Detection of Image Splicing Manipulation by Automated Classification of Color Temperature Distance
-
摘要: 拼接篡改是一类常见的图像伪造手段，现有取证方法难以实现图像中拼接篡改区域的自动检测与精确定位，导致拼接篡改伪造图像的取证长期依赖人工经验.基于图像中原始区域与拼接篡改区域所反映的光源色温的差异性，提出一种自动色温距离阈值分类的图像拼接篡改检测与定位方法.首先，变换待检验图像至YCbCr色彩空间，并按照Grid-based方式结构化分解为大小的子图像块；然后，利用自动白平衡（Automatic white balance，AWB）中的白点检测原理对每一个子图像块进行色温估计，计算子图像块与参考区域之间的色温距离；最后，采用最大类间方差法自适应地求取色温距离分类的最佳阈值，对子图像块进行分类标注，实现了图像拼接篡改区域的自动检测与精确定位.实验表明，该方法能够实现图像拼接篡改区域的自动检测与定位，具有较高的量化检测精度.Abstract: Splicing is a common types of tampering in image manipulation. As many authentication methods cannot detect and localize the manipulated area automatically in splicing images, authentication of splicing image has depended on human experience for a long time. In this paper, considering the inconsistency of color temperature between original area and splicing area, we propose an automated distance threshold classification method for splicing image detection and manipulation localization by color temperature estimation. At first, we transform suspicious image into YCbCr color space and divide it into blocks with grid-based manner. Then, we estimate color temperature of each block using automatic white balance (AWB) theory, and calculate Euclidean distance between reference area and suspicious area. Finally, we localize the splicing area with an automated estimated optimal threshold of color temperature distance. Experiments indicate that our method can detect splicing images and localize splicing area effectively and automatically with a quantitative result.1) 本文责任编委 桑农
-
图 3 哥伦比亚数据库的检测效果(a-1) $\sim$ (a-8)为哥伦比亚数据库中的拼接篡改图像; (b-1) $\sim$ (b-8)为本文方法的检测结果
Fig. 3 Detection on DVMM (a-1) $\sim$ (a-8) Splicing images in experiments; (b-1) $\sim$ (b-8) Detection results with proposed method
图 4 拼接篡改伪造图像检测的主观视觉评价((a-2) $\sim$ (e-2)本文方法检测结果; (a-3) $\sim$ (e-3)拼接篡改图像MASK; (a-4) $\sim$ (e-4)文献[18]方法的检测结果; (a-5) $\sim$ (e-5)文献[18]光照方向检测2-D模型)
Fig. 4 Visual evaluation of detection on splicing images ((a-2) $\sim$ (e-2) Detection results with proposed method; (a-3) $\sim$ (e-3) MASK of splicing images; (a-4) $\sim$ (e-4) Detection results with [18]; (a-5) $\sim$ (e-5) 2-D model of illumination direction in [18])
图 5 不同类型拼接篡改伪造图像的识别准确率、召回率与综合评测因子随$\varphi$值的变化曲线
Fig. 5 Curves of $R$, $P$ and $F_1$ on different splicing images with variation of $\varphi$
表 1 方法流程图中出现的变量及其描述
Table 1 Description of variables in framework
变量名 含义描述 $f(x, y)_{\rm RGB} $ RGB色彩空间的待检验图像 $f(x, y)_{\rm YCbCr}$ YCbCr色彩空间的待检验图像 ${\rm Block}_{ij}(x, y)$ YCbCr色彩空间的子图像块 $C_{ij} $ 每一个子图像块所对应的色温估计值 ${\rm Area}_R $ 参考区域, 由子图像块构成的假设无篡改区域 ${\rm Area}_S $ 嫌疑区域, 可能包含篡改区域的子图像块集合 $D_{ij} $ 嫌疑区域与参考区域之间的色温距离 $T$ 自动估计的色温距离阈值 $R_{\rm MAP} $ 比较色温距离与色温距离阈值后确定的篡改区域 
下载: 导出CSV
表 2 实验参数设置
Table 2 Configuration of experiments
$m$ $w_{ij}$ $\varphi$ $Th$ (a-1) 5 0.4, 0.3, 0.2, 0.15，0.05 43 0.34118 (b-1) 58 0.35686 (c-1) 44 0.32157 (d-1) 81 0.21569 (e-1) 54 0.23725
下载: 导出CSV
表 3 $\varphi$遍历寻优后的量化实验结果
Table 3 Detection results with ergodic optimization of $\varphi$
参数 自动+ 日光 阴天 阴影 荧光灯 钨丝灯 $F_1$ 0.566 6 0.484 4 0.620 5 0.631 9 0.823 7 $R$ 0.886 5 0.967 0 0.937 7 0.478 8 0.728 0 $P$ 0.416 4 0.323 2 0.463 7 0.929 2 0.948 4 $\varphi_{\rm best}$ 43 58 44 81 54
下载: 导出CSV
-
