Image Watermarking Algorithm Based on q-logarithm Component

Abstract

A new embedding component for image watermarking in the spatial domain is presented in this paper as a solution to copyright protection and to verify the real owner of digital image. The embedding component called q-logarithm is extracted from a host color image, and is used directly to embed watermark information. In the extraction process of the embedded watermark, the original version of non-watermarked component is first estimated from the average value of watermarked components in a small image area, so that the blind watermark extraction can be achieved. The watermarked image’s quality is assessed by weight Peak Signal to Noise Ratio (wPSNR), and the extracted watermark’s accuracy is assessed by Bit Correction Error (BCR). Investigations for optimal parameters are carried out to obtain q value that gives the most accurate extracted watermark. In the experiments, the performance of the proposed method is compared to the relevant three previous methods. A well-known benchmark called Stirmark is also used to assess the robustness of the embedded watermark. It is shown accordingly that the proposed method can achieve better performance in terms of accuracy and robustness even under attacks.