A Study On Reversible Watermarking

With rapid development and application of digital technologies, network multimedia and communication technologies, copyright protections and authentications are more and more important. Digital watermarking technology provides good solutions for such problems. Ordinary watermark have bad effect on the original image after the embedding and extracting process, it is acceptable if the effect is slight. However, for certain applications, such as medical images and engineering drawings used to guide industrial productions, there cann't be any distortion on the original image after extraction processing because of the strict using fields, In this circumstance, a new watermark technology called reversible watermarking, or lossless watermarking, which can restore the original image without any distortion, is appeared. In this article, we focus on this kind of watermarking technology, the main contents of this article include:A reversible watermarking algorithm with large-capacity for color images based on edge detection operator is proposed. A hot spot in watermarking research is to achieve large amount of data capacity, Researchers employ repeating embedding method to achieve this goal. In our method, we first decomposition of the RGB components of one color image, and then select the region with complex texture using edge detection operator on each component, the region we choose is suitable for embedding data without cause excessive visual redundancy according to HVS. As the strong correlation of three components, the embedding areas are mostly the same, so after even times embedding, about 48% of the embedd pixels are restored to the original value, thus reduces the visual redundancy when achieving large amount of data capacity Experimental results show that the algorithm achieves very good results.A reversible watermarking algorithm with good visual quality based on gray-scale-shifting is proposed. With gray- scale -shifting image watermarking technology, to achieve the desired visual effect must be reduce gray-scale-shifting range. Upon the previous work, we use the formula between capacity and storage of additional information as the judgement, from the largest "peaks" P1 in histogram, find the point P1' fit the formula both from the left and right respectively, and then shift gray-scale within [P1+1, P1'+1], embed watermarks in P1 and P1+1; continue to search for sub-biggest "peak" P2 and corresponding P2', embed information in accordance with the previous method until all of the gray zone is unavailable or no points meet the formula. Compared with the existing algorithms, our method effectively improves the visual quality, and avoids the complexity brought by dynamic programming approach.A reversible watermarking algorithm based on difference expansion for 2-D engineering drawings is proposed. The algorithm is based on the DE algorithm suitable for general digital images, for the feature of the irregular data in engineering drawings; we improve the algorithm module so that it can be applied to such irregular data. Taking into account the crafts constraint and error tolerance in engineering drawings, we introduce a parameterτto make error control effectively. Experimental results show that visual effect of the algorithm is very good, can resist some common attacks, and can also achieve blind watermark extraction. The complexity of the algorithm is low, and easy to implement.