A zero-watermarking scheme with embedding timestamp in vector maps for Big Data computing

Digital watermarking is a frontier technology for information security and also one of the most effective approaches to protect the copyright security of vector maps. Zero-watermarking, a sort of constructed watermarking, is different from the traditional embedded watermarking. Due to the lack of embedding watermark into the vector data, zero watermark exists an obvious difficulty to resist interpretation attack. In this paper, we present an improved zero-watermarking scheme for vector maps by introducing the timestamp method to resist the interpretation attack. The proposed scheme does not need any modifications on the original map data, providing the following feature: adding timestamp to the binary sequence of the scrambled watermarking image, performing the exclusive OR operation on scrambled sequence and extracted feature points, and conducting a reverse extraction detection and a timestamp validation on the zero-watermarking sequence. The experimental results show that the proposed scheme could not only resist interpretation attack in zero-watermarking effectively but also is of robustness and invisibility for the geometric attacks and map updating attacks.