WMNet: A Lossless Watermarking Technique Using Deep Learning for Medical Image Authentication
Traditional watermarking techniques extract the watermark from a suspected image, allowing the copyright information regarding the image owner to be identified by the naked eye or by similarity estimation methods such as bit error rate and normalized correlation. However, this process should be more...
Main Authors: | , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2021-04-01
|
Series: | Electronics |
Subjects: | |
Online Access: | https://www.mdpi.com/2079-9292/10/8/932 |
_version_ | 1797412189068328960 |
---|---|
author | Yueh-Peng Chen Tzuo-Yau Fan Her-Chang Chao |
author_facet | Yueh-Peng Chen Tzuo-Yau Fan Her-Chang Chao |
author_sort | Yueh-Peng Chen |
collection | DOAJ |
description | Traditional watermarking techniques extract the watermark from a suspected image, allowing the copyright information regarding the image owner to be identified by the naked eye or by similarity estimation methods such as bit error rate and normalized correlation. However, this process should be more objective. In this paper, we implemented a model based on deep learning technology that can accurately identify the watermark copyright, known as WMNet. In the past, when establishing deep learning models, a large amount of training data needed to be collected. While constructing WMNet, we implemented a simulated process to generate a large number of distorted watermarks, and then collected them to form a training dataset. However, not all watermarks in the training dataset could properly provide copyright information. Therefore, according to the set restrictions, we divided the watermarks in the training dataset into two categories; consequently, WMNet could learn and identify the copyright information that the watermarks contained, so as to assist in the copyright verification process. Even if the retrieved watermark information was incomplete, the copyright information it contained could still be interpreted objectively and accurately. The results show that the method proposed by this study is relatively effective. |
first_indexed | 2024-03-09T04:58:25Z |
format | Article |
id | doaj.art-b56507ed00444e8d9f6d52bf3273d6fc |
institution | Directory Open Access Journal |
issn | 2079-9292 |
language | English |
last_indexed | 2024-03-09T04:58:25Z |
publishDate | 2021-04-01 |
publisher | MDPI AG |
record_format | Article |
series | Electronics |
spelling | doaj.art-b56507ed00444e8d9f6d52bf3273d6fc2023-12-03T13:02:46ZengMDPI AGElectronics2079-92922021-04-0110893210.3390/electronics10080932WMNet: A Lossless Watermarking Technique Using Deep Learning for Medical Image AuthenticationYueh-Peng Chen0Tzuo-Yau Fan1Her-Chang Chao2Center for Artificial Intelligence in Medicine, Chang Gung Memorial Hospital, Linkou, Taoyuan 33305, TaiwanCenter for Artificial Intelligence in Medicine, Chang Gung Memorial Hospital, Linkou, Taoyuan 33305, TaiwanDepartment of Computer Science and Information Engineering, Ming Chuan University, Guei-Shan, Taoyuan 33348, TaiwanTraditional watermarking techniques extract the watermark from a suspected image, allowing the copyright information regarding the image owner to be identified by the naked eye or by similarity estimation methods such as bit error rate and normalized correlation. However, this process should be more objective. In this paper, we implemented a model based on deep learning technology that can accurately identify the watermark copyright, known as WMNet. In the past, when establishing deep learning models, a large amount of training data needed to be collected. While constructing WMNet, we implemented a simulated process to generate a large number of distorted watermarks, and then collected them to form a training dataset. However, not all watermarks in the training dataset could properly provide copyright information. Therefore, according to the set restrictions, we divided the watermarks in the training dataset into two categories; consequently, WMNet could learn and identify the copyright information that the watermarks contained, so as to assist in the copyright verification process. Even if the retrieved watermark information was incomplete, the copyright information it contained could still be interpreted objectively and accurately. The results show that the method proposed by this study is relatively effective.https://www.mdpi.com/2079-9292/10/8/932convolutional neural networkdeep learningwatermarking technique |
spellingShingle | Yueh-Peng Chen Tzuo-Yau Fan Her-Chang Chao WMNet: A Lossless Watermarking Technique Using Deep Learning for Medical Image Authentication Electronics convolutional neural network deep learning watermarking technique |
title | WMNet: A Lossless Watermarking Technique Using Deep Learning for Medical Image Authentication |
title_full | WMNet: A Lossless Watermarking Technique Using Deep Learning for Medical Image Authentication |
title_fullStr | WMNet: A Lossless Watermarking Technique Using Deep Learning for Medical Image Authentication |
title_full_unstemmed | WMNet: A Lossless Watermarking Technique Using Deep Learning for Medical Image Authentication |
title_short | WMNet: A Lossless Watermarking Technique Using Deep Learning for Medical Image Authentication |
title_sort | wmnet a lossless watermarking technique using deep learning for medical image authentication |
topic | convolutional neural network deep learning watermarking technique |
url | https://www.mdpi.com/2079-9292/10/8/932 |
work_keys_str_mv | AT yuehpengchen wmnetalosslesswatermarkingtechniqueusingdeeplearningformedicalimageauthentication AT tzuoyaufan wmnetalosslesswatermarkingtechniqueusingdeeplearningformedicalimageauthentication AT herchangchao wmnetalosslesswatermarkingtechniqueusingdeeplearningformedicalimageauthentication |