计算机工程与应用 ›› 2012, Vol. 48 ›› Issue (18): 193-196.

• 图形、图像、模式识别 • 上一篇    下一篇

基于成像传感器特征的相机源识别

蒋英春   

  1. 桂林电子科技大学 数学与计算科学学院,广西 桂林 541004
  • 出版日期:2012-06-21 发布日期:2012-06-20

Source camera identification based on characteristics of imaging sensor

JIANG Yingchun   

  1. School of Mathematics and Computational Science, Guilin University of Electronic Technology, Guilin, Guangxi 541004, China
  • Online:2012-06-21 Published:2012-06-20

摘要: 基于模式分类的方法,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。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,设计一个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、小波域子带统计矩以及通道颜色特征。实验结果表明所设计的算法不仅可以有效地正确辨识相机品牌/型号,还具有良好的鲁棒性。

关键词: 相机源辨识, 图像特征, 支持向量机, 成像传感器

Abstract: Based on the principle of pattern recognition, a robust method for source camera identification is proposed by using imaging sensor. The characteristics of imaging process of the camera are analysed, and a robust classifier is constructed to determine the brand/model of the camera by extracting noise statistical features of imaging sensor. The proposed features include image denoising difference,statistical moment of wavelet sub-bands and color feature. Experimental results demonstrate that the design of classifier can effectively identify the correct camera brand/model with good robustness.

Key words: source camera identification, image features, support vector machine, imaging sensor