Text Hiding in Coded Image Based on Quantization Level Modification and Chaotic Function

Abstract

A text hiding method in codded image is presented in this paper that based on quantization level modification. The used image is transformed into wavelet domain by DWT and coefficient of transform is partitioned into predefined block size. Specific threshold has been used to classify these blocks into two types named smooth and complex. Each type has its own method of text hiding (binary data), for smooth blocks, secret bits which represent the text data are switched by the bitmap. In order to reduce distortion, the quantization levels are modified. To reach extra embedding payload the quantization level could carry extra two bits depending on other threshold. The complex block carry one data bit on each block and quantization levels are swapped to reduce distortion with bitmap flipping. The proposed method result shows a high signal to noise ratio, with studying capacity as important in this work.