CNN Based Image Forgery Detection Using Pre-trained AlexNet Model

6 Pages Posted: 20 Mar 2019

See all articles by Amit Doegar

Amit Doegar

National Institute of Technical Teachers Training and Research (NITTTR)

Maitreyee Dutta

National Institute of Technical Teachers Training and Research (NITTTR)

Kumar Gaurav

Magma Research and Consultancy Services

Date Written: March 19, 2019

Abstract

Image forgery detection is an approach for detection and localization of forged component from a manipulated image. To find manipulation or tampering in the original image, an adequate number of features are required to classify the given image is either a forged or non-forged. To achieve this convolutional neural network (CNN) based pre-trained AlexNet model's deep features have been utilized which are efficient and effective, as compared to the existing state-of-the-art approaches on publicly available benchmark dataset MICC-F220. The experiment result shows that the proposed approach using a pre-trained AlexNet model based deep features with Support Vector Machine (SVM) classifier has achieved 93.94% accuracy.

Suggested Citation

Doegar, Amit and Dutta, Maitreyee and Gaurav, Kumar, CNN Based Image Forgery Detection Using Pre-trained AlexNet Model (March 19, 2019). International Journal of Computational Intelligence & IoT, Vol. 2, No. 1, 2019. Available at SSRN: https://ssrn.com/abstract=3355402

Amit Doegar (Contact Author)

National Institute of Technical Teachers Training and Research (NITTTR) ( email )

Sector 26
CHANDIGARH, 160019
India

Maitreyee Dutta

National Institute of Technical Teachers Training and Research (NITTTR)

Sector 26
CHANDIGARH, 160019
India

Kumar Gaurav

Magma Research and Consultancy Services ( email )

Ambala, 133006
India

Here is the Coronavirus
related research on SSRN

Paper statistics

Downloads
297
Abstract Views
894
rank
107,339
PlumX Metrics