Blended Learning Using Virtual RealityEnvironments

Abstract

Abstract:The huge explosion of information over World Wide Web forces us to use information security methods to keep it away from intruders. One of these security methods is information hiding method. Advantage of this method over other security methods is hiding existence of data using carrier to hold this data embedding inside it. Image-based information hiding represents one of widely used hiding methods due to the image capability of holding large amount of data as well as its resistance to detectable distortion. In last decades, statistical methods (types of stego-analysis methods) are used to detect existing of hidden data. Therefore, areas that have color variation (edges area) are used to hide data instead of smooth areas. In this paper, Corners points are proposed to hide data instead of edges, this to avoid statistical attacks that are used to expose hidden message. Additionally, this paper proposes clearing least significant bit (CLSB) method to retrieve data from stego-image without sending pixels' map; this will increase security of the proposed cornerbased hiding method. Experimental results show that the proposed method is robust against statistical attacks compared with edge-and sequential-based hiding methods. SVM classifier also confirms the outperformance of the proposed method over the previous methods by using Corel-1000 image dataset.