벡터 맵 데이터의 안전한 전송과 저장을 위한 압축 및 크립토마킹 기법

Abstract

With the increase in the number of applications using digital vector data and the development of surveying techniques such as CAD, GIS, video games or computer graphics and so on, a large volume of vector data having high accuracy and precision is being produced. However, to achieve their effective transmission while preserving their high quality, these high capacity vector data needs to be compressed. Also most of vector data a high commercial value due to industrial or technical expertise with regard to their design of them. Accordingly, It is important for storage and transmission of vector data to protect securely. This thesis presents two important themes about compression and security for vector data for GIS. First, this paper presents a compression method for vector map data that has the high-level accuracy and exact precision demanded for vector map data handling. To achieve this, the proposed method a priori divides a map into rectangular local regions and classifies the bits of each object in the local regions to three types of bins. Then, it encodes objects progressively using the properties of the classified bins, such as adjacency and orientation, to obtain the optimum compression ratio. The types of bins are defined as category bin, direction bin, and accuracy bin. Experimental results verify that the proposed compression method can encode vector map data constituting less than 20% of the original map data at a 1cm accuracy level and that constituting less than 9% at a 1m accuracy level. In addition, its coding efficiency is greater than that of conventional methods, whereas its complexity is lower for close to real-time applications. About the vector map data security as the second theme, as a new security technique combined a semi-blind watermarking and perceptual contents encryption based on vector map data, cryptomarking is proposed for copyright protection, user authorization and data security for vector map. To achieve this, a semi-blind watermarking that embed watermark using sum and mean of vertices’ Euclidean distances of each vector object is designed for satisfying perceptual invisibility and robustness against variable geometrical attacks. And a progressive contents encryption about each vector map region, each object in a region, and each vertex in a object for vector data having a huge data volume, is designed to satisfy real-time processing and high lever of perceptual security in this paper. Then, the target vector map data can be protected securely by the cryptomarking technique applying proposed watermarking and encryption algorithms simultaneously and using mutual independence the these two algorithms. Experimental results verify that the proposed cryptomarking can strengthen the weak points of the watermarking algorithm against attacks to remove the embedded watermark and verify that this method can minimize the chances of encryption key leakage in the course of watermark detection by enabling the copyright owner to detect watermark on encrypted vector map domain. Also, its watermarking and encryption algorithms have lower complexities for close to real-time applications.