为解决现有篡改定位网络随着深度加深不易收敛的问题，提出一种基于残差全卷积网络的图像拼接定位算法.所提算法一方面迁移残差思想，在全卷积神经网络（fullyconvolutional network，FCN）的部分卷积层中引入shortcut连接，使其输出的不仅是输入的映射，还是输入映射与输入的叠加.另一方面结合条件随机场（conditional random field，CRF）对定位结果进行后处理，并将FCN与CRF整合在一个端到端的学习系统中，进一步提高定位精度.此外，所提算法还融合3种FCN（FCN8、FCN16、FCN32）的预测结果.在实验中，随机选取公开数据集CASIA v2.0的5/6篡改图像作为训练集，然后对剩余1/6进行测试.为了测试提出算法的泛化性能，采用训练好的模型在公开数据集CASIA v1.0和DVMM上进行交叉测试.在3个数据集上的测试结果表明，所提算法的性能优于现有一些方法.
In order to solve the problem that the existing forgery localization networks are not easy to converge with the increase of the depth of networks, an image splicing localization algorithm based on fully convolution residual networks is proposed in this paper. On the one hand, the proposed algorithm transfers the idea of residual structure and introduces shortcut connection into part of convolution layers in fully convolutional network (FCN), so that the output is not a mapping input alone but the superposition of a mapping input and an input itself. On the other hand, conditional random field (CRF) is used as post-processing operation to improve splicing localization accuracy. Moreover, FCN and CRF are integrated in an end-to-end learning system. In addition, the proposed algorithm combines the prediction results of three kinds of FCNs (FCN8, FCN16 and FCN32). In our experiment, 5/6 of the spliced images in the public available dataset CASIA v2.0 are randomly selected as training set and the rest are used for testing. In order to test generalization performance of the proposed algorithm, the trained model is also cross-tested on another two public available datasets CASIA v1.0 and DVMM. The overall test results on three datasets show that the proposed algorithm performs better than some existing algorithms.
[1] 陈园园,于在河. 数字图像被动取证技术综述[J]. 吉林大学学报(信息科学版),2014, 32(6):690-698. Chen Y Y, Yu Z H. Survey on passive forensics techniques for digital images[J]. Journal of Jilin University (Information Science Edition), 2014, 34(6):690-698. (in Chinese)
[2] 赵洁,王光夫,赵建楠,等. 面向拼接与克隆篡改的数字图像内容取证系统设计[J]. 中央民族大学学报(自然科学版), 2018, 27(3):61-64. Zhao J, Wang G F, Zhao J N, et al. Design of digital image content forensics system for mosaic and clone tampering[J]. Journal of Minzu University of China (Natural Sciences Edition), 2018, 27(3):61-64. (in Chinese)
[3] Liu Z C, He J F, Tang X O, et al. Fast, automatic and fine-grained tampered JPEG image detection via DCT coefficient analysis[J]. Pattern Recognition, 2009, 42(11):2492-2501.
[4] Bianchi T, Piva A. Image forgery localization via block-grained analysis of JPEG artifacts[J]. IEEE Transactions on Information Forensics and Security, 2012, 7(3):1003-1017.
[5] Ye S Y S, Sun Q S Q, Chang E C. Detecting digital image forgeries by measuring inconsistencies of blocking artifact[C]//IEEE International Conference on Multimedia and Expo. IEEE, 2007:487-492.
[6] Li W, Yuan Y, Yu N. Passive detection of doctored JPEG image via block artifact grid extraction[J]. Signal Processing, 2009, 89(9):1821-1829.
[7] Ferrara P, Bianchi T, De Rosa A, et al. Image forgery localization via fine-grained analysis of CFA artifacts[J]. IEEE Transactions on Information Forensics and Security, 2012, 7(5):1566-1577.
[8] Dirik A E, Memon N D. Image tamper detection based on demosaicing artifacts[C]//IEEE International Conference on Image Processing. IEEE, 2009:354-358.
[9] Hsu Y F, Chang S F. Image splicing detection using camera response function consistency and automatic segmentation[C]//IEEE International Conference on Multimedia and Expo. IEEE, 2007:28-31.
[10] Stamm M C, Liu K J. Forensic detection of image manipulation using statistical intrinsic fingerprints[J]. IEEE Transactions on Information Forensics and Security, 2010, 5(3):492-506.
[11] Ng T T, Chang S F, Sun Q. Blind detection of photomontage using higher order statistics[C]//Proceedings of the 2004 International Symposium on Circuits and Systems. IEEE, 2004:688-691.
[12] Johnson M K. Exposing digital forgeries in complex lighting environments[J]. IEEE Transactions on Information Forensics and Security, 2007, 2(3):450-461.
[13] Mahdian B, Saic S. Using noise inconsistencies for blind image forensics[J]. Image and Vision Computing, 2009, 27(10):1497-1503.
[14] Lü S, Pan X, Zhang X. Exposing region splicing forgeries with blind local noise estimation[J]. International Journal of Computer Vision, 2014, 110(2):202-221.
[15] Rao Y, Ni J Q. A deep learning approach to detection of splicing and copy-move forgeries in images[C]//IEEE International Workshop on Information Forensics and Security. IEEE, 2016:254-258.
[16] Wang Q, Zhang R. Double JPEG compression forensics based on a convolutional neural network[J]. EURASIP Journal on Information Security, 2016, 2016(1):23-25.
[17] Amerini I, Uricchio T, Ballan L. Localization of JPEG double compression through multidomain convolutional neural networks[J]. IEEE CVPR Workshop Media Forensics, 2017, 3(10):1865-1871.
[18] Liu B, Pun C M. Locating splicing forgery by fully convolutional networks and conditional random field[J]. Signal Processing Image Communication, 2018, 66(3):103-112.
[19] Chen B J, Qi X M, Wang Y T. An improved splicing localization method by fully convolutional networks[J]. IEEE Access, 2018, 6:69472-69480.
[20] He K, Zhang X Y, Ren S Q, et al. Deep residual learning for image recognition[C]//IEEE Conference on Computer Vision and Pattern Recognition. IEEE, 2016:603-607.
