CNN-Based Deep Learning Approach Over Image Data for Cyber Forensic Investigation

CNN-Based Deep Learning Approach Over Image Data for Cyber Forensic Investigation

Aishwary Awasthi, Priyanksha Das, Rupal Gupta, Raj Varma, Shilpa Sharma, Ankur Gupta, Huma Khan
Copyright: © 2023 |Pages: 19
DOI: 10.4018/978-1-6684-8618-4.ch011
OnDemand:
(Individual Chapters)
Available
$37.50
Current Special Offers
No Current Special Offers
TOTAL SAVINGS: $37.50

Abstract

A deep learning approach is gaining popularity day by day in image data classification. The process of classification of graphical data considering training network is managed by conventional neural network. Such types of networks allow automatic classification by making use of CNN approach. But the issues that are faced during forensic investigation are slow in performance and lack accuracy. The major objective of work is to consider the CNN approach that is processing graphic data in order to perform cyber forensic investigation.
Chapter Preview
Top

1. Introduction

Cyber forensics is the discipline concerned with the acquisition, analysis, reporting, and presentation of digital evidence in computer-related legal proceedings. (Kumar, B., 2023) The hard disc or even deleted data may provide crucial evidence. Data acquisition and analysis from a system or device is what this term refers to, and it is done so that the information may be written down and submitted in court (S.Z.D. et al., 2022). The system's one-of-a-kind storage cell must be copied digitally or in some other non-physical form before the test may proceed. When a security breach occurs, it's important to undertake a comprehensive cyber forensics investigation to find out who's responsible. All of the analysis is done on the backup copy of the programme, which has no effect on the operational system. This work is considering role of deep learning, CNN and image processing in cyber forensic investigation (Naskar, R., 2018).

1.1 Deep Learning

Essentially, a NN with three or more layers is the DL subfield of machine learning. ANN “learn” using extensive datasets in an effort to mimic the human brain's operation; however they are still far from reaching the brain's level of sophistication. A single-layer neural network can still create approximations, but it's difficult to tune and tweak for precision without additional hidden layers. Some AI applications and services use DL to increase automation by taking over analytical and physical tasks that were formerly performed exclusively by humans (Nassa V. K., 2021).

1.2 Deep Learning Applications

In practise, DL applications improve people's daily lives, but they are generally embedded so unobtrusively in goods and services that consumers are blissfully ignorant of the complex data processing occurring in the background (Barik, L., 2020).

Figure 1.

Deep learning

978-1-6684-8618-4.ch011.f01

Complete Chapter List

Search this Book:
Reset