Deepfake Face Images Detection
No Thumbnail Available
Date
2024
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Bahrain Polytechnic
Abstract
Deepfake is a sort of AI that forges original image or video and create persuading images, audio and video. Deepfake media continues to gain ground online,
raising a number of ethical and moral questions about their use, in that deepfakes
can be used to undermine political elections, companies, individual and corporate
finances, reputation, and many more. The proposed system to solve this problem is to use the most popular algorithm in deep learning, Convolution Neural
Network (CNN), for detecting fake images. This will be achieved by training two
deep learning models and analyzing their performances in distinguishing between
the two classes of images “Real”,” Fake”.
Our main aim is to contribute a useful framework toward the detection of
deep-fake photos with deep learning. This thesis proposed convolutional neural
networks for the identification of genuine and deepfake pictures. In this study,
we have trained two models: DenseNet121 and ResNet50. The results will be
categorized by Four evaluation metrics: accuracy, precision, recall, and F1-score.
In that respect, DenseNet121 had the best performance with an accuracy of 94%.
Besides, we obtained 91% from the ResNet50.
Description
Chapter 1: Introduction
This chapter introduces the topic of research by providing an overview of
deepfake technology, its implications, and why it is critical to identify deep
fake face images. It also gives background information and context, defines
key terminologies, presents the problem statement, and outlines the aim
and research questions guiding the study.
Chapter 2: Literature Review
This chapter reviews existing research and technologies on detection models
for deepfake face images. It emphasizes a survey of methodologies, models,
and techniques developed and implemented in the subject area to date,
thereby situating this research within the appropriate context.
Chapter 3: Methodologies
This chapter details the methodologies used in developing the deepfake
face image detection models, including the processes of data collection, pre
processing, model architecture design, training procedures, and evaluation
metrics. The rationale behind the selection of specific approaches and models is also explained.
Chapter 4: Results and Discussion
This chapter elaborates on the research findings through the presentation
of performance metrics for the various tested models. It includes a discussion on the outcomes, the effectiveness of the techniques tested, and a
comparison of results with other related research. Additionally, this chapter
highlights trends, challenges, and key results.
Chapter 5: Conclusions and Future Work
This chapter provides a summary of the research findings, discusses the
limitations of the study, and offers recommendations for future research. It
broadens the impact of the research within the context of deepfake detection
and outlines potential directions for further advancements in the field.
Keywords
Deepfake, Detect, Neural Networks, Images, CNN