低复杂度的增强图像来源检测算法

doi:10.19665/j.issn1001-2400.2021.01.011

西安电子科技大学学报 ›› 2021, Vol. 48 ›› Issue (1): 96-106.doi: 10.19665/j.issn1001-2400.2021.01.011

低复杂度的增强图像来源检测算法

王俊祥¹(),黄霖¹(),张影¹(),倪江群^2,³(),林朗³()

1.景德镇陶瓷大学机械电子工程学院,江西景德镇 333403
2.中山大学数据科学与计算机学院,广东广州 510006
3.东南数字经济发展研究院,浙江衢州 324000

收稿日期:2020-08-10 出版日期:2021-02-20 发布日期:2021-02-03
作者简介:王俊祥(1985—),男,教授,博士,E-mail: wjx851113851113@163.com|黄霖(1997—),男,景德镇陶瓷大学硕士研究生, E-mail: 15083983519@163.com|张影(1996—),女,景德镇陶瓷大学硕士研究生, E-mail: 18279855273@163.com|倪江群(1963—),男,教授,博士, E-mail: issjqni@mail.sysu.edu.cn|林朗(1994—),男,硕士, E-mail: ll_linlang@163.com|王俊祥(1985—),男,教授,博士,E-mail: wjx851113851113@163.com|黄霖(1997—),男,景德镇陶瓷大学硕士研究生, E-mail: 15083983519@163.com|张影(1996—),女,景德镇陶瓷大学硕士研究生, E-mail: 18279855273@163.com|倪江群(1963—),男,教授,博士, E-mail: issjqni@mail.sysu.edu.cn|林朗(1994—),男,硕士, E-mail: ll_linlang@163.com
基金资助:
国家自然科学基金(61762054);国家自然科学基金(62062044);国家自然科学基金(U1736215);国家自然科学基金(61772573);广州市科技项目(201707010029);广州市科技项目(201804010265);衢州市科技项目(2019K12);国家自然科学基金(61762054);国家自然科学基金(62062044);国家自然科学基金(U1736215);国家自然科学基金(61772573);广州市科技项目(201707010029);广州市科技项目(201804010265);衢州市科技项目(2019K12)

Algorithm for the detection of a low complexity contrast enhanced image source

WANG Junxiang¹(),HUANG Lin¹(),ZHANG Ying¹(),NI Jiangqun^2,³(),LIN Lang³()

1. School of Mechanical and Electrical Engineering,Jingdezhen Ceramic Institute,Jingdezhen 333403,China
2. School of Data Science and Computer science,Sun Yat-sen University,Guangzhou,510006,China
3. Southeast Digital Economic Development Institute,Quzhou,324000,China

Received:2020-08-10 Online:2021-02-20 Published:2021-02-03

摘要/Abstract

摘要：

随着多媒体技术的飞速发展,增强图像因视觉效果好而被广泛使用。常规图像增强算法包含直方图均匀化、伽马校正等。然而,近期提出了一种基于对比度增强效果的可逆信息隐藏算法。该算法在往载体中嵌入一定比例秘密信息后,可生成与常规图像增强算法视觉效果一致的含密增强图像。此类增强图像的出现为后续不分辨增强图像来源而直接使用的操作带来了巨大的安全隐患。基于此,提出一种可识别可逆信息隐藏图像的低复杂度增强图像来源检测算法。通过分析可逆信息隐藏图像和多种常规对比度增强图像在直方图分布上的区别,设计了4个高效的特征,然后采用高效的支持向量机分类器完成增强图像的来源检测。实验结果表明,在多种对比度增强图像来源鉴定的场景下,这种算法均可获得较之当前主流方案更准确、更稳定的结果,优势明显。

关键词: 图像分析, 图像识别, 机器学习, 图像增强, 可逆信息隐藏, 支持向量机, 图像分析, 图像识别, 机器学习, 图像增强, 可逆信息隐藏, 支持向量机

Abstract:

