DEM数字水印算法研究

发布时间：2018-05-25 22:09

  本文选题：DEM + 数字水印　； 参考：《兰州交通大学》2017年硕士论文

【摘要】：数字高程模型(Digital Elevation Model,DEM)是地理空间数据的重要组成部分,已被广泛应用于测绘、地质、矿山工程、战场仿真等领域。然而,DEM的数字化形式为地形模拟分析带来便利的同时,因其易于存储、修改、传播的特性,使其篡改、侵权更加容易,由此产生的DEM版权纠纷问题日益严重。因此,如何使DEM数据得到合理共享的同时,又能有效保护所有者的版权利益,已成为一个迫在眉睫的现实问题。目前,数字水印技术已成为地理空间数据版权保护的重要手段,为DEM版权保护研究提供了可行的技术途径。本论文根据DEM的数据结构特点,结合其数字水印算法的特征和应用要求,分别对规则格网和不规则三角网两种类型DEM数据的数字水印算法进行了分析研究。主要研究成果总结如下:(1)对DEM数字水印特征及要求进行了分析研究。总结了目前DEM数字水印算法的研究现状,分别介绍了规则格网和不规则三角网两种类型DEM的数据结构与水印算法特性。分析了两种类型DEM数据最常遭受的水印攻击方式和数据处理方式,结合地形因子特征、数据误差与应用误差要求,对两种类型DEM数字水印算法的技术要求进行了研究总结,为DEM数字水印算法构建提供了理论基础与目标。(2)提出了一种抗高程平移和裁剪攻击的规则格网DEM盲水印算法。分析规则格网DEM最常遭受的水印攻击方式,结合其数据结构特点,算法对原始水印信息进行扩频处理,使其与高程差值形成映射关系。利用相邻格网的高程差值在高程平移攻击中的不变特性,以量化方式将扩频后水印按位嵌入到高程差值中,然后将扩张的差值分摊到相邻高程的后一位。实验分析结果表明:该水印算法具有良好的不可见性,对DEM数据精度和应用精度影响较小,对抵抗高程平移和裁剪攻击具有良好的鲁棒性。(3)提出了一种运用特征点的不规则三角网DEM盲水印算法。分析不规则三角网DEM数据结构特性,针对其最常遭受的简化攻击和裁剪攻击两种水印攻击类型,结合地形因子特征,利用不规则三角网DEM抽稀技术提取特征三角网节点,以特征三角网节点作为水印信息嵌入载体。对节点坐标进行放大处理,取有效位作为水印映射载体,根据误差控制原则,通过量化的方式将水印嵌入到合适的坐标位中。提取水印时采用投票原则,无需原始数据参与。通过实验结果分析表明,该水印算法对DEM数据精度和应用精度影响较小,具有良好的不可见性,对抵抗简化攻击和裁剪攻击具有很好的鲁棒性,误差可控,并且保证了以较少的嵌入误差代价实现水印的嵌入和良好的鲁棒性。
[Abstract]:Digital elevation Elevation Model (DEM) is an important part of geospatial data, which has been widely used in surveying, geology, mine engineering, battlefield simulation and other fields. However, the digital form of Dem brings convenience for terrain simulation analysis, because of its easy to store, modify, spread, make it easier to tamper, infringement, the resulting DEM copyright dispute is becoming more and more serious. Therefore, how to share DEM data reasonably while effectively protecting the copyright interests of the owner has become an urgent and realistic problem. At present, digital watermarking technology has become an important means of geospatial data copyright protection, which provides a feasible technical approach for the research of DEM copyright protection. According to the characteristics of DEM data structure and the characteristics and application requirements of its digital watermarking algorithm, this paper analyzes and studies the digital watermarking algorithms for two types of DEM data, regular grid and irregular triangular network, respectively. The main research results are summarized as follows: (1) the characteristics and requirements of DEM digital watermarking are analyzed and studied. This paper summarizes the current research status of DEM digital watermarking algorithm, and introduces the data structure and watermarking algorithm of two types of DEM, regular grid and irregular triangular network, respectively. Two kinds of DEM data are analyzed, which are most often attacked by watermark and data processing methods. Combined with terrain factor characteristics, data error and application error requirements, the technical requirements of two types of DEM digital watermarking algorithm are studied and summarized. This paper provides a theoretical basis for the construction of DEM digital watermarking algorithm and proposes a rule-based DEM blind watermarking algorithm against height translation and clipping attacks. Based on the analysis of the most common watermarking attacks on regular grid DEM and the characteristics of its data structure, the algorithm performs spread spectrum processing of the original watermark information to form a mapping relationship with the elevation difference. Using the invariant property of the elevation difference of the adjacent grid in the height translation attack, the spread spectrum watermark is embedded into the elevation difference bit by bit in a quantized way, and then the expanded difference value is distributed to the next bit of the adjacent elevation. The experimental results show that the watermarking algorithm has good invisibility and has little effect on the accuracy of DEM data and its application. This paper presents a DEM blind watermarking algorithm using feature points to resist height translation and clipping attacks. This paper analyzes the characteristics of DEM data structure in irregular triangulation network, aiming at the two kinds of watermark attack types, namely simplification attack and clipping attack, and combining with terrain factor feature, the feature triangulation node is extracted by irregular triangulation DEM thinning technique. The feature triangulation node is used as the embedding carrier of watermark information. The node coordinate is amplified and the effective bit is taken as the watermark mapping carrier. According to the error control principle, the watermark is embedded into the appropriate coordinate bit by quantization. When extracting the watermark, the voting principle is adopted, without the participation of the original data. The experimental results show that the proposed watermarking algorithm has little influence on the precision of DEM data and its application accuracy, and has good invisibility, good robustness against the simplified attack and clipping attack, and the error can be controlled. At the same time, the watermark embedding and good robustness are guaranteed at the cost of less embedding error.
【学位授予单位】：兰州交通大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP309.7;P208

