Article Preview
TopIntroduction
With the development of digital-content design and processing techniques, transferring and accessing 2D vector maps via internet become more and more popular. These highly accurate geo-spatial data are of great value (Lee & Kwon, 2013; López, 2002; Niu, Shao, & Wang, 2006; Peng, Yue, Wu, & Peng, 2014). However, thanks to powerful available tools and equipment, it is very easy even for an amateur to illegally modify these data and create “perfect” forgeries. Tools that help us verify the integrity and authenticity of the 2D vector map content are thus essential.
2D vector map integrity and authenticity can be guaranteed through the use of fragile watermarking techniques (Ren, Wang, & Zhu, 2014; Shao, Wang, & Xu, 2005; Wang & Men, 2012; Wang & Men, 2013; Yue, Peng, & Peng, 2014; Zheng, Li, Feng, & Liu, 2010; Zheng & You, 2009). In most cases of fragile watermarking, the original content are modified, and hence distorted in an irreversible way in order to embed the authentication data. The original data suffer some permanent destruction. However, due to the required high-precision nature of 2D vector maps, it is critical to reverse the embedded content back to the original version after authentication. To avoid permanent distortion, reversible (also referred to as invertible, lossless, or distortion-free) fragile watermarking, which allow the decoder to recover the original content upon authentication and locate malicious attacks have been proposed.
Currently, research on 2D vector map watermarking schemes is mainly focused on robust watermarking algorithms (Doncel, Nikolaidis, & Pitas, 2007; Lafaye, Béguec, Gross-Amblard, & Ruas, 2012; Lee & Kwon, 2013; López, 2002; Niu, Shao, & Wang, 2006; Ohbuchi, Ueda, & Endoh, 2002; Peng, Yue, Wu, & Peng, 2014; Wang, Peng, Peng, Yu, Wang, & Zhao, 2012) for copyright protection or reversible watermarking methods (Cao, Men, & Gao, 2013; Cao, Men, & Ji, 2013; Voigt, Yang, & Busch, 2004; Wang, Shao, Xu, & Niu, 2007; Wang, Zhang, & Men, 2014; Zhou, Hu, & Zeng, 2009) for content recovery. Few works have been done on fragile watermarking schemes for authentication. Ren et al. (2014) proposed a semi-fragile watermarking method that tolerates compression, noise addition and vertex deletion operations. A drawback of this method is that it may be not able to locate deleted Polylines. By embedding a group watermark and an object watermark into each object, Yue et al. (2014) described a fragile watermarking algorithm with modification type characterization. Although it guarantees a more precise detection granularity, the invisibility may be not desirable.