Digital media technology With the rapid development of computer technology and the increasing popularity of the Internet, all kinds of media contents are converted into digital forms to improve the efficiency and accuracy of information expression。 But the copyright protection of digital contents is also more difficult。 GIS digital vector map is a kind of national basic geographic information data which involves national security。 So as a special kind of digital products, the copyright protection of GIS digital vector map is more important than general intellectual property protection [1-2]。 

Digital watermark embedding technology is one of the most important methods of digital products copyright protection。 Compared to the copyright protection of general digital products, the watermark embedding technology of GIS digital vector map should have the following characteristics for its particularity in accuracy and application: 1) high accuracy: The embedding of watermark shouldn’t affect the using accuracy of vector map, especially in positioning accuracy。 2) imperceptibility: The watermark will not be perceivable in a variety of platforms after embedding into vector map, including distortion in vision and measurement。 It means that the existence of watermark shouldn’t affect the visual effects of vector map or lead to the element deformation。 So it is indispensable to inspect the qualities of watermarked GIS vector data [1, 3]。80682

At present, the research on watermarking algorithms for data quality is mainly focused on error analysis which mainly used in watermarking for digital images。 The indicators such as average absolute difference, mean square error, signal to noise ratio, peak signal to noise ratio, correlation quality and histogram similarity are used in measuring image distortion。 A perception measurement method was presented based on spatial domain JND [4]。 This model considered the effects of 3 main HVS attributes of frequency sensitivity, luminance sensitivity and contrast masking。 Christian (1996) presented a model named masked peak signal to noise ratio (MPSNR) which is a measurement method comprehensively considering the effect of several factors of the image, and is formed by integrating human visual characteristics [5]。 Voloshynovskiy (2001) presented a model named weighted peak signal to noise ratio (wPSNR) which gave different visual weights to different regions [6]。 Watson (1993) presented a model of image quality evaluation based on Watson standards [7]。 Watson standards are mainly used in measuring errors of block-based DCT transform coefficients by a certain sensitivity threshold [7]。 A method of image quality evaluation was presented which based on human visual system [8]。 It merges the errors of original image and degraded image and calculates SNR of human visual system using the characteristics that wavelet transform matches the multi-channel features of human visual system [8]。 However, the relevant research on data quality of watermarked vector data is relatively limited。 Users usually feel suspect of watermarking algorithms in different degrees for the lack of data quality inspection methods, error analysis methods and error analysis tools。 And it affected the application of watermarking to some extent。 Therefore, in this paper we will do some research on data inspection of watermarked GIS vector data on the basis of analyzing principles of watermark embedding, referring to methods of error analysis in image field。 And some data quality inspection tools are developed, so that it can provide effective support both in theory and technology for data inspection of watermarked GIS vector data。

Paper form  — 《Huang, Li ; Zhou, Wei; Jiang, Rui; Li, Anbo  Source: 2010 18th International Conference on Geoinformatics, Geoinformatics 2010, 2010, 2010 18th International Conference on Geoinformatics, Geoinformatics 2010》

REFERENCES

[1] D。 H。 Xu, C。 Q。 Zhu, and Q。 S。 Wang, “A Survey of the Research on Digital Watermarking for the Vector Digital Map,” Geomatics World, pp: 42-48, 2007。 [In Chinese]