Scribd
Upload
14 views6 pages

Presentation Phase-I

The document describes an image forgery detection system using convolutional neural networks. It discusses how CNNs are well-suited for this task due to their ability to extract subtle image features. The system consists of preprocessing, feature extraction via CNNs, and classification to detect tampered regions and assess forgery levels. Literature on the topic is reviewed, noting intrinsic camera features provide more reliable detection than software-based methods. The system architecture is presented, dividing work between preprocessing, model design/training, and analysis. References are provided on image forensics techniques.

Uploaded by

vishalkhendake1426
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
14 views6 pages

Presentation Phase-I

The document describes an image forgery detection system using convolutional neural networks. It discusses how CNNs are well-suited for this task due to their ability to extract subtle image features. The system consists of preprocessing, feature extraction via CNNs, and classification to detect tampered regions and assess forgery levels. Literature on the topic is reviewed, noting intrinsic camera features provide more reliable detection than software-based methods. The system architecture is presented, dividing work between preprocessing, model design/training, and analysis. References are provided on image forensics techniques.

Uploaded by

vishalkhendake1426
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 6

Dr. D. Y.

Patil Institute of Technology, Pimpri, Pune-18


Department of Electronics and Telecommunication
Engineering
Project Phase-I

Project Title: Image Forgery Detection System

Gauri Fule BEB05


Aditya Dhavale BEB15
Vishal Khendake BEB58

Guided By : Ms.Deepali Bendale


Introduction:

In the age of digital manipulation, the trustworthiness of images is at risk due to


rampant forgery. To combat this, we present an Image Forgery Detection System
powered by Convolutional Neural Networks (CNN). This system is designed to
automatically detect forged or manipulated regions in digital images.
CNNs, known for their prowess in image analysis, play a
pivotal role in our system. They excel in extracting intricate features, making them
ideal for identifying subtle anomalies introduced by forgers. Our system consists of
preprocessing, feature extraction, and CNN-based classification, providing
transparency in detecting tampered areas and assessing the level of forgery.
Observations from literature survey:
Ref No. Author Name Contribution Observations

1 Kunj Bihari Meena, Reviewed various techniques in digital Intrinsic features based methods of camera
Vipin Tyagi [2019] camera image forensics hardware are more reliable and better in terms
of accuracy as compared to methods based on
other camera software parts.
Camera identification methods outperform as
compared to other forgery detection methods.

2 Mohassin Ahmad Categorized the image forgery Due to the advancement of image manipulation
and Farida detection techniques into five groups tools, an arms race between the forger and
Khursheed. [2021] (pixel-based. format-based, camera- forensic analyst is inescapable.
based. physically based and geometric-
bused)

3 Marcello Zanardelli, Reviewed various method based on Existing methods produce considerably higher
Riccardo Leonardii, blind image forgery false positive rates than which are reported in
Nicola Adami. [2022] the existing papers.

Need to develop more reliable and robust


methods.

4 Zankhana J. Barad, Reviewed state of the art approaches Keypoint-based methods better than block-
Mukesh M. pertaining to copy-move forgery based methods in term of performance
Goswami. [2020] (execution time), however, block-bused
methods give better detection accuracy.
System Architecture :

Fig: System architecture


Work Distribution :

• Member 1: Led the literature review and gathered relevant research papers. Prepared the
front end part using HTML, CSS, Javascript. Explored challenges and suggested future
research directions. Investigated the challenges and potential future research directions

• Member 2: Designed and implemented the core CNN-based forgery detection model.
Assisted in data preprocessing and augmentation. Collected and preprocessed the dataset for
training and testing. Explored and implemented advanced CNN architectures for comparison.

• Member 3: Conducted an in-depth comparative analysis of different forgery detection


methods. Compiled the list of references and maintained proper citation throughout the report.
Ensured the consistency of the report's format, style, and citations.
Refrences:

1. Kunj Bihari Meena, Vipin Tyagi.: Image Forgery Detection: Survey and Future
Direction. IEEE International Conference Indian Inform, (2019)
2. Mohassin Ahmad and Farida Khursheed.: Digital Images Forgery Detection
Approaches. A Cataloging in Publication Record, Artech House Inc, Norwood, MA,
USA (2021)
3. Marcello Zanardelli, Riccardo Leonardii, Nicola Adami.: Image forgery detection: A
survey of recent Deep Learning approaches. IEEE Trans. Multimedia, (2022)
4. Zankhana J. Barad, Mukesh M. Goswami.: Image Forgery Detection using Deep
Learning: A Survey. In: IEEE International Conference on Image Processing. (2020)
5. Mohammed, Tajuddin Manhar Bunk, Jason Nataraj, Lakshmananent.: Boosting Image
Forgery Detection using Resampling Features and Copy-move Analysis, (2018)

You might also like