一种点约束分块的矢量地理数据精确认证算法
作者简介:任娜(1981-),女,博士,讲师,研究方向为空间数据安全。E-mail:renna1026@163.com
收稿日期: 2014-11-12
要求修回日期: 2014-12-02
网络出版日期: 2015-02-10
基金资助
国家自然科学基金项目(41301413)
江苏省自然科学基金项目(BK20130903)
测绘遥感信息工程国家重点实验室项目(12I02)
江苏高校优势学科建设工程项目
An Accurate Authentication Algorithm Based on Point Constraint Block for Vector Geographic Data
Received date: 2014-11-12
Request revised date: 2014-12-02
Online published: 2015-02-10
Copyright
脆弱水印可以实现对数据内容真实性的认证,并对发生变化的位置进行定位,鉴此,本文提出了一种点约束分块的矢量地理数据精确认证算法。在认证内容嵌入过程中,首先,按照点约束的方法对矢量地理数据进行分块,并将每块的数据点进行空间位置关系的“之字形”排序,建立数据点之间的位置关系;然后,将相邻点生成的脆弱水印信息嵌入到当前点。在内容精确认证时,比较原始的水印信息与提取的水印信息,判断二者是否相同来鉴别数据是否更新。将本文算法与均匀分块的方法进行了对比实验,结果表明,本文提出的认证算法能实现在数据更新时的精确认证,并可实现要素删除的精确认证,在检测数据变化时可对修改位置进行定位与标记。
任娜 , 吴维 , 朱长青 . 一种点约束分块的矢量地理数据精确认证算法[J]. 地球信息科学学报, 2015 , 17(2) : 166 -171 . DOI: 10.3724/SP.J.1047.2015.00166
Fragile watermarking technique has been viewed as an effective method to achieve content authentication, which not only detect any modifications that occurred, but also locate the modified areas. Based on fragile watermarking technology, an accurate authentication scheme based on point constraint block is proposed, which detects the malicious attacks with high accuracy while ensuring to locate exactly the tampered area for vector geographic data. Our innovative strategy is based on point constraint block and uses the spatial relationships between the elements of data points. In the authentication information embedding process, vector geographical data is divided into blocks according to the method of point constraint block, and the spatial positional relationship between the data points of each block is reordered by the "Zig-Zag" pattern, so as to establish an organized positional relationship between data points. Then, for each point, the fragile watermark information is generated by its adjacent point, and is embedded into the current point. In the process of content authentication, the extracted watermark information is compared with the generated watermark information, and the comparative result is used to judge whether the data have been updated. The proposed algorithm is furthermore compared with the method that is based on a uniform block, and the experimental results show that the proposed authentication algorithm can accomplish accurate authentication when the data is updated, and it has the ability to achieve the accurate authentication of deleted elements. Meanwhile, when data modifications are detected, it can locate and mark the modification positions.
Fig. 1 Flowchart of accurate authentication algorithm图1 精确认证算法的流程图 |
Fig. 2 Original data图2 原始数据 |
Fig. 3 Visual results before and after embedding a fragile watermark (partial enlarged view)图3 嵌入脆弱水印前后的数据可视化对比结果(局部放大图) |
Tab. 1 Error analysis表1 误差分析 |
精度位变化大小 | 数据点个数(个) | 占总数比例(%) |
---|---|---|
0 | 12081 | 35.24 |
0-1 | 14015 | 40.88 |
1-2 | 6750 | 19.69 |
2-3 | 1436 | 4.19 |
>3 | 0 | 0 |
Fig. 4 Accurate authentication results图4 精确认证结果 |
Fig. 5 Authentication results after adding elements to watermarked data elements图5 含水印数据增加要素后的认证结果 |
Fig. 6 Authentication results after translating elements to watermarked data elements图6 含水印数据平移要素后的认证结果 |
Fig. 7 Authentication results after deleting elements from watermarked data elements图7 含水印数据删除要素后的认证结果 |
The authors have declared that no competing interests exist.
[1] |
|
[2] |
|
[3] |
|
[4] |
|
[5] |
|
[6] |
|
[7] |
|
[8] |
|
[9] |
|
[10] |
|
[11] |
|
[12] |
|
[13] |
|
[14] |
|
[15] |
|
[16] |
|
[17] |
|
[18] |
|
/
〈 |
|
〉 |