Research Of Image Information Hiding Capacity

At present, the digital watermarking and information hiding techniques have formed a theoretical basis, at the same time, a large number of digital watermarking algorithms had been proposed. The capacity of digital watermarking has become a major concern. The capacity, the invisibility, as well as the robustness of watermarking interact with one another. On condition of a certain capacity, it is the main point that how to improve the robustness of the watermarking, and maintain a higher detection rate without affecting the quality of the original image.This paper focuses on the capacity of information hiding and digital watermarking. There are three contributions in this thesis as follows:1. An algorithm of capacity adaptive digital image watermarking based on JPEG quantity was put forward. First, from the principle of JPEG quantity, the embedding positions could be selected. Then, the maximum watermark amplitude and watermark capacity were obtained by using the human vision system and image feature. At last, a simple amending polarity method was used to embed and extract the watermark. The simulation results show that this algorithm not only show good performance for embedding and compression requirements, but also have good invisibility and robustness. And this algorithm has realized the adaptive capacity. Thus, the relationship among the invisibility, robustness and capacity has been well balanced.2. Based on RGB color, a multi-channel image watermarking algorithm was proposed. The image could embed in three different watermarks, which were resilient to cropping, noising, bluring, compression and other attacks. Each watermark can afford specific attacks, so this algorithm has a more overall robustness. Compared to a single watermarking algorithm, this algorithm has larger capacity and robustness for copyright protection.3. A blind watermarking algorithm for color images was put forward, which used the brightness of the image to embed information. Combine with the texture characteristics of different images to select each sub-block's hiding positions. Then put down the image texture features as a key for extraction. With good invisibility and robustness, this algorithm has realized the adaptive capacity.