Local Contrast Criterion for Verifiable Image Enhancement Network: Layered Difference Representation Loss

Deep neural networks (DNN) have made significant improvements in image processing, particularly in media forensic investigations. However, the resulting images or frames from DNN-based algorithms are typically not admissible as evidence because these algorithms do not precisely verify the internal p...

Full description

Saved in:
Bibliographic Details
Main Authors: Jin-Hwan Kim, Hanul Kim, Jae Sung Lim, Nam In Park
Format: Article
Language:English
Published: IEEE 2024-01-01
Series:IEEE Access
Subjects:
Online Access:https://ieeexplore.ieee.org/document/10786199/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850100823608000512
author Jin-Hwan Kim
Hanul Kim
Jae Sung Lim
Nam In Park
author_facet Jin-Hwan Kim
Hanul Kim
Jae Sung Lim
Nam In Park
author_sort Jin-Hwan Kim
collection DOAJ
description Deep neural networks (DNN) have made significant improvements in image processing, particularly in media forensic investigations. However, the resulting images or frames from DNN-based algorithms are typically not admissible as evidence because these algorithms do not precisely verify the internal processes from input to output. This study proposes an efficient local contrast enhancement criterion for a layered difference representation (LDR) loss, a verifiable image enhancement network. The LDR originally derives a transformation function based on neighboring pixel value differences. However, appending an additional constraint, such as image similarity to the ground truth, is challenging. To address this, we utilize DNNs and introduce a novel criterion, LDR loss, for image enhancement. The LDR loss aims to increase neighboring pixel differences, whereas the image loss ensures similarity with the ground truth, thus enhancing both global and local contrasts of the image. Experimental results demonstrate that the proposed algorithm outperforms conventional image enhancement algorithms.
format Article
id doaj-art-5bfde236024b48b19e45ba97c03efaba
institution DOAJ
issn 2169-3536
language English
publishDate 2024-01-01
publisher IEEE
record_format Article
series IEEE Access
spelling doaj-art-5bfde236024b48b19e45ba97c03efaba2025-08-20T02:40:11ZengIEEEIEEE Access2169-35362024-01-011219168219169410.1109/ACCESS.2024.351341910786199Local Contrast Criterion for Verifiable Image Enhancement Network: Layered Difference Representation LossJin-Hwan Kim0https://orcid.org/0009-0003-5930-7795Hanul Kim1https://orcid.org/0000-0001-7450-6600Jae Sung Lim2https://orcid.org/0009-0004-5886-3850Nam In Park3Digital Analysis Division, National Forensic Service, Wonju, South KoreaDept. of Applied Artificial Intelligence, Seoul National University of Science and Technology, Seoul, South KoreaDigital Analysis Division, National Forensic Service, Wonju, South KoreaDigital Analysis Division, National Forensic Service, Wonju, South KoreaDeep neural networks (DNN) have made significant improvements in image processing, particularly in media forensic investigations. However, the resulting images or frames from DNN-based algorithms are typically not admissible as evidence because these algorithms do not precisely verify the internal processes from input to output. This study proposes an efficient local contrast enhancement criterion for a layered difference representation (LDR) loss, a verifiable image enhancement network. The LDR originally derives a transformation function based on neighboring pixel value differences. However, appending an additional constraint, such as image similarity to the ground truth, is challenging. To address this, we utilize DNNs and introduce a novel criterion, LDR loss, for image enhancement. The LDR loss aims to increase neighboring pixel differences, whereas the image loss ensures similarity with the ground truth, thus enhancing both global and local contrasts of the image. Experimental results demonstrate that the proposed algorithm outperforms conventional image enhancement algorithms.https://ieeexplore.ieee.org/document/10786199/Contrast enhancementdeep neural networkimage enhancementtransformation function
spellingShingle Jin-Hwan Kim
Hanul Kim
Jae Sung Lim
Nam In Park
Local Contrast Criterion for Verifiable Image Enhancement Network: Layered Difference Representation Loss
IEEE Access
Contrast enhancement
deep neural network
image enhancement
transformation function
title Local Contrast Criterion for Verifiable Image Enhancement Network: Layered Difference Representation Loss
title_full Local Contrast Criterion for Verifiable Image Enhancement Network: Layered Difference Representation Loss
title_fullStr Local Contrast Criterion for Verifiable Image Enhancement Network: Layered Difference Representation Loss
title_full_unstemmed Local Contrast Criterion for Verifiable Image Enhancement Network: Layered Difference Representation Loss
title_short Local Contrast Criterion for Verifiable Image Enhancement Network: Layered Difference Representation Loss
title_sort local contrast criterion for verifiable image enhancement network layered difference representation loss
topic Contrast enhancement
deep neural network
image enhancement
transformation function
url https://ieeexplore.ieee.org/document/10786199/
work_keys_str_mv AT jinhwankim localcontrastcriterionforverifiableimageenhancementnetworklayereddifferencerepresentationloss
AT hanulkim localcontrastcriterionforverifiableimageenhancementnetworklayereddifferencerepresentationloss
AT jaesunglim localcontrastcriterionforverifiableimageenhancementnetworklayereddifferencerepresentationloss
AT naminpark localcontrastcriterionforverifiableimageenhancementnetworklayereddifferencerepresentationloss