A Research On Image Forgery Forensics And The Evaluation Of Its Influences On The Original Image

Images,unlike text,represent an effective and natural communication media for humans,due to their immediacy and the easy way to understand the image content.At the same time,the large availability of image editing software tools makes it extremely simple to alter the content of the images,or to create new ones.As a result,doctored images are appearing with a growing frequency in different application f ields,and thus today's digital technology has begun to erode the trust on visual content,so that apparently seeing is no longer believing.All these issues will get worse as processing tools become more and more sophisticated.Image forensics has been focusing on low-level visual features,paying little attention to high-level semantic information of the image.We propose the framework for image forgery detection based on high-level semantics with three components of image understanding module,the normal rule bank(NR)holding semantic rules that comply with our common sense,and the abnormal rule bank(AR)holding semantic rules that don't.Firstly,image understanding module is integrated by a dense image caption model,with no need for human intervention and more hierarchical features.Secondly,our proposed framework can generate thousands of semantic rules automatically for NR.Thirdly,besides NR,we also propose to construct AR.In this way,not only can we frame image forgery detection as anomaly detection with NR,but also as recognition problem with AR.The experimental results demonstrate our framework is effective and performs better.Traditional image forensic algorithm can at best tell whether the probe image is authentic or fake,and there is no doubt that some visual content has been maliciously tampered to achieve illegal purpose,while some modifications are benign,just for fun,for enhancing artistic value,or effectiveness of news dissemination.So beyond the tampering detection,how to evaluate the influence of image tampering is on schedule.In this work,we study the problem of automatically assessing the influence of image tampering by examining whether the modification affects the dominant visual content,and utilize saliency mechanism to assess how harmful the tampering is.The experimental results demonstrate the effectiveness of our method.In summary,this article provides the latest development of image tampering detection.Furthermore,it further studies the impact of image tampering on the image itself.