ZWNet: A Deep-Learning-Powered Zero-Watermarking Scheme with High Robustness and Discriminability for Images
In order to safeguard image copyrights, zero-watermarking technology extracts robust features and generates watermarks without altering the original image. Traditional zero-watermarking methods rely on handcrafted feature descriptors to enhance their performance. With the advancement of deep learnin...
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2024-01-01
|
| Series: | Applied Sciences |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2076-3417/14/1/435 |
| _version_ | 1797359087461072896 |
|---|---|
| author | Can Li Hua Sun Changhong Wang Sheng Chen Xi Liu Yi Zhang Na Ren Deyu Tong |
| author_facet | Can Li Hua Sun Changhong Wang Sheng Chen Xi Liu Yi Zhang Na Ren Deyu Tong |
| author_sort | Can Li |
| collection | DOAJ |
| description | In order to safeguard image copyrights, zero-watermarking technology extracts robust features and generates watermarks without altering the original image. Traditional zero-watermarking methods rely on handcrafted feature descriptors to enhance their performance. With the advancement of deep learning, this paper introduces “ZWNet”, an end-to-end zero-watermarking scheme that obviates the necessity for specialized knowledge in image features and is exclusively composed of artificial neural networks. The architecture of ZWNet synergistically incorporates ConvNeXt and LK-PAN to augment the extraction of local features while accounting for the global context. A key aspect of ZWNet is its watermark block, as the network head part, which fulfills functions such as feature optimization, identifier output, encryption, and copyright fusion. The training strategy addresses the challenge of simultaneously enhancing robustness and discriminability by producing the same identifier for attacked images and distinct identifiers for different images. Experimental validation of ZWNet’s performance has been conducted, demonstrating its robustness with the normalized coefficient of the zero-watermark consistently exceeding 0.97 against rotation, noise, crop, and blur attacks. Regarding discriminability, the Hamming distance of the generated watermarks exceeds 88 for images with the same copyright but different content. Furthermore, the efficiency of watermark generation is affirmed, with an average processing time of 96 ms. These experimental results substantiate the superiority of the proposed scheme over existing zero-watermarking methods. |
| first_indexed | 2024-03-08T15:11:11Z |
| format | Article |
| id | doaj.art-9b11d5e9b3f44869b82f8bbb0753f3f6 |
| institution | Directory Open Access Journal |
| issn | 2076-3417 |
| language | English |
| last_indexed | 2024-03-08T15:11:11Z |
| publishDate | 2024-01-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Applied Sciences |
| spelling | doaj.art-9b11d5e9b3f44869b82f8bbb0753f3f62024-01-10T14:52:08ZengMDPI AGApplied Sciences2076-34172024-01-0114143510.3390/app14010435ZWNet: A Deep-Learning-Powered Zero-Watermarking Scheme with High Robustness and Discriminability for ImagesCan Li0Hua Sun1Changhong Wang2Sheng Chen3Xi Liu4Yi Zhang5Na Ren6Deyu Tong7College of Information Engineering, Nanjing University of Finance and Economics, Nanjing 210023, ChinaHunan Engineering Research Center of Geographic Information Security and Application, Changsha 410007, ChinaHunan Engineering Research Center of Geographic Information Security and Application, Changsha 410007, ChinaHunan Engineering Research Center of Geographic Information Security and Application, Changsha 410007, ChinaHunan Engineering Research Center of Geographic Information Security and Application, Changsha 410007, ChinaHunan Engineering Research Center of Geographic Information Security and Application, Changsha 410007, ChinaJiangsu Center for Collaborative Innovation in Geographical Information Resource Development and Application, Nanjing 210023, ChinaCollege of Information Engineering, Nanjing University of Finance and Economics, Nanjing 210023, ChinaIn order to safeguard image copyrights, zero-watermarking technology extracts robust features and generates watermarks without altering the original image. Traditional zero-watermarking methods rely on handcrafted feature descriptors to enhance their performance. With the advancement of deep learning, this paper introduces “ZWNet”, an end-to-end zero-watermarking scheme that obviates the necessity for specialized knowledge in image features and is exclusively composed of artificial neural networks. The architecture of ZWNet synergistically incorporates ConvNeXt and LK-PAN to augment the extraction of local features while accounting for the global context. A key aspect of ZWNet is its watermark block, as the network head part, which fulfills functions such as feature optimization, identifier output, encryption, and copyright fusion. The training strategy addresses the challenge of simultaneously enhancing robustness and discriminability by producing the same identifier for attacked images and distinct identifiers for different images. Experimental validation of ZWNet’s performance has been conducted, demonstrating its robustness with the normalized coefficient of the zero-watermark consistently exceeding 0.97 against rotation, noise, crop, and blur attacks. Regarding discriminability, the Hamming distance of the generated watermarks exceeds 88 for images with the same copyright but different content. Furthermore, the efficiency of watermark generation is affirmed, with an average processing time of 96 ms. These experimental results substantiate the superiority of the proposed scheme over existing zero-watermarking methods.https://www.mdpi.com/2076-3417/14/1/435zero-watermarkingdeep learningrobustnessdiscriminabilityConvNeXtLK-PAN |
| spellingShingle | Can Li Hua Sun Changhong Wang Sheng Chen Xi Liu Yi Zhang Na Ren Deyu Tong ZWNet: A Deep-Learning-Powered Zero-Watermarking Scheme with High Robustness and Discriminability for Images Applied Sciences zero-watermarking deep learning robustness discriminability ConvNeXt LK-PAN |
| title | ZWNet: A Deep-Learning-Powered Zero-Watermarking Scheme with High Robustness and Discriminability for Images |
| title_full | ZWNet: A Deep-Learning-Powered Zero-Watermarking Scheme with High Robustness and Discriminability for Images |
| title_fullStr | ZWNet: A Deep-Learning-Powered Zero-Watermarking Scheme with High Robustness and Discriminability for Images |
| title_full_unstemmed | ZWNet: A Deep-Learning-Powered Zero-Watermarking Scheme with High Robustness and Discriminability for Images |
| title_short | ZWNet: A Deep-Learning-Powered Zero-Watermarking Scheme with High Robustness and Discriminability for Images |
| title_sort | zwnet a deep learning powered zero watermarking scheme with high robustness and discriminability for images |
| topic | zero-watermarking deep learning robustness discriminability ConvNeXt LK-PAN |
| url | https://www.mdpi.com/2076-3417/14/1/435 |
| work_keys_str_mv | AT canli zwnetadeeplearningpoweredzerowatermarkingschemewithhighrobustnessanddiscriminabilityforimages AT huasun zwnetadeeplearningpoweredzerowatermarkingschemewithhighrobustnessanddiscriminabilityforimages AT changhongwang zwnetadeeplearningpoweredzerowatermarkingschemewithhighrobustnessanddiscriminabilityforimages AT shengchen zwnetadeeplearningpoweredzerowatermarkingschemewithhighrobustnessanddiscriminabilityforimages AT xiliu zwnetadeeplearningpoweredzerowatermarkingschemewithhighrobustnessanddiscriminabilityforimages AT yizhang zwnetadeeplearningpoweredzerowatermarkingschemewithhighrobustnessanddiscriminabilityforimages AT naren zwnetadeeplearningpoweredzerowatermarkingschemewithhighrobustnessanddiscriminabilityforimages AT deyutong zwnetadeeplearningpoweredzerowatermarkingschemewithhighrobustnessanddiscriminabilityforimages |