earticle

영어

This paper presents a RST (rotation, scaling and translation) invariant reversible watermarking method for 2D vector maps. Firstly, the proposed algorithm selects two reference vertices to calculate the normalized quantization step. Then, for each vertex, the Euclidean distance between a reference vertex and the vertex is divided into equal segments using the normalized quantization step. According to the segment which the vertex is divided into, a watermark is embedded by moving the vertex within its corresponding segment in a revertible manner. This algorithm not only recovers the original content after watermark extraction, but also correctly extracts the embedded watermarks after RST transformations. In addition, to control the distortions introduced by watermark embedding, the embedding parameter is carefully selected. Theoretical analysis and experimental results show that the proposed scheme provides RST invariance property, and good reversibility, invisibility, computational complexity and data capacity.