Hybrid Signal Processing and Deep Feature Fusion for Image Forgery Detection

Shalini Kumari; Komal Yadav

Hybrid Signal Processing and Deep Feature Fusion for Image Forgery Detection

Author(s)	Ms. Shalini Kumari, Dr. Komal Yadav
Country	India
Abstract	The rapid advancement of image editing tools and artificial intelligence has led to a significant increase in sophisticated image forgeries, raising serious concerns regarding the credibility of digital media, forensic analysis, and information security. Traditional image forgery detection techniques based on handcrafted signal processing features often lack robustness against geometric transfor-mations, compression, and post-processing operations. Conversely, deep learning–based approaches, although effective in learning high-level semantic representations, commonly suffer from dataset de-pendency, high computational complexity, and limited generalization to unseen forgery types. To address these challenges, this paper proposes a hybrid image forgery detection framework that integrates signal processing–based features with deep learning features to exploit their comple-mentary strengths. In the proposed approach, low-level statistical and multi-resolution features are extracted using signal processing techniques to capture subtle manipulation artifacts, while high-level discriminative features are learned using a convolutional neural network to model complex structural and semantic information. The extracted features are fused into a unified representation and refined using dimensionality reduction to reduce redundancy and improve computational efficiency. A su-pervised classification model is then employed to distinguish between authentic and forged images. Extensive experiments conducted on benchmark image forgery datasets demonstrate that the proposed framework achieves an average detection accuracy of 98.2%, with precision, recall, and F1-score values of 97.9%, 98.4%, and 98.1%, respectively. Moreover, the proposed method shows strong robustness against post-processing attacks, including JPEG compression, scaling, and noise addition, with less than 2% performance degradation under challenging conditions. Cross-dataset evaluation further confirms improved generalization, outperforming existing classical and deep learning–based methods by 3–6% in accuracy. These results validate that hybrid signal processing and deep feature fusion provides an effective, robust, and reliable solution for real-world image forgery detection.
Field	Computer > Artificial Intelligence / Simulation / Virtual Reality
Published In	Volume 8, Issue 3, May-June 2026
Published On	2026-05-13

View / Download PDF File

E-ISSN 2582-2160

doi

CrossRef DOI is assigned to each research paper published in our journal.

IJFMR DOI prefix is
10.36948/ijfmr

Downloads

Research Paper Format Copyright Permission Form and Undertaking Form Cover Page Vol 8 Isu 2 Cover Page Vol 8 Isu 1 Cover Page Vol 7 Isu 6

All research papers published on this website are licensed under Creative Commons Attribution-ShareAlike 4.0 International License, and all rights belong to their respective authors/researchers.

CC-BY-SA

About IJFMR Fees & Payment Current Issue Publication Archive	Submit Research Paper Track Submission Status Publication Guidelines Publication Ethics Peer Review & Plagiarism	Join as a Reviewer Editors & Reviewers Reviewer Referral Program Get Reviewer Membership Certi.	Website/Journal Policies Usage Policy Content Policies Privacy Policy

Contact Us		+91-9687-828-838	editor@ijfmr.com

International Journal For Multidisciplinary Research

A Widely Indexed Open Access Peer Reviewed Multidisciplinary Bi-monthly Scholarly International Journal

Hybrid Signal Processing and Deep Feature Fusion for Image Forgery Detection

Share this