Design Of Digital Image Forensic System Based On The Chromatic Aberration

Under the need of trial of computer-related crimes, a number of computer forensic tools appear. There are also some theoretical results. However, comparing with the need of combating computer-related crimes, we still have a lot of works to do in the practice. At present, there are many digital photos in which watermarks haven't been embedded. If they are been invaded, it is difficult to obtain evidences. In fact, the shortcomings of optical system decide that a digital camera imaging itself is not perfect. This is because of chromatic aberration. That is to say, there is local relevance between the pixels of a digital photo. When tampering with a digital photo, it is bound to change the original contacts between the pixels. So the digital forgeries can be found by detecting the changes of the inherent attributes of the image. It is on this basis, this paper designs a digital image blind forensics system.Firstly, under the urgent need of fighting against computer crime, this paper designs a digital image blind forensics software tool for the detection intrusion in digital photos. Although academics have achieved some theory results in this field, the actual situation is there is lack of software tool for the detection tampering in digital images, in particularly blind forensic tool is blank.Secondly, from the application point of view, several blind forensic algorithms are analyzed and compared carefully. They are being adopted and authenticated each other.Thirdly, some algorithms of digital image forensic have been adapted. The complexity of the algorithm has been reduced. Meanwhile, the accuracy rate also has increased slightly.Fourthly, with the need of fighting against the crime of digital images forgeries, we provide a framework for the digital image blind forensic system. We talk about the probability with the concrete algorithm models. For the development of such software tools, a solid foundation has been laid.Fifthly, we summary and analysis the types of the digital image forged operations. To the features of various operations, corresponding algorithm models have been adopted. Because image forgery often involves several operations, such as copy, paste, resample and fuzzy, the system uses various types of models for the forensic analysis. So the accuracies of analysis have been greatly improved.In short, starting from the needs of the market, this paper put forward a digital image blind forensic system. For the promotion of the development of the digital image forensic software tool, it may play a important role. We expect it can give some help to the difficult digital image forensic works.