UPM Institutional Repository

Splicing image forgery identification based on artificial neural network approach and texture features


Mohd Omar, Nur Fareha Amira (2019) Splicing image forgery identification based on artificial neural network approach and texture features. Masters thesis, Universiti Putra Malaysia.


In this technology area, manipulation an image become an easy task due to the availability of open source image handling software and becomes a great challenge to determine whether an image has been manipulated or not. Moreover, the authenticity of digital image experience extreme dangers because the capable of altering images software that effectively adjust the image without leaving any obvious hint of such change. Therefore, image integrity is becoming questionable especially when images have influential power for example, in a court of law or news report. Manipulating the original image content is called digital image forgery. Splicing image forgery is one of technique to forgery an image. The splicing image forgery is replicated one or more are from source image and paste into an objective picture to create a composite image. This study present combination of features extraction to produce good vector to describe the image and feed the image to the multilayer perceptron. This study is try to improve the accuracy identification on splicing image based on anchor paper. The finding outcome from this study have shown improved approach for identification splicing image. The identification accuracy in the technique used is about 100% and 98% based on dataset.

Download File

[img] Text
FSKTM 2019 30 IR.pdf

Download (896kB)

Additional Metadata

Item Type: Thesis (Masters)
Subject: Image processing - Digital techniques
Subject: Image processing
Call Number: FSKTM 2019 30
Chairman Supervisor: Puan Zaiton Binti Muda
Divisions: Faculty of Computer Science and Information Technology
Depositing User: Mas Norain Hashim
Date Deposited: 24 Jul 2020 02:27
Last Modified: 24 Jul 2020 02:27
URI: http://psasir.upm.edu.my/id/eprint/82953
Statistic Details: View Download Statistic

Actions (login required)

View Item View Item