【参考文献】

中国期刊全文数据库 前10条

1 刘爱利;丁浒;田丹;王丽;王少峰;;基于坡度和坡向分析的DCT域DEM数字水印算法[J];武汉大学学报(信息科学版);2016年07期

2 张黎明;闫浩文;齐建勋;张永忠;;运用特征点的矢量空间数据盲水印算法[J];测绘科学;2016年04期

3 张黎明;闫浩文;齐建勋;张永忠;;基于DFT的可控误差矢量空间数据盲水印算法[J];武汉大学学报(信息科学版);2015年07期

4 任娜;朱长青;佟德宇;崔翰川;;地理空间数据数字水印系统设计与实现[J];南京师范大学学报(工程技术版);2013年04期

5 符浩军;朱长青;赵毅;任娜;;面向网络环境的地理空间数据数字水印模型[J];测绘学报;2013年06期

6 王奇胜;朱长青;符浩军;;利用数据点定位的矢量地理数据数字水印算法[J];测绘学报;2013年02期

7 朱长青;符浩军;缪剑;王玉海;;一种自适应的数字栅格地图可见水印算法[J];测绘学报;2013年02期

8 张驰;李安波;闾国年;林冰仙;;以夹角调制的矢量地图可逆水印算法[J];地球信息科学学报;2013年02期

9 符浩军;朱长青;李永练;徐惠宁;管大鹏;;抗几何变换的二值栅格地图盲水印算法[J];测绘科学技术学报;2013年01期

10 张佐理;夏守行;郑胜峰;王本轶;;基于点模式匹配的矢量地图水印算法[J];计算机应用与软件;2013年02期

中国博士学位论文全文数据库 前5条

1 张黎明;地理空间矢量数据数字水印算法研究[D];兰州交通大学;2016年

2 孙鸿睿;矢量地图无损数字水印技术和算法研究[D];中南大学;2013年

3 王志伟;DEM数字水印模型与算法研究[D];解放军信息工程大学;2011年

4 杨成松;矢量地理数据数字水印模型与算法研究[D];解放军信息工程大学;2011年

5 任娜;遥感影像数字水印算法研究[D];南京师范大学;2011年

中国硕士学位论文全文数据库 前2条

1 陆尘;不规则三角网DEM数字水印算法研究[D];南京师范大学;2015年

2 缪志修;基于机载LiDAR数据的DEM抽稀算法研究[D];西南交通大学;2010年

，

本文编号：1934820