Abstract
Arabic diacritic marks represent efficient carriers to hide information in plain text. The ability of diacritics to be invisibly superimposed on each other when typed multiple times consecutively makes them suitable for robust data-hiding applications. In this paper, we propose two different algorithms to map secret messages into repeated diacritics in a non-wasteful fashion, where the number of extra diacritics is defined in fixed and variable size fashions. Therefore, the size of the outputted text is decided by the encoding flexibility. Both steganographic algorithms are characterized by several advantages over their existing counterparts. Finally, we provide a detailed performance analysis of both algorithms in terms of embedding capacity, robustness and file-size measures.