基于动态差分扩展的强鲁棒数据库水印算法研究

doi:10.11959/j.issn.2096-109x.2023065

摘要/Abstract

摘要：

科技行业的快速发展带来信息量的暴增，各行各业都需要收集和应用大量的数据，海量数据在发挥价值的同时，给数据安全领域带来了史无前例的挑战。关系型数据库作为数据的底层存储载体之一，其存储的数据规模大、数据内容丰富、数据隐私度高。数据库的数据一旦泄露将会造成巨大的损失，保护数据库的所有权，确认数据的归属刻不容缓。对于现有的数据库水印技术来说，提高水印嵌入容量和减小数据失真之间存在固有矛盾问题，为了缓解此问题且进一步提高水印的鲁棒性，提出了一种基于动态差分扩展的强鲁棒数据库水印算法。该算法选取QR码作为水印，利用经过Haar小波变换的图像低频部分进行奇异值分解（SVD，singular value decomposition），提取部分特征值，用取余后的特征值作为待嵌入的水印序列，使得相同长度的水印序列包含更多信息，缩短了嵌入水印的长度。该算法结合自适应差分进化算法和最小差值算法选择最佳嵌入属性位，以缓解传统差分扩展技术在嵌入水印时计算效率低、数据失真大、鲁棒性差的问题，提高水印嵌入容量的同时减少了数据的失真。实验结果表明，该算法保证高水印嵌入率的同时数据失真较低，能够抵御多种攻击，具有良好的鲁棒性，追踪溯源的能力强，且与现有的算法对比优势明显，在数据安全领域具有广阔的应用前景。

关键词: 数据库水印, 差分进化, 差分扩展, SVD, Haar小波变换, QR码

Abstract:

A surge in the amount of information comes with the rapid development of the technology industry.Across all industries, there is a need to collect and utilize vast amounts of data.While this big data holds immense value, it also poses unprecedented challenges to the field of data security.As relational databases serve as a fundamental storage medium for data, they often contain large-scale data rich in content and privacy.In the event of a data leak, significant losses may occur, highlighting the pressing need to safeguard database ownership and verify data ownership.However, existing database watermarking technologies face an inherent tradeoff between improving watermark embedding capacity and reducing data distortion.To address this issue and enhance watermark robustness, a novel robust database watermarking algorithm based on dynamic difference expansion was introduced.The QR code was employed as the watermark, the SVD decomposition of the low frequency part of the image was utilized after Haar wavelet transform.By extracting specific feature values and using residual feature values as the watermark sequence, it was ensured that the same-length watermark sequence contains more information and the embedded watermark length can be reduced.Furthermore, by combining the adaptive differential evolution algorithm and the minimum difference algorithm, the optimal embedding attribute bits were selected to alleviate the problems of low computational efficiency, high data distortion and poor robustness of traditional difference expansion techniques in embedding watermarks, and to improve the embedding capacity of watermarks while reducing the distortion of data.Experimental results demonstrate that the proposed algorithm achieves a high watermark embedding rate with low data distortion.It is resilient against multiple attacks, exhibiting excellent robustness and strong traceability.Compared to existing algorithms, it offers distinct advantages and holds great potential for broad application in the field of data security.

Key words: database watermarking, differential evolution, difference expansion, singular value decomposition, Haar wavelet transform, quick response code

中图分类号:

TP393

汪天琦, 张迎周, 邸云龙, 李鼎文, 朱林林. 基于动态差分扩展的强鲁棒数据库水印算法研究[J]. 网络与信息安全学报, 2023, 9(5): 150-165.

Tianqi WANG, Yingzhou ZHANG, Yunlong DI, Dingwen LI, Linlin ZHU. Research on strong robustness watermarking algorithm based on dynamic difference expansion[J]. Chinese Journal of Network and Information Security, 2023, 9(5): 150-165.

图/表 19

图1

图2

图3

表1

嵌入数字QR码的小波变换对比Table 1 Comparison of wavelet transform with embedded digital QR code"

小波变换	QR码像素为64×64		QR码像素为128×128		QR码像素为256×256
小波变换	压缩率	绝对最大差异	压缩率	绝对最大差异	压缩率	绝对最大差异
Haar	70.7 001 953	$5 . 4001247$	$78 . 3752441$	7.3 896 444	$93 . 3868408$	$9 . 094947$
Daubechies	68.8 720 703	1.0 516 032	74.2 797 851	1.2 221 335	82.588 195	1.4 687 962
Coiflet	$71 . 508789$	1.4 861 683	75.8 422 851	7.6 738 615	84.6 649 169	1.8 301 996
Symlet	69.2 626 953	1.0 461 462	74.1 210 937	$9 . 0750518$	84.1 064 453	2.1 753 364

