Research On Fragile Watermarking For 2D Vector Map

These days,2D vector maps,the fundamental data of geographical information system(GIS),have been widely applied in military,civil cartography,urban planning,navigation,etc.These vector geo-spatial data can be easily replicated,manipulated and distributed using powerful available tools and public networks.On the other hand,the ease and extent of such manipulations and potential security loopholes of the public networks emphasize the need for2 D vector map authentication techniques in applications where verification of integrity and authenticity of the 2D vector map content are essential.The high cost of acquiring these vector geo-spatial data further underscores this need.As a potential solution,fragile watermarking is paid extensive attention by researchers at home and abroad and becomes an international academic focus.The thesis discusses the key problems,i.e.,the tamper localization,security and data embedding problems,in 2D vector map fragile watermarking techniques in detail,and researches mainly on the key points in 2D vector map fragile watermarking with high tamper localization accuracy.The main contributions are as follows:Firstly,to accurately locate tampered feature groups and recover the original content after authentication,a reversible fragile watermarking algorithm for 2D vector maps based on feature marking is proposed.Because a reversible watermarking technique is applied to each group for the authentication watermark embedding,the original 2D vector map content can be recovered after authentication.Besides,a marking method based on vertex addition is described for correctly identifying the original features of each group in the watermark verification stage.Using the mark of each feature,the approach can get the exact original features for each group,and provide high tamper localization accuracy.To tolerate rotation,uniform scaling and translation operations,a rotation,uniform scaling and translation invariant fragile watermarking algorithm for 2D vector maps is presented.By exploiting the feature marking technique to store the original location of each feature,the method can locate tampered data with high accuracy.Using the hash results of the differences of the log-radiuses to calculate an authentication watermark for each feature group,and embedding the watermark using a rotation,uniform scaling and translation invariant watermarking approach,the algorithm provides both sensitivity to malicious manipulationsand invariance to rotation,uniform scaling and translation transformations.After that,to accurately locate tampered regions and recover the original content after authentication,a reversible fragile watermarking method for locating tampered blocks in 2D vector maps is presented.In the proposed approach,the vector map is divided into two categories of blocks: normal and complex.To decrease the embedding distortion for the normal blocks,we describe a reversible data hiding approach based on LSD(least-significant-digit)planes.To authenticate the complex blocks,a fragile watermarking method based on vertex insertion is proposed.While the reversible watermarking algorithm ensures original content recovery,the independent block authentication watermarks locate tampered blocks ith high accuracy.Moreover,to further decrease the computational complexity of the block-based reversible fragile watermarking for 2D vector maps,a reversible fragile watermarking algorithm is proposed based on the idea of virtual coordinates.The proposed scheme not only locates tampered blocks precisely,but also recovers the original content.In addition,this algorithm provides low computational complexity,high capacity,and good invisibility.The fragile watermarking schemes proposed in this thesis provide good performances in authentication and tamper localization.By combining the proposed reversible watermarking algorithms,these fragile watermarking methods can not only recover the original content after authentication and ensure the accurate use of the vector map data,but also have high data capacity and good invisibility.The proposed schemes have good practicability.