摘要
针对基于模式噪声的数码相机源检测中存在的问题,提出了一种改进的模式噪声提取算法。在Lukas算法的基础上,首先利用Wallis预滤波处理抑制低频噪声并增强模式噪声;再利用Sobel边缘检测算子对图像的复杂边缘纹理区进行有效剔除,最后利用经过上述处理得到的图像模式噪声与相机参考模式噪声模板进行相关检测。实验结果表明,提出的改进算法可以有效提高源相机检测的性能。
Aiming at the problems existing in source digital camera detection based on sensor pattern noise,an improved al-gorithm for pattern noise extraction is presented. Based on Lukas’s algorithm,the Wallis pre-filtering is employed to suppressthe low frequency noise and enhance the sensor pattern noise,and then the Sobel operator based edge detection is used to re-move the complex edge and texture regions of an image. Finally,camera correlation-based detection is performed by calculatingthe correlation between the image pattern noise obtained by the above process and the camera reference pattern noise. The ex-perimental results show that the proposed algorithm can effectively improve the detection accuracy of the source camera.
出处
《现代电子技术》
2014年第6期27-31,共5页
Modern Electronics Technique
基金
国家高技术研究发展计划("863"计划)资助项目(2006AA09Z237)
关键词
数字图像取证
源相机检测
模式噪声
Wallis预滤波
相关性检测
digital image forensics
source camera detection
pattern noise
Wallis pre-filtering
correlation-based detec-tion