PoWBWM: Proof of work consensus cryptographic blockchain-based adaptive watermarking system for tamper detection applications
Image tamper detection is a challenging area in multimedia research. The advances in photography technology have made it possible to capture real-time high-dynamic-range (HDR) images through an iPhone or an Android device, which highlights the need for rigorous research on HDR images for tamper dete...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2025-01-01
|
| Series: | Alexandria Engineering Journal |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S1110016824011694 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832583094124675072 |
|---|---|
| author | P. Aberna L. Agilandeeswari |
| author_facet | P. Aberna L. Agilandeeswari |
| author_sort | P. Aberna |
| collection | DOAJ |
| description | Image tamper detection is a challenging area in multimedia research. The advances in photography technology have made it possible to capture real-time high-dynamic-range (HDR) images through an iPhone or an Android device, which highlights the need for rigorous research on HDR images for tamper detection, and localization. The algorithms built for standard images may affect the watermark visibility when it is applied to HDR images as this produces high perceptual variations in the image compared to the original. To tackle this problem, we presented a Proof of Work consensus blockchain watermarking scheme combined with a convolution attention model (PoWBWM) system for tamper detection and localization. The system utilizes a Convolution Attention model (CoAtNet) to generate robust watermarks. A quaternion graph-based transform (QGBT) for embedding, ensuring imperceptibility and robustness. A fuzzy inference system optimizes embedding regions and factors based on human visual system characteristics. The system's security is enhanced through blockchain's proof-of-work (consensus) mechanism, providing a semi-blind watermarking scheme that authenticates ownership and detects tampering efficiently. The security is ensured only when the embedded hash key is authentic with its previous block to proceed further extraction process. The proposed algorithm's performance is evaluated in terms of its visibility by Peak-Signal-to-Noise-Ratio (PSNR), Structural Similarity Index (SSIM), and the perceptual quality of an HDR image is additionally measured by the Visual Dynamic Predictor (VDP) metric. On the other hand, the robustness performance is measured by Normalized Correlation Coefficient (NCC) and Bit Error Rate (BER). The experimental results for CASIA images achieved the highest PSNR value of 63.84 dB, and the SSIM value of 1.000, whereas the maximum VDP value obtained for HDR images is 98.02. In comparison with the existing system, the experimental findings of the suggested model show an effective tamper detection watermarking system as well as a robust against both intentional and unintentional attacks with an average NCC value of 0.98. |
| format | Article |
| id | doaj-art-6e0bd68d84354ad78052af25f2540367 |
| institution | Kabale University |
| issn | 1110-0168 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Alexandria Engineering Journal |
| spelling | doaj-art-6e0bd68d84354ad78052af25f25403672025-01-29T05:00:04ZengElsevierAlexandria Engineering Journal1110-01682025-01-01112510537PoWBWM: Proof of work consensus cryptographic blockchain-based adaptive watermarking system for tamper detection applicationsP. Aberna0L. Agilandeeswari1School of Computer Science Engineering and Information Systems, Vellore Institute of Technology, Vellore, Tamil Nadu, IndiaCorresponding author.; School of Computer Science Engineering and Information Systems, Vellore Institute of Technology, Vellore, Tamil Nadu, IndiaImage tamper detection is a challenging area in multimedia research. The advances in photography technology have made it possible to capture real-time high-dynamic-range (HDR) images through an iPhone or an Android device, which highlights the need for rigorous research on HDR images for tamper detection, and localization. The algorithms built for standard images may affect the watermark visibility when it is applied to HDR images as this produces high perceptual variations in the image compared to the original. To tackle this problem, we presented a Proof of Work consensus blockchain watermarking scheme combined with a convolution attention model (PoWBWM) system for tamper detection and localization. The system utilizes a Convolution Attention model (CoAtNet) to generate robust watermarks. A quaternion graph-based transform (QGBT) for embedding, ensuring imperceptibility and robustness. A fuzzy inference system optimizes embedding regions and factors based on human visual system characteristics. The system's security is enhanced through blockchain's proof-of-work (consensus) mechanism, providing a semi-blind watermarking scheme that authenticates ownership and detects tampering efficiently. The security is ensured only when the embedded hash key is authentic with its previous block to proceed further extraction process. The proposed algorithm's performance is evaluated in terms of its visibility by Peak-Signal-to-Noise-Ratio (PSNR), Structural Similarity Index (SSIM), and the perceptual quality of an HDR image is additionally measured by the Visual Dynamic Predictor (VDP) metric. On the other hand, the robustness performance is measured by Normalized Correlation Coefficient (NCC) and Bit Error Rate (BER). The experimental results for CASIA images achieved the highest PSNR value of 63.84 dB, and the SSIM value of 1.000, whereas the maximum VDP value obtained for HDR images is 98.02. In comparison with the existing system, the experimental findings of the suggested model show an effective tamper detection watermarking system as well as a robust against both intentional and unintentional attacks with an average NCC value of 0.98.http://www.sciencedirect.com/science/article/pii/S1110016824011694Convolution attention modelFuzzy inference systemHigh dynamic range image (HDR)Proof of work consensus blockchainQuaternion graph-based transform (QGBT) |
| spellingShingle | P. Aberna L. Agilandeeswari PoWBWM: Proof of work consensus cryptographic blockchain-based adaptive watermarking system for tamper detection applications Alexandria Engineering Journal Convolution attention model Fuzzy inference system High dynamic range image (HDR) Proof of work consensus blockchain Quaternion graph-based transform (QGBT) |
| title | PoWBWM: Proof of work consensus cryptographic blockchain-based adaptive watermarking system for tamper detection applications |
| title_full | PoWBWM: Proof of work consensus cryptographic blockchain-based adaptive watermarking system for tamper detection applications |
| title_fullStr | PoWBWM: Proof of work consensus cryptographic blockchain-based adaptive watermarking system for tamper detection applications |
| title_full_unstemmed | PoWBWM: Proof of work consensus cryptographic blockchain-based adaptive watermarking system for tamper detection applications |
| title_short | PoWBWM: Proof of work consensus cryptographic blockchain-based adaptive watermarking system for tamper detection applications |
| title_sort | powbwm proof of work consensus cryptographic blockchain based adaptive watermarking system for tamper detection applications |
| topic | Convolution attention model Fuzzy inference system High dynamic range image (HDR) Proof of work consensus blockchain Quaternion graph-based transform (QGBT) |
| url | http://www.sciencedirect.com/science/article/pii/S1110016824011694 |
| work_keys_str_mv | AT paberna powbwmproofofworkconsensuscryptographicblockchainbasedadaptivewatermarkingsystemfortamperdetectionapplications AT lagilandeeswari powbwmproofofworkconsensuscryptographicblockchainbasedadaptivewatermarkingsystemfortamperdetectionapplications |