Study on benchmark system for copyright marking algorithms of GIS vector data

GIS vector data is widely used in economic construction, social life, management decision, military affairs, et al. The application of GIS vector data faces some unsolved issues such as digital copyright protecting, data authenticating, and data source tracing, etc. These issues could be obstacles restricting the further usage of digital vector data. The illegally forged, distributed or tampered data could pose a threat to the healthy development of geography information industry. With the increasing of copyright problems, copyright protection of GIS vector data has been drawing more and more attention. The technique of watermarking provides potential solutions for the problem. Comparing with the watermarking schemes designed for general multimedia data, watermarking GIS vector data has some distinct features due to the special data structures and application environments of GIS vector data. The special organization and storage characteristics of GIS vector data makes it difficult to simply apply existing copyright marking techniques to GIS vector data. Some research and applications have been done on copyright marking techniques of GIS vector data, mainly about basic theory, marking generation, marking embedding and marking detection. Today, the GIS watermarking community needs some advanced attacks and fair benchmarks in order to compare the performances of different copyright marking algorithms of GIS vector data. Study on Benchmark System for copyright marking algorithms of GIS vector data permit to find the weaknesses of an algorithm and consequently trigger further research in order to overcome the problem. In this paper, firstly, the construction of benchmark datasets is studied from two aspects of data source and data organization. Secondly, some attacks of GIS vector data watermarking is analyzed and the automated benchmark method for watermarking performances based on simulated attacks is proposed. Lastly, the system architecture of benchmark system for copyright marking algorithms of GIS vector data is presented.