摘要： 将模式噪声作为固定指纹特征应用于手机相机来源检测时，存在计算复杂、效率不高等问题。为此，提出一种基于模式噪声大分量信息的手机相机来源检测方法。利用光响应敏感点的成像特性及对模式噪声的影响，将其从模式噪声中分离出来，构成模式噪声的大分量信息，并作为新的模式噪声进行手机相机来源检测。实验结果表明，与传统基于模式噪声的图像来源检测方法相比，该方法不仅能有效辨识手机相机图像的来源，而且能减少检测计算量。

关键词: 手机相机, 来源检测, 被动取证, 模式噪声, 相机指纹, 成像传感器

Abstract: Due to the problems such as complex calculations and poor efficiency of the sensor pattern noise as the inherent fingerprint of mobile phone cameras when it is used in digital images origin forensics, a new method which based on large components of sensor pattern noise for mobile phone camera is proposed. According to the optical response characteristics and special influence of the light sensitive points, it makes up large components of imaging sensor pattern noise mainly for mobile camera origin detection. Experimental results show that, compared with the conventional image source detection method based on the pattern noise, the proposed method can not only identify the source of mobile phone camera images effectively, but also decrease the computational load for detection.

Key words: mobile phone camera, origin detection, passive authentication, pattern noise, camera fingerprint, imaging sensor

中图分类号: 

• TP391