journal

A Novel Image Tamper Detection Approach by Blending Forensic Tools and Optimized CNN: Sealion Customized Firefly Algorithm

Area/Stream: Computer Science,
Authors: Mohassin Ahmad, Farida Khursheed
Keywords: Image-Tamper Detection, Fused Framework for Feature Extraction, Optimized CNN, Sealion Customized Firefly Algorithm
Journal Name: Multimedia Tools and Applications

Year:2021,
Month:November,
Volume: 80,
Issue: 28-29

Page No:,
ISSN: 1573-7721,
DOI/Link: https://doi.org/10.1007/s11042-021-11529-0


Abstract:

In recent days, the digital images are manipulated more professionally and easily via common image processing tools. This has been highly practiced in diverse applications including the surveillance systems, where the tamper detection with higher reliability is essential. A novel image tamper detection framework is designed with two major phases: fused feature extraction framework and tamper detection. The collected data are subjected to the fused feature extraction framework, where the features like adaptive speeded up robust features (SURF), Discrete Wavelet Transform (DWT) based Patched Local Vector Pattern (LVP) features, Proposed Principal Component Analysis (PCA) based Histogram of Oriented Gradients (HoG) feature and Mode Based First Digit Feature (MBFDF) are extracted. Subsequently, the extracted features are fed as the input to Optimized Convolutional Neural Network (CNN), which results in the type of tampering in the image: copy-move, splicing, noise inconsistency and double compression. To make the detection more accurate, the weights of CNN are fine-tuned by a new hybrid optimization algorithm referred as Sealion Customized Firefly algorithm (SCFF). The proposed hybrid optimization algorithm is the amalgamation of the standard Sea Lion Optimization Algorithm (SLnO) and Firefly Algorithm (FF). Finally, a comparative evaluation is made between the proposed and existing works in terms of certain performance measures as well.

Cite this: Mohassin Ahmad, Farida Khursheed,"A Novel Image Tamper Detection Approach by Blending Forensic Tools and Optimized CNN: Sealion Customized Firefly Algorithm", Multimedia Tools and Applications, 80, 28-29, November, 2021, , 1573-7721, https://doi.org/10.1007/s11042-021-11529-0
Views: 4223 Download File
News

Index your research paper @ RSquareL

Call for research papers evaluation Volume 3 Issue 2

Get listed your profile under listing based on your RSquareL Value

Registration for Indexing Author Journal Publisher Conference Organizer
Research Recognition & Listing Young Researcher Young Achiever Research Excellence

Contact Us

RSquareL is the indexing platform developed by Global Academicians & Researchers Network (GARNet.). RSquareL is the abstract database of peer-reviewed scientific journals, books, and conference proceedings that covers research topics across all scientific, technical, and medical disciplines.

Contact Details

Contact Email: publish@rsquarel.org
Write to Us: Click Here
Counter Start Date: 27-12-2021 Flag Counter

© 2024 RSquareL