A Comparative Study of Secure Image Deduplication Techniques Using Hashing and Deep Learning Approaches

The rapid growth of image data in cloud storage environments has increased the need for efficient image deduplication methods that eliminate redundant information while supporting privacy, security, and storage efficiency. This paper presents a review and reported experimental comparison of image deduplication approaches, including cryptographic hashing, perceptual hashing, encryption-based methods, proof-of-ownership mechanisms, and deep learning techniques. Existing solutions are categorized according to their security functions, computational requirements, duplicate-detection capabilities, and suitability for cloud-based applications. The comparative evaluation uses the INRIA Holidays and Oxford Buildings datasets and considers SHA-256, perceptual hashing (pHash), and MobileNetV2. According to the reported results, SHA-256 identifies exact duplicate files with 100% accuracy but does not identify transformed near-duplicates, whereas pHash provides moderate near-duplicate detection. MobileNetV2 yields reported near-duplicate detection accuracies of 93.94% and 97.93% on INRIA Holidays and Oxford Buildings, respectively. These findings suggest that deep feature representations can improve the recognition of visually similar images under the specified testing conditions, although a similarity model alone does not establish cryptographic confidentiality or storage-system security. The study therefore distinguishes the accuracy of image matching from the security properties of a complete deduplication service, identifies limitations in the reported evaluation protocol, and discusses potential directions for privacy-preserving image management in cloud and Internet of Things (IoT) environments.

Authors

Publication Details

Journal
Archives des sciences a multidisciplinary journal
Published
2026-11-20
DOI
https://doi.org/10.68304/as/76506
Primary Topic
Data Quality and Management
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

A Comparative Study of Secure Image Deduplication Techniques Using Hashing and Deep Learning Approaches

Rahul Shah
Archives des sciences a multidisciplinary journal
Data Quality and Management
article

A Comparative Study of Secure Image Deduplication Techniques Using Hashing and Deep Learning Approaches

Rahul Shah
article en

Abstract

The rapid growth of image data in cloud storage environments has increased the need for efficient image deduplication methods that eliminate redundant information while supporting privacy, security, and storage efficiency. This paper presents a review and reported experimental comparison of image deduplication approaches, including cryptographic hashing, perceptual hashing, encryption-based methods, proof-of-ownership mechanisms, and deep learning techniques. Existing solutions are categorized according to their security functions, computational requirements, duplicate-detection capabilities, and suitability for cloud-based applications. The comparative evaluation uses the INRIA Holidays and Oxford Buildings datasets and considers SHA-256, perceptual hashing (pHash), and MobileNetV2. According to the reported results, SHA-256 identifies exact duplicate files with 100% accuracy but does not identify transformed near-duplicates, whereas pHash provides moderate near-duplicate detection. MobileNetV2 yields reported near-duplicate detection accuracies of 93.94% and 97.93% on INRIA Holidays and Oxford Buildings, respectively. These findings suggest that deep feature representations can improve the recognition of visually similar images under the specified testing conditions, although a similarity model alone does not establish cryptographic confidentiality or storage-system security. The study therefore distinguishes the accuracy of image matching from the security properties of a complete deduplication service, identifies limitations in the reported evaluation protocol, and discusses potential directions for privacy-preserving image management in cloud and Internet of Things (IoT) environments.

Archives des sciences a multidisciplinary journalVol. 76(5)
Data Quality and Management
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon