Research On Adapt Digital Image Watermarking Technology

With the rapid development of network and multimedia technology, the copyright protection and anti-tampering issue of image data becomes more and more important. Digital watermarking technology, as an effective method of copyright protection and anti-tampering has been a hotspot in academia. The thesis mainly studies digital watermarking algorithms. It introduces the origin, conception, characteristics, and application fields of digital watermarking algorithm. Based on the full analysis of existing digital image watermarking algorithms, the processing of watermarking signal is divided into two course about pretreatment and postprocessing. The main contribution of this thesis is to design two adaptive image watermarking algorithms used to different conditions toward two different classes of watermark signals.A binary image is the focus in the first algorithm based on human visual characteristics, the algorithm proposed a new watermarking scheme which based on DWT and DCT domain. In order to ensure the clarity of algorithm and improve robust, through post-processing of Hopfield neutral network the extracting watermark signal is resumed , and the extracting accuracy of watermark is increased. The simulation results show that this algorithm can accurately recover the watermarking signal and effectively resist the attacks of noise, tampering, filtering, cropping and JPEG compression.Audio signal as watermarking signal is the main object in the second algorithm. Based on the differences of sensitivity on cepstrum coefficients, an image watermarking algorithm based on hearing characteristic is bring forward. Through selecting proper Cepstrum coefficient, audio signal is embedded in two dimension cepstrum region. The simulation results demonstrate that this algorithm has better clarity and robust.These two methods can be used to protect the copyright of digital image.