Designing Statistical Model-based Discriminator for Identifying Computer-generated Graphics from Natural Images

Designing Statistical Model-based Discriminator for Identifying Computer-generated Graphics from Natural Images
复制标题

设计基于统计模型的鉴别器,用于从自然图像中识别计算机生成的图形

DOI:
--
复制
发表时间:
2019
影响因子:
1
通讯作者:
Zheng Ning
Zheng Ning
中科院分区:
计算机科学4区
文献类型:
--
作者:
Huang Mingying;Xu Ming;Qiao Tong;Wu Ting;Zheng Ning

文献摘要

参考文献

相似文献

本文的目的是区分自然图像(NI)的数码相机和计算机生成的图形(CG)创建的计算机图形渲染软件。本文的主要贡献有三个方面。首先,我们建议利用两种不同的去噪滤波器来获取被检图像的一阶和二阶噪声,并假设残余噪声遵循所提出的统计模型来分析其特性。其次,在假设检验理论的框架下,将NI和CG的识别问题平滑地转化为已知干扰参数的似然比检验(LRT)的设计问题,并对LRT的性能进行了理论研究。第三,在实际分类中,利用估计的模型参数,我们提出建立广义似然比检验(GLRT)。仿真和真实的数据的大规模实验结果直接验证了我们提出的测试具有识别CG和NI的能力,具有较高的检测性能,并显示出与一些现有技术相当的有效性。此外,考虑到一些后处理技术产生的攻击,所提出的分类器的鲁棒性进行了验证。
The purpose of this paper is to differentiate between natural images (NI) acquired by digital cameras and computer-generated graphics (CG) created by computer graphics rendering software. The main contributions of this paper are threefold. First, we propose to utilize two different denoising filters for acquiring the first-order and second-order noise of the inspected image, and analyze its characteristics with assuming that residual noise follows the proposed statistical model. Second, under the framework of the hypothesis testing theory, the problem of identifying between NI and CG is smoothly transferred to the design of the likelihood ratio test (LRT) with knowing all the nuisance parameters, and meanwhile the performance of the LRT is theoretically investigated. Third, in the practical classification, using the estimated model parameters, we propose to establish a generalized likelihood ratio test (GLRT). A large scale of experimental results on simulated and real data directly verify that our proposed test has the ability of identifying CG from NI with high detection performance, and show the comparable effectiveness with some prior arts. Besides, the robustness of the proposed classifier is verified with considering the attacks generated by some post-processing techniques.
基于多重分形和回归分析的自然图像和计算机生成图形的区分
DOI: 10.1016/j.aeue.2016.11.009
发表时间: 2017-01-01
影响因子: 3.2
作者:
Peng, Fei;Zhou, Die-lan;Sun, Xing-ming
通讯作者: Sun, Xing-ming
DOI: 10.1007/978-1-4757-1923-9
发表时间: 1959
影响因子: 7.3
作者:
E. L. Lehmann
通讯作者: E. L. Lehmann
DOI: 10.1145/2482513.2482514
发表时间: 2013-06
期刊: --
影响因子: --
作者:
Vojtech Holub;J. Fridrich
通讯作者: Vojtech Holub;J. Fridrich
DOI: 10.1109/icip.2014.7026116
发表时间: 2014-10
期刊: 2014 IEEE International Conference on Image Processing (ICIP)
影响因子: --
作者:
T. Qiao;Cathel Zitzmann;F. Retraint;R. Cogranne
通讯作者: T. Qiao;Cathel Zitzmann;F. Retraint;R. Cogranne
DOI: 10.1016/j.diin.2013.04.007
发表时间: 2013-10
期刊: Digit. Investig.
影响因子: --
作者:
G. Birajdar;V. Mankar
通讯作者: G. Birajdar;V. Mankar