表1

表2

嵌入文本QR码的小波变换对比Table 2 Comparison of wavelet transform with embedded text QR code"

小波变换	QR码像素为64×64		QR码像素为128×128		QR码像素为256×256
小波变换	压缩率	绝对最大差异	压缩率	绝对最大差异	压缩率	绝对最大差异
Haar	$69 . 5751953$	5.9 685 589	$83 . 203125$	7.3 896 444	$89 . 979553$	$7 . 6738615$
Daubechies	67.8 955 071	$9 . 3791641$	77.1 911 621	1.1 937 117	81.054 687	1.4 164 767
Coiflet	68.2 617 187	1.3 672 206	79.461 669	1.4 278 214	81.561 279	1.938 388
Symlet	68.2 617 187	5.6 616 581	77.746 582	$7 . 7305807$	81.381 225	2.1 638 108

表2

图4

图5

图6

图7

图8

图9

表3

水印嵌入前后不同数据库水印算法的平均值和标准差统计结果Table 3 Mean and standard deviation statistics of different database watermarking algorithms before and after watermark embedding"

属性	原始数据库		ADE-DDEW算法		GADEW算法		RF-GADEW算法		GAHSW算法		RF-GAHSW算法
属性	Mean	Std	Mean	Std	Mean	Std	Mean	Std	Mean	Std	Mean	Std
T9	19.483	2.015	19.476	2.009	19.082	2.382	19.575	2.133	$19 . 483$	$2 . 015$	19.453	2.035
RH_9	41.548	4.151	$41 . 541$	$4 . 143$	41.383	4.601	41.476	4.285	41.572	4.139	41.563	4.148
T_out	7.4	5.317	$7 . 389$	$5 . 308$	7.201	5.528	7.523	5.203	7.443	5.348	7.432	5.329
Press_mm_hg	79.769	7.399	79.757	7.405	79.829	7.021	79.892	7.256	79.737	7.371	$79 . 748$	$7 . 385$
RH_out	755.523	14.901	$755 . 518$	$14 . 901$	755.401	14.710	755.437	14.834	755.556	14.850	755.549	14.924
Windspeed	4.041	2.451	$4 . 052$	$2 . 459$	4.284	2.649	4.184	2.392	4.086	2.411	4.067	2.428
Visibility	38.342	11.794	$38 . 35$	$11 . 789$	38.742	11.620	38.492	11.630	38.393	11.756	38.361	11.775
Tdewpoint	3.753	4.195	$3 . 747$	$4 . 204$	3.932	4.348	3.629	4.275	3.702	4.264	3.737	4.218
rv1	24.986	14.496	$24 . 972$	$14 . 511$	24.847	14.730	24.816	14.583	24.932	14.558	24.971	14.526

表3

表4

水印嵌入前后不同数据库水印算法的平均值、标准差变化和平均绝对误差统计结果Table 4 Mean, standard deviation change and mean absolute error statistics of different database watermarking algorithms before and after watermark embedding"

属性	ADE-DDEW算法		GADEW算法		RF-GADEW算法		GAHSW算法		RF-GAHSW算法
属性	D_M	D_S	D_M	D_S	D_M	D_S	D_M	D_S	D_M	D_S
T9	0.001	0.002	0.023	0.063	0.002	0.192	$0$	$0$	0.003	0.045
RH_9	$0 . 001$	$0 . 003$	0.019	0.320	0.187	0.073	0.006	0.072	0.084	0.065
T_out	$0 . 002$	$0 . 001$	0.380	0.754	0.003	0.018	0.038	0.047	0.026	0.194
Press_mm_hg	0.002	0.006	0.049	0.128	0.218	0.007	0.014	0.005	$0$	$0$
RH_out	$0 . 003$	$0 . 002$	0.020	0.870	0.010	0.501	0.106	0.023	0.008	0.073
Windspeed	$0 . 001$	$0 . 001$	0.076	0.018	0.009	0.129	0.080	0.007	0.017	0.072
Visibility	$0 . 002$	$0 . 001$	0.890	0.342	0.227	0.284	0.004	0.039	0.080	0.160
Tdewpoint	$0 . 004$	$0 . 001$	0.017	0.050	0.001	0.547	0.072	0.216	0.021	0.015
rv1	$0 . 003$	$0 . 002$	0.430	0.081	0.039	0.028	0.193	0.074	0.047	0.007
MAE	$0 . 019$		2.914		1.315		0.852		0.516

