A Data Hiding Scheme Based on LBPROT Pattern

Texture classification is mainly used for segmentation of texture regions and content-based access to image databases. Lately these texture classification patterns have been applied in an additional area of computer science, the data hiding field. In this paper the use of Rotation - Invariant Local Binary Pattern (LBPROT) is described, in order to invisibly embed secret data into a cover image. Pixel values are modified in such a way that the texture satisfies the message requirements. The method is presented and studied in terms of capacity, image quality and robustness to geometrical attacks. This semi-blind data hiding scheme in spatial domain, was proven to be robust only in rotation attack, strictly rotated by angles of kπ/2. The LBPROT method is compared to the Local Binary Pattern (LBP), which is also used for data hiding. Comparison revealed that both LBPROT and LBP schemes perform well in images with smooth areas and can be used for authentication and secret communicatio...

• Type: Journal
• Year: 2011
• Where: HOUJOI
• Authors: E. E. Varsaki, V. Fotopoulos, A. N. Skodras
• Keywords: Data Hiding, LBPROT, Steganography, Texture Classification