With the rapid development of multimedia techniques,enhanced images,such as mobile phone pictures,are widely used due to its good visual quality,In general,conventional image enhancement algorithms include histogram equalization,gamma correction,and so on.Recently,a new reversible data hiding algorithm with the content enhancement function (denoted as RDH_CE) is proposed,which could achieve identical visual enhancement quality as conventional enhancement algorithms do when a certain amount of secret data is embedded.It is easy to have some security risk when one enhanced image with some suspicious code embedded in it is utilized.Therefore,an effective algorithm for identifying some suspicious RDH_CE and other regular ones (i.e.,histogram equalization and gamma correction) is proposed in this paper.By analyzing their implementation process,four features are designed and then SVM is employed to identify their source.Experimental results indicate that the proposed scheme can achieve a better performance compared with other state-of-art algorithms in terms of the accuracy and stability.

Key words: image analysis, image recognition, machine learning, image enhancement, reversible data hiding, support vector machine (SVM), image analysis, image recognition, machine learning, image enhancement, reversible data hiding, support vector machine (SVM)

中图分类号:

TP309

王俊祥,黄霖,张影,倪江群,林朗. 低复杂度的增强图像来源检测算法[J]. 西安电子科技大学学报, 2021, 48(1): 96-106.

WANG Junxiang,HUANG Lin,ZHANG Ying,NI Jiangqun,LIN Lang. Algorithm for the detection of a low complexity contrast enhanced image source[J]. Journal of Xidian University, 2021, 48(1): 96-106.