[1] Farid H. Creating and Detecting Doctored and Virtual Images: Implications to the Child Pornography Prevention Act. Technical Report TR2004-518, Department of Computer Science, Dartmouth College, USA, 2004. [2] 黄太云.刑事诉讼法修改释义.人民检察, 2012, (8):10-73 http://www.bookask.com/book/938711.htmlHuang Tai-Yun. Interpretation of the amendment of the Criminal Procedure Law. People's Procuratorial Semimonthly, 2012, (8):10-73 http://www.bookask.com/book/938711.html [3] Al-Qershi O M, Khoo B E. Passive detection of copy-move forgery in digital images:state-of-the-art. Forensic Science International, 2013, 231(1-3):284-295 doi: 10.1016/j.forsciint.2013.05.027 [4] Cao Y J, Gao T G, Fan L, Yang Q T. A robust detection algorithm for copy-move forgery in digital images. Forensic Science International, 2012, 214(1-3):33-43 doi: 10.1016/j.forsciint.2011.07.015 [5] Amerini I, Ballan L, Caldelli R, Del Bimbo A, Serra G. A SIFT-based forensic method for copy-move attack detection and transformation recovery. IEEE Transactions on Information Forensics and Security, 2011, 6(3):1099-1110 doi: 10.1109/TIFS.2011.2129512 [6] Bay H, Ess A, Tuytelaars T, Van Gool L. Speeded-up robust features (SURF). Computer Vision and Image Understanding, 2008, 110(3):346-359 doi: 10.1016/j.cviu.2007.09.014 [7] Fridrich J, Soukal D, Lukáš J. Detection of copy-move forgery in digital images. International Journal, 2003, 3(2):652-663 http://www.mendeley.com/catalog/detection-copymove-forgery-digital-images/ [8] Wang W, Dong J, Tan T N. Exploring DCT coefficient quantization effects for local tampering detection. IEEE Transactions on Information Forensics and Security, 2014, 9(10):1653-1666 doi: 10.1109/TIFS.2014.2345479 [9] Popescu A C, Farid H. Exposing digital forgeries by detecting duplicated image regions. Technical Report TR2004-515, Dartmouth College, USA, 2004 [10] Khan S, Kulkarni A. An efficient method for detection of copy-move forgery using discrete wavelet transform. International Journal on Computer Science and Engineering, 2010, 2(5):1801-1806 [11] 吕颖达, 申铉京, 陈海鹏.具有几何不变性的图像复制——粘贴盲鉴别算法.电子学报, 2016, 44(11):2592-2599 doi: 10.3969/j.issn.0372-2112.2016.11.005Lv Ying-Da, Shen Xuan-Jing, Chen Hai-Peng. Blind forensic for image copy-paste tampering with geometric invariance. Acta Electronica Sinica, 2016, 44(11):2592-2599 doi: 10.3969/j.issn.0372-2112.2016.11.005 [12] 朱叶, 申铉京, 陈海鹏.基于彩色LBP的隐蔽性复制-粘贴篡改盲鉴别算法.自动化学报, 2017, 43(3):390-397 http://www.aas.net.cn/CN/abstract/abstract19017.shtmlZhu Ye, Shen Xuan-Jing, Chen Hai-Peng. Covert copy-move forgery detection based on color LBP. Acta Automatica Sinica, 2017, 43(3):390-397 http://www.aas.net.cn/CN/abstract/abstract19017.shtml [13] Johnson M K, Farid H. Exposing digital forgeries by detecting inconsistencies in lighting. In: Proceedings of the 7th Workshop on Multimedia and Security, MM&Sec 2005. New York, NY, USA: ACM Press, 2005. 1-10 [14] Chen M, Fridrich J, Goljan M, Lukas J. Determining image origin and integrity using sensor noise. IEEE Transactions on Information Forensics and Security, 2008, 3(1):74-90 doi: 10.1109/TIFS.2007.916285 [15] Popescu A C, Farid H. Exposing digital forgeries in color filter array interpolated images. IEEE Transactions on Signal Processing, 2005, 53(10):3948-3959 doi: 10.1109/TSP.2005.855406 [16] Johnson M K, Farid H. Exposing digital forgeries through chromatic aberration. In: Proceedings of the 8th Workshop on Multimedia and Security, Mm&Sec 2006. Geneva, Switzerland: ACM Press, 2006. 