Data Hiding Method for VQ image using Tree-Structured Links

Abstract

In recent years, various data hiding techniques using Vector Quantization (VQ) have been proposed. However, these methods sometimes could not use optimal output vectors and quality of the output image is thus degraded. In this study, we propose a new data hiding technique using VQ. This method uses tree structures to represent connectivity and distance between output vectors. One node corresponds to an output vector and another one nearest to it are linked by an edge. A suitable vector is chosen for data hiding amongst the output vector itself and its linked vector depending on the embedding information. Simulation results show that our method indicates higher PSNR and SSIM value than the conventional method under the same amounts of embedded data.