New Pixel Sorting Method for Palette Based Steganography and Color Model Selection

In this article, we propose a new steganographic approach for palette-based images. This new method has the advantage of embedding secure data, within the index, the palette or both, using special sorting scheme. In the index, information can be embedded using a double pseudo random key that selects blocks and pixels within the image randomly. A stego-bit is embedded at each pixel of the selected blocks. Embedding is accomplished by selecting the closest color using a weighted distance measure that contains the desired modulus-bit. This algorithm also gives you the option of selecting the best color models (i.e. YCbCr and HSV) for embedding purposes. The presented technique also incorporates the use of a cover image measure in order to select the best of the candidates for the insertion of the stego information. This has the advantage of introducing lower distortion and higher embedding capacity than EzStego and Fridrich’s method. Computer simulations have shown that the presented algorithms carry with them the following advantages: 1) Introduces 8-10 times less distortion than EzStego [4] and 4-5 times less distortion than Fridrich’s [3] method. 2) Provides additional security through a simple selective color and cover image algorithm. 3) Provides the closest color substitutes using different color models and a weighted distance measure. 4) Finally, it offers an increased capacity by embedding only on the best cover images.