基于量化表估计的JPEG篡改图像盲检测  

Blind Detection Method of Doctored JPEG Image Based on Estimating Quantization Table

作　　者：赵彦涛[1] 郝丽娟[1] 闫彦霖[1] 

机构地区：[1]燕山大学电气工程学院,河北秦皇岛066004

出　　处：《计算机仿真》2012年第10期23-25,356,共4页Computer Simulation

基　　金：国家自然科学基金(61107039);燕山大学博士基金(B505);秦皇岛市科学技术研究与发展计划(201001A078)

摘　　要：传统JPEG篡改图像盲检测算法采用不同质量因子对待测图像进行尝试性再压缩。为了缩短检测时间,减少计算量,提出一种基于量化表估计的JPEG篡改图像盲检测有效算法。首先通过DCT系数直方图估计出篡改图像某一部分量化表,然后通过计算小块DCT系数中分别是对应量化步长整数倍的个数进一步判定该小块是否属于篡改区域。实验结果表明,提出的量化步长估计算法复杂度小,正确率高;检测算法能准确检测出copy-move和合成类图像篡改。此外,改进方法不仅对标准量化表压缩的JPEG篡改图像具有非常好的检测结果,而且对非标准量化表压缩的JPEG篡改图像检测也有效。In order to shorten the testing time and reduce computation cost,an efficient algorithm was proposed for localizing the doctored JPEG image by estimating the quantization table in this paper.The quantization table of the spliced portion was firstly estimated by the histogram of the DCT coefficients.Then the numbers of the DCT coefficients were added up respectively,which is an integer multiple of the quantization step,and the doctored area was determined by the numbers.The experimental results show that the quantization table estimation algorithm proposed is less time consuming and of higher accuracy rate.The detection algorithm can expose various kinds of JPEG image forgeries including copy-move and composite.Furthermore,the proposed algorithm successfully detects the tempered image with JPEG compression using not only the standard quantization table,but also the Non-standard quantization tables

关 键 词：盲检测 压缩 系数 量化表 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]