图/表 16

图1

图2

图3

图4

图5

表1

图6

图7

表2

图8

图9

图10

表3

表4

表5

表6

参考文献 22

[1]	冯玉颖, 王俊平, 马塾亮, 等. COPLIP新模型及其图像增强新算法[J]. 西安电子科技大学学报, 2018,45(1):66-71.
	FENG Yuying, WANG Junping, MA Shuliang, et al. New COPLIP Model and Its Image Enhancement Algorithm[J]. Journal of Xidian University, 2018,45(1):66-71.
[2]	张明, 白松浩. 利用上下边界估计的图像增强方法[J]. 西安电子科技大学学报, 2015,42(1):149-154.
	ZHANG Ming, BAI Songhao. Image Enhancement Based on Estimating the Upper and Lower Boundary[J]. Journal of Xidian University, 2015,42(1):149-154.
[3]	KAMEL M S, GUAN L. Histogram Equalization Utilizing Spatial Correlation for Image Enhancement [C]//Proceedings of SPIE:1199.Bellingham:SPIE, 1989: 712-723.
[4]	WU H T, DUGELAY J L, SHI Y Q. ReversibleImage Data Hiding with Contrast Enhancement[J]. IEEE Signal Processing Letters, 2015,22(1):81-85.
[5]	WU H T, TANG S, HUANG J, et al. A Novel Reversible Data Hiding Method with Image Contrast Enhancement[J]. Signal Processing:Image Communication, 2018,62:64-73.
[6]	QIAN R, LI W, YU N, et al. Image Forensics with Rotation-tolerant Resampling Detection [C]//Proceedings of the 2012 IEEE International Conference on Multimedia and Expo Workshops.Washington:IEEE Computer Society, 2012: 61-66.
[7]	KUMAWAT C, PANKAJAKSHAN V. A Robust JPEG Compression Detector for Image Forensics[J]. Signal Processing:Image Communication, 2020,89:116008.
[8]	BHARDWAJ D, PANKAJAKSHAN V. A JPEG Blocking Artifact Detector for Image Forensics[J]. Signal Processing:Image Communication, 2018,68:155-161.
[9]	SHAN W, YI Y, QIU J, et al. Robust Median Filtering Forensics Using Image Deblocking and Filtered Residual Fusion[J]. IEEE Access, 2019,7:17174-17183.
[10]	YUANH D. Blind Forensics of Median Filtering in Digital Images[J]. IEEE Transactions on Information Forensics and Security, 2011,6(4):1335-1345.
[11]	HUANG H Y, CIOU A J. Copy-move Forgery Detection for Image Forensics Using the Superpixel Segmentation and the Helmert Transformation[J]. EURASIP Journal on Image and Video Processing, 2019,2019(1):68.
[12]	LI Y M, ZHOU J. Fast and Effective Image Copy-move Forgery Detection via Hierarchical Feature Point Matching[J]. IEEE Transactions on Information Forensics and Security, 2019,14(5):1307-1322.
[13]	WU Y, ABD-ALMAGEED W, NATARAJAN P. Image Copy-move Forgery Detection via an End-to-end Deep Neural Network [C] //Proceedings of the 2018 IEEE Winter Conference on Applications of Computer Vision.Piscataway:IEEE, 2018: 1907-1915.
[14]	STAMM MC, LIU K J R.Forensic Detection of Image Manipulation Using Statistical Intrinsic Fingerprints[J]. IEEE Transactions on Information Forensics and Security, 2010,5(3):492-506.
[15]	CAO G, ZHAO Y, NI R, et al. Contrast Enhancement-based Forensics in Digital Images[J]. IEEE Transactions on Information Forensics and Security, 2014,9(3):515-525.
[16]	DE ROSA A, FONTANI M, MASSAI M, et al. Second-order Statistics Analysis to Cope with Contrast Enhancement Counter-forensics[J]. IEEE Signal Processing Letters, 2015,22(8):1132-1136.
[17]	SUN J Y, KIM S W, LEE S W, et al. A Novel Contrast Enhancement Forensics Based on Convolutional Neural Networks[J]. Signal Processing:Image Communication, 2018,63:149-160.
[18]	SHAN W, YI Y, HUANG R, et al. Robust Contrast Enhancement Forensics Based on Convolutional Neural Networks[J]. Signal Processing:Image Communication, 2019,71:138-146.
[19]	CHEN Y, LYU Z X, KANG X, et al. A Rotation-invariant Convolutional Neural Network for Image Enhancement Forensics [C]// Proceedings of the 2018 IEEE International Conference on Acoustics,Speech and Signal Processing.Piscataway:IEEE, 2018: 2111-2115.
[20]	CHANG C C, LIN C J. Libsvm:a Library for Support Vector Machines[J]. ACM Transactions on Intelligent Systems and Technology, 2011,2(3):27.
[21]	BAS P, FILLER T, PEVNY T. Break Our Steganographic System:the Ins and Outs of Organizing BOSS [C]//Lecture Notes in Computer Science:6958.Heidelberg:Springer Verlag, 2011: 59-70.
[22]	GONEN, M. Receiver Operating Characteristic (ROC) Curves[J]. Journal of Clinical Nursing (Wiley-Blackwell), 2002,11(1):136.

特征	直方图均匀化	伽马矫正	RDH_CE
零点个数	增加	增加	减少
零点分布	分散于直方图中	集中在单侧	分散在直方图两侧
包络线	整体平缓(中间散落零点)	存在单侧峰(起伏大)	整体平缓(中间无零点)

特征组合	HE、G、RDH_0.03	HE、G、RDH_0.10	HE、G、RDH_0.70
f₁	67.70	72.56	81.44
f₁+f₂	92.84	93.05	96.32
f₁+f₂+f₃	93.50	93.95	97.18
f₁+f₂+f₃+f₄	95.75	96.15	97.39

检测对象	文献[14]	文献[15]	文献[16]	文中算法
HE、G、RDH_0.005	69.23	68.79	81.52	95.41
HE、G、RDH_0.030	56.67	68.78	81.57	95.75
HE、G、RDH_0.050	63.22	68.76	81.85	95.95
HE、G、RDH_0.100	64.90	68.75	80.33	96.15
HE、G、RDH_0.500	67.08	67.49	83.10	97.43
HE、G、RDH_0.700	65.90	68.53	85.40	97.39

