Improved Intelligent Partitioning Scheme for Audio Data Hiding

Abstract

Reversible data hiding method can be implemented not only for images but also on other media such as audio. Data dimension changes are required to make an image-based data hiding method works in the audio domain and vice versa. For instance, 16-bit audio must be converted into an 8-bit image. An intelligent partitioning (IP) method was utilized to split the audio signals into segments of partitions. It modifies the size of the expansion media in audio data; thus, it applies to the image data hiding scheme. In this experiment, the image data hiding scheme is supported by an improved IP-based method. Two IP methods are proposed, aiming to maintain audio quality after embedding. The experimental results find that the second proposed IP method produces higher capacity values, and the first proposed IP method generates better quality results. This study also discovers that audio genres influence the performance of the proposed methods.