Video-Based Evidence Analysis and Extraction in Digital Forensic Investigation

Abstract

As a result of the popularity of smart mobile devices and the low cost of surveillance systems, visual data are increasingly being used in digital forensic investigation. Digital videos have been widely used as key evidence sources in evidence identification, analysis, presentation, and report. The main goal of this paper is to develop advanced forensic video analysis techniques to assist the forensic investigation. We first propose a forensic video analysis framework that employs an efficient video/image enhancing algorithm for the low quality of footage analysis. An adaptive video enhancement algorithm based on contrast limited adaptive histogram equalization (CLAHE) is introduced to improve the closed-circuit television (CCTV) footage quality for the use of digital forensic investigation. To assist the video-based forensic analysis, a deeplearning-based object detection and tracking algorithm are proposed that can detect and identify potential suspects and tools from footages. INDEX TERMS Forensics investigation, forensic video analysis, video/image enhancement, object detection, deep learning.

A Survey on the Internet of Things (IoT) Forensics: Challenges, Approaches, and Open Issues

“…Therefore, the forensics discipline needs video analysis frameworks that employ efficient image enhancing algorithms for low quality footage. An approach that addresses the aforementioned issue was presented by Xiao et al [145]. Their work introduces a video enhancement algorithm based on contrast limited adaptive histogram equalization (CLAHE).…”

“…Unfortunately, the majority of the recordings (e.g., security and monitoring cameras) differ from gallery images because of their low resolution quality, angle of view, color, data rate, etc. In order to guarantee a precise forensic identification process, new techniques for video quality improvement need to be developed, for example advanced robust evidence extraction and subject detection methods [145]. Intelligent techniques like deep learning could also improve and shorten the evidence identification phase.…”

“…Because of that, forensics professionals are expected to use an extended amount of available forensics techniques, such as facial ageing, marks, near-infrared face recognition or forensics sketch recognition, etc. [145]. With the growing number of devices and data volume, there will be an even greater need for further in-depth authenticity examinations and validation procedures [68].…”

“…Eventually, there have been many attempts to integrate artificial intelligence (including machine learning and deep learning approaches) in the security and digital forensics practices. Intelligent methods have been applied for anomaly detection [59], forensics video analysis [145], rule extraction [186], intrusion classification [177]. Buczak and Guven [187] for example, presented a literature survey, which discusses application areas of different intelligent methods, with focus on data mining techniques for intrusion detection.…”

Novel framework for optimized digital forensic for mitigating complex image attacks

“…The authors have also used the method for the identification of median filtering. Xiao et al [12] have developed a machine learning approach of histogram equalization towards improving the captures of surveillance system aiming forensic application. Ma et al [13] have used a convolution neural network for assisting in image retrieval.…”