Research On Rich-information Security Watermarking Algorithm For Vector Geographic Data

With the explosive growth of spatiotemporal big data in recent years,data applications linked by time,space and attributes have flourished,including spatial media data,remote sensing image data,geodetic survey and heavy magnetic survey data.As an important component of spatiotemporal big data,vector geographic data plays a pivotal role in geographic information,land planning and infrastructure construction because of its high data accuracy,high production cost and high use value.However,as a kind of digital product,vector geographic data is easy to transmit and tamper with,and the convenience of the network also brings security problems to vector geographic data transmission,especially in response to natural disasters and other situations when vector geographic data transmission is urgently needed,it is easy to face problems such as data leakage and tampering.It is necessary to pay attention to the problem of secure transmission of vector geographic data,so watermarking technology is introduced to combine vector geographic data and watermarking algorithm.Most of the existing watermarking algorithms based on vector geographic data focus on embedding algorithms,while ignoring the preprocessing process of original watermark information and the self-optimization process of watermark extraction results.Moreover,the amount of watermark information in most watermarking algorithms is small,and the amount of copyright information embedded in vector geographic data is limited.Therefore,the security of the watermarking algorithm for vector geographic data can be improved from two aspects: the original watermark information preprocessing process and the watermark extraction result selfoptimization process.In this paper,the rich information security watermarking algorithm for vector geographic data is studied,and the digital watermarking technology of vector geographic data is studied in depth from two aspects: watermark information construction and optimal detection,which improves the security guarantee ability of vector geographic data.The main research contents of the paper are as follows:(1)After the vector geographic data model and spatial characteristics are summarized,the problems in the watermark system based on the vector geographic data watermark embedding system and detection system are studied,which provides a theoretical basis for the subsequent introduction of the rich information security watermarking algorithm based on vector geographic data.(2)Based on the performance improvement strategy of vector geographic data watermarking embedding algorithm,a vector geographic data watermarking algorithm based on watermark information preprocessing is proposed.In order to prevent the thief from obtaining the watermark information,the original watermark information is encrypted.QR code is used to compress the original rich-information watermarking,and then the visual password is used to further encrypt the watermark information.Finally,an improved discrete Fourier watermarking algorithm is used to embed the preprocessed watermarking information.Experimental results show that the algorithm has good invisibility and robustness.(3)Based on the performance improvement strategy of vector geographic data watermarking detection algorithm,a vector geographic data watermarking algorithm based on self-optimization of watermark extraction results is proposed.In order to enable the receiver to further obtain clear and accurate watermark information,the extracted watermark information is optimized by removing noise points.In this paper,a denoising algorithm combining octal neighborhood algorithm and connected domain algorithm is used to remove single and multiple connected noise points.Experimental results show that the algorithm has good denoising ability.