48-55 [17] Farid H. Image forgery detection. IEEE Signal Processing Magazine, 2009, 26(2):16-25 doi: 10.1109/MSP.2008.931079 [18] Johnson M K, Farid H. Exposing digital forgeries in complex lighting environments. IEEE Transactions on Information Forensics and Security, 2007, 2(3):450-461 doi: 10.1109/TIFS.2007.903848 [19] Kee E, O'Brien J F, Farid H. Exposing photo manipulation from shading and shadows. ACM Transactions on Graphics, 2014, 33(5):Article No. 165 doi: 10.1145/2629646 [20] 牛少彰, 黄艳丽, 孙晓婷.投影与光照方向一致性的图像篡改检测.北京邮电大学学报, 2014, 37(5):61-65 http://www.cqvip.com/QK/91520A/201405/663517069.htmlNiu Shao-Zhang, Huang Yan-Li, Sun Xiao-Ting. Image tampering detection by consistency of projection and lighting direction. Journal of Beijing University of Posts and Telecommunications, 2014, 37(5):61-65 http://www.cqvip.com/QK/91520A/201405/663517069.html [21] 陈海鹏, 申铉京, 吕颖达, 金玉善.基于Lambert光照模型的图像真伪盲鉴别算法.计算机研究与发展, 2011, 48(7):1237-1245 http://mall.cnki.net/magazine/Article/JFYZ201107017.htmChen Hai-Peng, Shen Xuan-Jing, Lv Ying-Da, Jin Yu-Shan. Blind identification for image authenticity based on Lambert illumination model. Journal of Computer Research and Development, 2011, 48(7):1237-1245 http://mall.cnki.net/magazine/Article/JFYZ201107017.htm [22] 孙鹏, 杨洪臣, 代雪晶, 沈喆.拼接篡改图像的色温估计取证方法.计算机辅助设计与图形学学报, 2012, 24(9):1193-1196 http://mall.cnki.net/magazine/Article/JSJF201209012.htmSun Peng, Yang Hong-Chen, Dai Xue-Jing, Shen Zhe. An authentification method for splicing images with color temperature estimation. Journal of Computer-Aided Design & Computer Graphics, 2012, 24(9):1193-1196 http://mall.cnki.net/magazine/Article/JSJF201209012.htm [23] de Carvalho T J, Riess C, Angelopoulou E, Pedrini H, de Rezende Rocha A. Exposing digital image forgeries by illumination color classification. IEEE Transactions on Information Forensics and Security, 2013, 8(7):1182-1194 doi: 10.1109/TIFS.2013.2265677 [24] Gijsenij A, Lu R, Gevers T. Color constancy for multiple light sources. IEEE Transactions on Image Processing, 2012, 21(2):697-707 doi: 10.1109/TIP.2011.2165219 [25] Nakano N, Nishimura R, Sai H, Nishizawa A, Komatsu H. Digital still camera system for mega pixel CCD. IEEE Transactions on Consumer Electronics, 1998, 44(3):581-586 doi: 10.1109/30.713166 [26] Lee J S, Jung Y Y, Kim B S, Ko S J. An advanced video camera system with robust AF, AE, and AWB control. IEEE Transactions on Consumer Electronics, 2001, 47(3):694-699 doi: 10.1109/30.964165 [27] 周荣政, 何捷, 洪志良.自适应的数码相机自动白平衡算法.计算机辅助设计与图形学学报, 2005, 17(3):529-533 http://www.docin.com/p-426388884.htmlZhou Rong-Zheng, He Jie, Hong Zhi-Liang. Adaptive algorithm of auto white balance for digital camera. Journal of Computer-Aided Design & Computer Graphics, 2005, 17(3):529-533 http://www.docin.com/p-426388884.html [28] Christlein V, Riess C, Jordan J, Riess C, Angelopoulou E. An evaluation of popular copy-move forgery detection approaches. IEEE Transactions on Information Forensics and Security, 2012, 7(6):1841-1854 doi: 10.1109/TIFS.2012.2218597 [29] Ng T T, Hsu J, Chang S F. Columbia image splicing detection evaluation dataset[Online]. availabe: http://www.ee.columbia.edu/ln/dvmm/downloads/AuthSplicedDataSet/AuthSplicedDataSet.htm, July 12, 2018 -
图(5) / 表(4)
计量
- 文章访问数: 3220
- HTML全文浏览量: 387
- PDF下载量: 938
- 被引次数: 0
