基于成像传感器特征的相机分类器设计  

Design of camera classifier based on imaging sensor features

作　　者：李俊强[1] 孟雷[1] 

机构地区：[1]许昌学院计算机科学与技术学院,河南许昌461000

出　　处：《传感器与微系统》2011年第8期92-94,共3页Transducer and Microsystem Technologies

基　　金：国家自然科学基金资助项目(10871217)

摘　　要：在对现有成像设备源辨识算法分析研究的基础上,提出一种利用成像传感器特征进行相机源辨识的鲁棒性方法。基于模式分类的原理,首先分析数码相机成像的特点,提取传感器噪声信息的统计特征,设计一种鲁棒的分类器来确定相机的品牌/型号。所提取的图像特征包括图像去噪差值和小波域分析。结果表明:所设计的分类器可以有效地正确辨识相机品牌/型号,且具有良好的鲁棒性。On the basis of analysis,investigation,and modification of existing imaging device identification algorithm,a robust method for source camera identification is proposed based on imaging sensor features.Based on the principle of pattern recognition,the characteristics of imaging process of the camera is analyzed,and a robust classifier is constructed to determine the brand/model of the camera by extracting noise＇s statistical features of sensor.The proposed features include image denoising subtraction and wavelet analysis.Experimental results demonstrate that the design of classifier can effectively identify the correct camera brand/model with good robustness.

关 键 词：数字图像取证 源相机辨识 图像特征 支持向量机 成像传感器 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]