DWT-QR based blind image watermarking method using firefly algorithm

Abstract

Digital watermarking is a method used to protect intellectual property rights and copyrights. This study represents a blind method based on Discrete Wavelet Transform (DWT) and QR decomposition. In this method, the cover image is firstly divided into non-overlapping blocks, and then the blocks with low standard deviation are selected to embed the watermark. After each of the selected blocks is decomposed by DWT, low frequency sub-bands are subjected to QR decomposition. The first row of the R matrix resulting from QR decomposition is used to embed the watermark bits. Firefly Algorithm (FA) is utilized to optimize the gain factor of embedding, which is effective in determining the robustness and imperceptibility of the watermarking scheme. To appraise the performance of the study, Peak Signal to Noise Ratio (PSNR) is used to measure the perceptual quality and Normalized Correlation (NC) is used to evaluate robustness. According to experimental results, the perceptual quality of the proposed method is found to be higher than that of similar studies in the literature. In addition, it is established that the robustness of the method against attacks such as noise, cropping, compression, and filtering is better.