表4

图10

图11

图12

图13

表5

图14

参考文献 18

[1]	HOLT L , MAUFE B G , WIENER A . Encoded marking of a recording signal[P]. U L Patent GB2196167A, 1988.
[2]	HIMANSHU A , FAROOQ H . Development of payload capacity enhanced robust video watermarking scheme based on symmetry of circle using lifting wavelet transform and SURF[J]. Journal of Information Security and Applications, 2021,59.
[3]	SHOHIDULISLAM M , NAQVI N , ABBASI A T ,et al. Robust dual domain twofold encrypted image-in-audio watermarking based on SVD[J]. Circuits,Systems,and Signal Processing, 2021 (prepublish).
[4]	赵春雨 . 提高嵌入容量的水印算法研究[D]. 郑州:郑州大学, 2011.
	ZHAO C Y . Research on algorithm for improving the capacity of digital watermarking[D]. Zhengzhou:Zhengzhou University, 2011.
[5]	ZHANG Y , YANG B , NIU X M . Reversible watermarking for relational database authentication[J]. Journal of Computers, 2006,17(2): 59-66.
[6]	HU D , ZHAO D , ZHENG S . A new robust approach for reversible database watermarking with distortion control[J]. IEEE Transactions on Knowledge and Data Engineering, 2018,31(6): 1024-1037.
[7]	GUPTA G , PIEPRZYK J . Reversible and blind database watermarking using difference expansion[J]. International Journal of Digital Crime and Forensics, 2009,1(2): 42-54.
[8]	JAWAD K , KHAN A . Genetic algorithm and difference expansion based reversible watermarking for relational databases[J]. Journal of Systems and Software, 2013,86(11): 2742-2753.
[9]	邱升红 . 面向关系数据库的可逆水印方法研究[D]. 西安:西安电子科技大学, 2018.
	QIU S H . Research on reversible watermarking methods for relational databases[D]. Xi'an:Xi'an University of Electronic Science and Technology, 2018.
[10]	宋岩, 沈泉江, 杨洪山 . 基于布谷鸟算法的可逆数据库水印方案[J]. 计算机应用与软件, 2021,38(12): 304-313.
	SONG Y , SHEN Q H , YANG H S . A reversible database watermarking scheme based on cuckoo algorithm[J]. Computer Application and Software, 2021,38(12): 304-313.
[11]	孔嘉琪, 王利明, 葛晓雪 . 基于模拟退火和粒子群混合改进算法的数据库水印技术[J]. 信息网络安全, 2022,22(5): 37-45.
	KONG J Q , WANG L M , GE X X . Database watermarking technique based on simulated annealing and particle swarm hybrid improvement algorithm[J]. Information Network Security, 2022,22(5): 37-45.
[12]	SHI Y , LI X , ZHANG X ,et al. Reversible data hiding:advances in the past two decades[J]. IEEE Access, 2016,4(10): 3210-3237.
[13]	GE C , SUN J , SUN Y ,et al. Reversible database watermarking based on random forest and genetic algorithm[C]// 2020 International Conference on Cyber-Enabled Distributed Computing and Knowledge Discovery (CyberC). 2020.
[14]	陈青, 夏兰婷, 卜莹 . 基于两级奇异值分解的鲁棒水印算法[J]. 应用科学学报, 2020,38(6): 966-975.
	CHEN Q , XIA L T , BU Y . Robust watermarking algorithm based on two-level singular value decomposition[J]. Journal of Applied Sciences, 2020,38(6): 966-975.
[15]	关虎, 张桂煊, 张树武 ,等. 基于二维码的鲁棒图像水印技术及应用研究[J]. 有线电视技术, 2018(10): 20-26.
	GUAN H , ZHANG G X , ZHANG S W ,et al. Research on robust image watermarking technology and application based on QR code[J]. Cable TV Technology, 2018(10): 20-26.
[16]	KLEMA V , LAUB A . The singular value decomposition:its computation and some applications[J]. Automatic Control IEEE Transactions on, 1980,25(2): 164-176.
[17]	STORNR P K . Differential evolution:a simple and efficient adaptive scheme for global optimization over continuous spaces[R]. 1995.
[18]	TIAN J . Reversible data embedding using a difference expansion[J]. IEEE Transactions on Circuits ＆ Systems for Video Technology, 2003,13(8): 890-896.

攻击比例	删除攻击		修改攻击		插入攻击
攻击比例	文本QR码	数字QR码	文本QR码	数字QR码	文本QR码	数字QR码
10%
90%