that an improvement in the decoding in terms of robustness is obtained.
The second system is proposed by considering watermarking as a communication with side information, where the knowledge of the host image is exploited at the encoder to reduce the host interference. we implement a blind image watermarking system based on the scalar costa scheme SCS and uses quantization for encoding and decoding of the watermark.
Finally, a new watermarking algorithm that uses Watson`s perceptual model and based on a hybrid combination between the scalar costa scheme SCS quantization system and spread spectrum SS concept called scalar costa scheme spread transform SCS-ST is proposed. The performance of the proposed quantization systems is evaluated and compared to other quantization based methods and to our proposed DSSS system. The experimental results show that SCS-ST has a better performance than the other watermarking schemes.