计算机科学 ›› 2014, Vol. 41 ›› Issue (Z11): 136-138.

• 模式识别与图像处理 • 上一篇    下一篇

基于背景噪声的图像盲篡改检测

刘丽娟,林小竹   

  1. 北京石油化工学院信息工程学院 北京102617;北京化工大学信息科学与技术学院 北京100026;北京石油化工学院信息工程学院 北京102617
  • 出版日期:2018-11-14 发布日期:2018-11-14
  • 基金资助:
    本文受北京市属高等学校人才强教计划资助

Image Forgery Detection Using Characteristics of Background Noise

LIU Li-juan and LIN Xiao-zhu   

  • Online:2018-11-14 Published:2018-11-14

摘要: 数字图像在成像过程中会产生特定的背景噪声,如果两幅不同噪声的图像拼接在一起,篡改区域和其他区的噪声会有差异。提出一种基于偏度统计特性的背景噪声估计算法,其通过对图像分块计算每块的噪声标准差,从而检测出噪声异常部分以达到篡改检测的目的。算法利用DCT变换去除原图细节部分,利用偏度统计特性估计噪声,利用条件最小值法求出噪声的标准差。算法改进了迭代求条件最小值法,利用微分方法求取最小值,避免了初始值设定问题,提高了算法的准确率。实验结果表明,提出的噪声估计算法正确率高,且对拼接篡改图像篡改检测有明显效果。

关键词: 图像取证,背景噪声,偏度,标准差

Abstract: There is a part of background noise in digital image which comes from imaging process.The noise characteristics between image forgery area and the other area are different if images with different noise levels are spliced together.This paper proposed a background noise estimation algorithm based on the statistical properties of skewness.We detected the forgery parts by dividing image into some sub-blocks and computing the noise variance of each ones.This algorithm removes the original image details with DCT transform,estimates noise with the statistical properties of skewness,and estimates the standard deviation of noise with condition of the minimum method.This algorithm improves iterative conditional minimum value method using differential method to calculate the minimum value.This algorithm avoids the problem of setting the initial value,and improves the accuracy of the algorithm.The experimental results show that the proposed noise estimation algorithm has high accuracy and effectiveness in detecting forgery part in spliced images.

Key words: Image forensics,Background noise,Skewness,The standard deviation

[1] Gou H,Swaminathan A,Wu M.Intrinsic sensor noise featuresfor forensic analysis on scanners and scanned images[J].IEEE Transactions on Information Forensics and Security,2009,4(3):476-491
[2] 李哲,郑江滨.基于噪声分布规律的伪造图像盲检测算法[J].计算机应用研究,2009,6(3):1092-1094
[3] Donoho D L,Johnstone J M.Ideal spatial adaptation by waveletshrinkage[J].Biometrika,1994,81(3):425-455
[4] Mahdian B,Saic S.Using noise inconsistencies for blind imageforensics[J].Image and Vision Computing,2009,27(10):1497-1503
[5] Li T,Wang M,Li T.Estimating noise parameter based on thewavelet coefficients estimation of original image[C]∥2010 International Conference on Challenges in Environmental Science and Computer Engineering (CESCE).IEEE,2010,1:126-129
[6] Luká J,Fridrich J,Goljan M.Detecting digital image forgeries using sensor pattern noise[C]∥International Society for Optics and Photonics Electronic Imaging 2006.2006:60720Y-60720Y-11
[7] Popescu A C,Farid H.Statistical tools for digital forensics[C]∥Information Hiding.Springer Berlin Heidelberg,2005:128-147
[8] 张晖,张荣,尹东.使用背景噪声盲估计的图像真伪鉴别[J].中国图像图形学报,2010,5(12):1738-1741
[9] 卢燕飞,鞠娅莉,于跃.基于图像背景噪声特性的篡改检测[J].信号处理,2012,8(9):1299-1307
[10] Zoran D,Weiss Y.Scale invariance and noise in natural images[C]∥2009 IEEE 12th International Conference on Computer Vision.IEEE,2009:2209-2216
[11] Pan X,Zhang X,Lyu S.Exposing image forgery with blind noise estimation[C]∥Proceedings of the Thirteenth ACM Multimedia Workshop on Multimedia and Security.ACM,2011:15-20
[12] Pan X,Zhang X,Lyu S.Exposing image splicing with inconsistent local noise variances[C]∥2012 IEEE International Conference on Computational Photography (ICCP).IEEE,2012:1-10
[13] Franzen R.Kodak lossless true color image suite.1999http://r0k.us/graphics/kodak
[14] Schaefer G,Stich M.UCID-An Uncompressed Colour Image Database[C]∥Storage and Retrieval Methods and Applications for Multimedia.SPIE,2004:472-480
[15] Ng T T,Chang S F.Columbia Image Splicing Detection Evaluation Dataset[EB/OL].http://www.ee.columbia.edu/ln/dvmm/downloads/AuthSplicedDataSet/photographers.htm

No related articles found!
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
No Suggested Reading articles found!