Video forensic tools exploiting features from video-container to video-encoder level
Year: 2018
Type: dissertation
Abstract: The escalation of multimedia contents exchange, especially of videos belonging to mobile \ndevices, and the availability of a great amount of editing software has raised grave \ndoubts on their digital life-cycle. In this thesis, we firstly introduce a new dataset \nfor multimedia forensics and then develop forensic tools that analyse the \nvideo-container and the video-signal in order to evaluate possible tampering that have \nbeen introduced in the life-cycle of a video content. \n \nThe first contribution consists on the release of a new \nDataset of videos and images captured from to 35 modern smartphones/tablets \nbelonging to 11 different brands: Apple, Asus, Huawei, Lenovo, LG electronics, \nMicrosoft, Oneplus, Samsung, Sony, Wiko and Xiaomi. \nOverall, we collected 11732 native images; 7565 of them were shared \nthrough Facebook, in both high and low quality, and through WhatsApp, resulting in a \ntotal of 34427 images. Furthermore we acquired 648 native videos, \n622 of which were shared through YouTube at the maximum \navailable resolution, and 644 through WhatsApp, resulting in a total of 1914 videos. \nThe uniqueness of the VISION dataset was tested on well known forensic tool, i.e., \nthe detection of the Sensor Pattern Noise (SPN) left by the acquisition \ndevice for the source identification of \nnative/social media contents. \n \nThe second contribution is based on the analysis of the container structure of videos acquired by means of mobile devices. We argue that the atoms belonging to the container, in terms of order and value, are fragile and that it is more difficult to hide their modifications than the regular metadata. This characteristic can be exploited to perform Source Identification and Integrity Verification of videos taken from devices belonging to well known operating systems and manufactures. \n \nIn the third contribution we focus on the video-signal and on its encoding \nprocess. We used codecs that perform a hybrid video coding scheme, \nand developed a classification technique able to perform group of picture length estimation \nand double compression detection. The proposed technique is one of the fastest \napproaches that use videos encoded with B-frames, with both constant bit rate and \nvariable bit rate.
Author: Dasara Shullani
Language: English
Subfield: Computer Vision and Pattern Recognition
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W2802380096
