摘要
真实图像和计算机生成图像的鉴别是数字图像盲取证中的一个重要问题。提出了一种真实图像和计算机生成图像的鉴别方法。借用模式识别中的二分类概念,采用图像质量评价量来建立模型,以捕获真实图像和计算机生成图像的统计差异,选用支持向量机作为分类器进行训练和测试。实验结果表明,该方法计算方便、实用性强、可靠性高。
Discrimination between real images and computer generated images is an important issue in digital image forensics. This paper proposed a new detection scheme to discriminate computer generated images from real images. The detection could be regarded as a two-class pattern recognition problem, and the model was established using image quality metrics （IQM） extracted from the given test image. This model could capture statistical differences between real images and computer generated images. Kernel-based Support Vector Machine （SVM） was chosen as a classifier to train and test the given images. Experimental results demonstrated that this new detection scheme had some advantages of easy calculation, wide applications and high reliability.
出处
《测绘科学技术学报》
北大核心
2008年第5期355-358,共4页
Journal of Geomatics Science and Technology
基金
国家自然科学基金(60473022)
关键词
真实图像
计算机生成图像
图像质量评价量
支持向量机
real images
computer generated images
image quality metrics
support vector machine （SVM）
