Technologies Of Digital Video Passive Forensics On The Objects

In recent years,with the rapid development of multimedia technology represented by digital image and digital video and the high quality digital camera equipment widely used,a variety of digital multimedia resources have brought some conveniences and enjoyment to our daily life and work.At the same time,a large number of multimedia video editing software is coming up.However,someone with criminal motivation modify arbitrarily video and spread fake videos by using the digital video own characteristics of being tampered,copyed and diffused extremely easy,which gives rise to many information security problems and brought serious challenges to information security.Therefore,how to quickly and effectively determine the authenticity and integrity of the video resources has been one of the research hotspots in the field of information security at present.Under such background,the technology of digital video passive forensics is carried on,which aims to detect traces of tampering without using prior information.In this paper,firstly,the research status of the digital video passive forensics technology is summarized,and then the key issues of video content authenticity detection are studied and explored.In content tampering,object tampering is the most confusing.Object level tampering usually make the foreground object as the target to tamper with the digital video,which can be used to distinguish the whole frame tampering.In the paper,we research on the key technology of digital video forensics.The main research contents are as follows.1.Aim at the condition that a video may be tampered by removing the video object in a static background,we presented a kind of video tamper detection algorithm based on high frequency component difference degree.In order to achieve the goal of visual coherence,the repair operation was usually introduced,which reduced the difference of the high frequency component in the tampered area.Algorithm was not operated on the basis of pixel,but transferred to the frequency domain,it's perfect to resist the quantization noise impact brought by the secondary compression.The leakage alarm rate,false alarm rate based on a subset reached 7.20% and 3.07%respectively.2.On the basis of the above methods,a spatial and temporal detection algorithm for detecting object deletion is proposed to mine residual information from the aspects of temporal domain and spatial domain respectively.At the first time,we obtained a good observation view by making use of the video profile to locate the video tampering region,when extracting residual information from temporal domain.Moreover,the algorithm can transforms the 3D video into 2D image effectively.The experiments show that the proposed approach achieves more excellent results on detection compared to study 1.3.By using the characteristics of the optical flow which is easy to be influenced by the shootin g environment,we compared the optical flow from different foreground objects in the video to dete rmine whether the foreground objects came from the same shooting environment.Heterogeneous v ideo synthesis tampering is the operation of putting the foreground objects from different videos int o the same video background.In addition to the influence of the speed of the object,the optical flo w would also be affected by the light intensity under different shooting conditions.So the backgrou nd environment can be characterised by the optical flow value after removing the influence of the s peed factor.