Photorealistic Computer Graphics Identification Based on Features from Multi-domains
As the advent and growing popularity of image rendering software, photorealistic computer graphics are becoming more and more visually indistinguishable from photographic images. If the computer graphics are abused as faked images, it may lead to potential social, legal or private consequences. To this end, it”s very essential and also challenging to find effective methods to differentiate between them. In this paper, two kinds of features are extracted from image gradient domain and wavelet domain, and a feature fusion approach is applied construct the detection method. The proposed method can efficiently distinguish between photographic images and computer generated ones. Results of experiments indicate that the validity of the method. Meanwhile, the detection accuracy remain high after feature reduction.
digital image forensics computer graphics identification image gradient leading digit law
Bo XU Guangjie LIU Yuewei DAI
School of Automation, Nanjing University of Science & Technology, Nanjing 210094, China
国内会议
北京
英文
148-153
2012-04-01（万方平台首次上网日期，不代表论文的发表时间）