检测对象	文献[14]	文献[15]	文献[16]	文中算法
O与RDH_0.005	91.06	64.00	76.99	95.26
O与RDH_0.030	93.18	65.81	78.48	95.05
O与RDH_0.050	93.81	66.65	77.96	95.09
O与RDH_0.100	95.31	67.98	77.84	95.18
O与RDH_0.500	98.65	76.14	84.44	95.78
O与RDH_0.700	98.93	79.02	87.60	96.39

检测对象	文献[14]	文献[15]	文献[16]	文中算法
O与HE	99.15	98.40	99.27	99.92
O与G_0.5	99.00	99.20	92.87	99.46
O与G_0.7	99.40	99.46	94.60	99.20
O与G_0.9	97.26	99.06	84.00	93.70
O与G_1.1	94.86	88.67	79.27	90.67
O与G_1.3	93.20	91.60	81.30	89.80
O与G_1.5	98.13	92.80	90.86	98.47
O与G_2.0	92.70	88.40	81.00	97.30

低复杂度的增强图像来源检测算法

Algorithm for the detection of a low complexity contrast enhanced image source

RichHTML

PDF (PC)

赞

可视化

摘要/Abstract

引用本文

使用本文

图/表 16

参考文献 22

相关文章 15

Metrics

本文评价

推荐阅读 10

[1]	乔文欣,卢昱,刘益岑,李志伟,李玺. 空天地协同的边缘云服务功能链动态编排方法[J]. 西安电子科技大学学报, 2022, 49(2): 79-88.
[2]	孙彦景,魏力,张年龙,云霄,董锴文,葛敏,程小舟,侯晓峰. 联合DD-GAN和全局特征的井下人员重识别方法[J]. 西安电子科技大学学报, 2021, 48(5): 201-211.
[3]	曾勇,吴正远,董丽华,刘志宏,马建峰,李赞. 加密流量中的恶意流量识别技术[J]. 西安电子科技大学学报, 2021, 48(3): 170-187.
[4]	张树栋,高海昌,曹曦文,康帅. 针对ASR系统的快速有目标自适应对抗攻击[J]. 西安电子科技大学学报, 2021, 48(1): 168-175.
[5]	李海,尚金雷,孙婷逸,冯青,庄子波. MC-DTSVMs的双偏振气象雷达降水粒子分类方法[J]. 西安电子科技大学学报, 2020, 47(4): 132-140.
[6]	王晓明,张淑艳,张婕,原思聪. 一种深度残差学习的含噪图像轮廓重建方法[J]. 西安电子科技大学学报, 2020, 47(3): 66-71.
[7]	闫林,刘凯,段玫妤. 一种用于点云分类的轻量级深度神经网络[J]. 西安电子科技大学学报, 2020, 47(2): 46-53.
[8]	刘志强,张铭津,池源,李云松. 一种深度学习的硬件木马检测算法[J]. 西安电子科技大学学报, 2019, 46(6): 37-45.
[9]	安亚利,周水生,陈丽,王保军. 鲁棒支持向量机及其稀疏算法[J]. 西安电子科技大学学报, 2019, 46(1): 64-72.
[10]	马苗;许西丹;武杰. 一种耦合深度信念网络的图像识别方法[J]. 西安电子科技大学学报, 2018, 45(5): 102-107.
[11]	冯玉颖;王俊平;马塾亮;李超;伍尧. COPLIP新模型及其图像增强新算法[J]. 西安电子科技大学学报, 2018, 45(1): 66-71.
[12]	李美玲;胡耀垓;周晨;赵正予;张援农;刘静;邓忠新. 支持向量机用于电离层foF2的短期区域预报[J]. J4, 2015, 42(5): 147-153+206.
[13]	蔡艳艳;宋晓东. 针对非平衡数据分类的新型模糊SVM模型[J]. J4, 2015, 42(5): 120-124+160.
[14]	张明;白松浩. 利用上下边界估计的图像增强方法[J]. J4, 2015, 42(1): 149-154.
[15]	姜建国;赵媛;孟宏伟;李博. 采用类电磁机制算法的SVM决策树多分类策略[J]. J4, 2014, 41(6): 83-88.