STM Journals Peer-reviewed research from Consortium e-Learning Network

Journal of Image Processing & Pattern Recognition Progress

Data Embedding Technique for Palette Scale Image using Derived Genetic Algorithm (DEP-DGA)

  1. P.M. Siva Raja
  2. E. Baburaj

Abstract

Steganography method is used for exchanging the hidden information. It is defined as the study of hidden communication that deals with the existence of communicated messages. The existing methodology for minimizing additive distortion in steganography fails in measuring the level of distortion for the given source palette digital image. Perfect sphere shaped palette digital images are not suitable for selection, as cover models because the hidden data might be easily observed even after specific modification by embedding method. To shape constraints in palette digital images and to observe the distortion, data embedding technique using derived genetic algorithm for palette scale image has been proposed. Row major order is used to embed huge quantity of stego message on source palette digital image. Four bits of data hiding message is embedded per byte of the source palette digital image onto the rightmost 4 bit of each pixel. Mutation is applied on the embedded palette digital image. In the process of embedding, dimension of the data hiding message is followed by the content of the original palette based digital image. Reverse procedure is followed during decoding. Derived Genetic algorithm is used to enhance the security level without any extraction of data by attackers. Several statistical parameters computed are compared with the existing additive distortion syndrome trellis code method using Statlog Dataset from UCI repository. DEP-DGA approach obtained better results in terms of security level of hidden information, 3.50 % improved throughput on using the palette based digital image for embedding and minimize the RMS error between the stego message and the palette based original image.

Keywords

Support