A New Forensic Video Database for Source Smartphone Identification: Description and Analysis

Abstract

In recent years, the field of digital imaging has made significant progress, so that today every smartphone has a built-in video camera that allows you to record high-quality video for free and without restrictions. On the other hand, rapidly growing internet technology has contributed significantly to the widespread use of digital video via web-based multimedia systems and mobile smartphone applications such as YouTube, Facebook, Twitter, WhatsApp, etc. However, as the recording and distribution of digital videos have become affordable nowadays, security issues have become threatening and spread worldwide. One of the security issues is identifying source cameras on videos. There are some new challenges that should be addressed in this area. One of the new challenges is individual source camera identification (ISCI), which focuses on identifying each device regardless of its model. The first step towards solving the problems is a popular video database recorded by modern smartphone devices, which can also be used for deep learning methods that are growing rapidly in the field of source camera identification. In this paper, a smartphone video database named Qatar University Forensic Video Database (QUFVD) is introduced. The QUFVD includes 6000 videos from 20 modern smartphone representing five brands, each brand has two models, and each model has two identical smartphone devices. This database is suitable for evaluating different techniques such as deep learning methods for video source smartphone identification and verification. To evaluate the QUFVD , a series of experiments to identify source cameras using a deep learning technique are conducted. The results show that improvements are essential for the ISCI scenario on video.

Source Camera Device Identification from Videos

“…We also conduct experiments on the newly available Qatar University Forensic Video Database (QUFVD) data set [1]. The data set consists of 20 camera devices such that there are 5 brands, 2 camera models for each brand and 2 identical devices for each camera model.…”

“…In comparison to VISION data set, QUFVD contains more recent smartphones. Furthermore, Akbari et al [1] explicitly divide the data set into train, validation, and test sets. This allows for fair evaluation of the SCI methods on the QUFVD data set.…”

“…Law Enforcement Agencies (LEAs) have a special interest in the developments in this field as the knowledge of the source camera, extracted from the digital media, can provide additional intelligence in the fight against child sexual abuse content. Our work is part of the EU-funded 4NSEEK project 1 which is aimed at the development of cyber-tools to assist LEAs in identifying the source of illicit content involving minors.…”

“…Moreover, to the best of our knowledge, this is the largest subset of devices from the VISION data set used to conduct experiments for SCI using only videos and trained using a single model. Furthermore, we also set a new benchmark result on the QUFVD data set [1] using all the 20 camera devices. Finally, we demonstrate the runtime efficiency of our algorithm during deployment which becomes necessary to conduct forensic investigations in time critical situations.…”

PRNU-Net: a Deep Learning Approach for Source Camera Model Identification based on Videos Taken with Smartphone

“…Most of the databases used for source video identification provide recordings captured with video cameras, and only two databases offer recordings with smartphones, namely Daxing [1] and QUFVD [30]. Therefore, exploring the methods based on smartphone databases, which are developed with a rapid growth, can show whether the existing methods are useful on the new databases.…”