Blind Watermark Approach for Map Authentication Using Support Vector Machine

Abstract

This paper presents a blind and robust watermark approach for authentication 2D Map based on polar coordinates mapping and support vector machine is presented. The proposed system is composed of three phases. Firstly, in the preprocessing phase, the proposed algorithm mapped all vertices into polar coordinate system. Then, in the support vector machine phase, the watermark portable points will be chosen using support vector machine to reduce the number of these points which increases the imperceptibility without any effect on the robustness of the watermark. Afterwards, in the watermarking algorithm phase, the watermark is embedded into the map vertices using the random table of the decimal valued of the polar coordinates through the digit substitution of the decimal part. Finally, in the theoretical analysis and experimental results shows that the presented approach is robust against a various attacks including rotation, scaling, and translation. The proposed approach attained high imperceptibility. © Springer-Verlag Berlin Heidelberg 2013.