Scribd
Upload
10 views

Fake Image Detection

The document discusses the detection of fake images using machine learning techniques, particularly through neural networks and Error Level Analysis. It highlights the importance of identifying manipulated images to combat cyber crimes and outlines the use of Convolutional Neural Networks for effective classification of original and forged images. Additionally, it specifies the hardware and software requirements needed for implementing the detection system.

Uploaded by

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

Fake Image Detection

The document discusses the detection of fake images using machine learning techniques, particularly through neural networks and Error Level Analysis. It highlights the importance of identifying manipulated images to combat cyber crimes and outlines the use of Convolutional Neural Networks for effective classification of original and forged images. Additionally, it specifies the hardware and software requirements needed for implementing the detection system.

Uploaded by

achudasth
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 3

FAKE IMAGE DETECTION USING MACHINE LEARNING

ABSTRACT

Many fake images are spreading through digital media nowadays. Detection of
such fake images is inevitable for the unveiling of the image based cyber
crimes. Forging images and identifying such images are promising research
areas in this digital era. The tampered images are a detected using neural
network which also recognizes the regions of the image that have been
manipulated and reveals the segments of the original image. It can be
implemented on Android platform and hence made available to common users.
The compression ratio of the foreign content in a fake image is different from
that of the original image and is detected using Error Level Analysis. Another
feature used along with compression ratio is image metadata. Although it is
possible to alter metadata content making it unreliable on its own, here it is used
as a supporting parameter for error level analysis decision .

MODULES

1.Error Level Analysis

Error level analysis is one technique for knowing images that have been
manipulated by storing images at a certain quality level and then calculating the
difference from the compression level . When JPEG was first saved, then it will
compress the image the first time, most editing software like adobe photoshop,
gimp, and adobe lightroom support JPEG compressing operation.If the image is
rescheduled using image editing software, then compressed again.

So it shows that the original image when the first image is taken using a digital
camera has been compressed twice, first use the camera and the second is

1
editing software. When viewed with the naked eye the image looks the same,
but by using this method it will look the difference between a forgery image
with the original image.

2. Convolutional Neural Network

CNN has a high ability in large-scale image classification. Cnn consists of three
layers: convolutional layer, pooling layers, and fully connection layers .A
Convolutional layer and pooling layer is the most important layer on CNN.
Convolutional layer is used for extract feature by combining the image area
with many filters. Pooling layer reduces the size of the output map of the
convolution layer and prevents overfitting. Through these two layers the
number of neurons, parameters, and connections is much less than there is a
CNN model. This makes CNN more efficient compared with BP networks with
similar layers.Applying CNN for fake image classification and original image
converted into error level form on image. We know through the previous
literature that CNN can achieve competitive performance and even better than
humans in some visual problems, and we wanted to test CNN's ability to
classify forgery image and original images via Error Level Analysis.

HARDWARE AND SOFTWARE REQUIREMENTS

HARDWARE REQUIREMENTS

System : Pentium IV 2.4 GHz


Hard Disk : 200GB.
Monitor : 15 VGA Colour.
Mouse : Logitech.

2
SOFTWARE REQUIREMENT

Operating system : Windows 7 and above.


Coding Language : python
Database : Mysql
Ram : 4GB

You might also like

pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy