Research On Fragile Watermarking Algorithm For Vector Map

In recent years,with the rapid development of digital map surveying and mapping disciplines,digital vector map with its rich content and convenience of operation has been widely employed in military surveying and mapping,GPS navigation,marine development,digital city and many other areas.However,the vector map information security issues come one after another,some users of vector map hope to certificate the authenticity and integrity of the content before using the received vector map.In this regard,how to effectively carry out the content authentication of vector map is the unremitting pursuit of the researchers worldwide.Fragile watermarking technology has become a popular choice to solve these problems.In the current domestic and foreign studies,most of the research is to block or group the features of the vector map and embed the fragile watermark.In the watermark authentication phase,the block or group will be marked out in which the tampered content is located and then the better tampering accuracy is achieved.However,the vector map may lose the entire feature group after the batch deletion attack against features.These watermarking algorithms that perform content authentication by group can not effectively detect and mark this tampering attack.In order to solve the above problem,this thesis proposes a fragile watermarking scheme of anti-deleting features for 2D vector map authentication.The proposed algorithm first divide the vector features of the vector map according to the record number,mark the position of each feature and use the feature group correlation technique designed in this thesis to embed the position information of each feature group in the adjacent group.Then applying the log-polar mapping method and the folding method to construct the hash function to generate the fragile watermark with rotation,scaling and translation(RST)invariance.Afterwards,the RST invariant fragile watermark embedding algorithm is used to embed the watermark into the vector map.In the watermark authentication phase,the location of each feature and the correlation mark of the feature group can accurately identify the grouping of the original map feature and the important location information for the missing group.And then the extracted watermark is compared with the reconstructed watermark information to achieve the exact localization of the tampering content.Through the experimental analysis and comparison,the fragile watermarking algorithm proposed in this thesis has good invisibly and high tamper localization accuracy.The embedding of watermark has less perturbation on the vector map and can resist the addition,deletion and modification of vertex or linear feature.It can resist the feature rearrangement and vertex inversion attack,and detect the batch features deletion attack effectively on the basis of the RST invariance of the previous watermarking algorithm.