A Secure Learned Image Codec for Authenticity Verification via Self-Destructive Compression
In the era of deepfakes and AI-generated content, digital image manipulation poses significant challenges to image authenticity, creating doubts about the credibility of images. Traditional image forensics techniques often struggle to detect sophisticated tampering, and passive detection approaches...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2025-01-01
|
| Series: | Big Data and Cognitive Computing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2504-2289/9/1/14 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832589147854864384 |
|---|---|
| author | Chen-Hsiu Huang Ja-Ling Wu |
| author_facet | Chen-Hsiu Huang Ja-Ling Wu |
| author_sort | Chen-Hsiu Huang |
| collection | DOAJ |
| description | In the era of deepfakes and AI-generated content, digital image manipulation poses significant challenges to image authenticity, creating doubts about the credibility of images. Traditional image forensics techniques often struggle to detect sophisticated tampering, and passive detection approaches are reactive, verifying authenticity only after counterfeiting occurs. In this paper, we propose a novel full-resolution secure learned image codec (SLIC) designed to proactively prevent image manipulation by creating self-destructive artifacts upon re-compression. Once a sensitive image is encoded using SLIC, any subsequent re-compression or editing attempts will result in visually severe distortions, making the image’s tampering immediately evident. Because the content of an SLIC image is either original or visually damaged after tampering, images encoded with this secure codec hold greater credibility. SLIC leverages adversarial training to fine-tune a learned image codec that introduces out-of-distribution perturbations, ensuring that the first compressed image retains high quality while subsequent re-compressions degrade drastically. We analyze and compare the adversarial effects of various perceptual quality metrics combined with different learned codecs. Our experiments demonstrate that SLIC holds significant promise as a proactive defense strategy against image manipulation, offering a new approach to enhancing image credibility and authenticity in a media landscape increasingly dominated by AI-driven forgeries. |
| format | Article |
| id | doaj-art-d9c8f16efb1a4dbb9faca0df6ed1d017 |
| institution | Kabale University |
| issn | 2504-2289 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Big Data and Cognitive Computing |
| spelling | doaj-art-d9c8f16efb1a4dbb9faca0df6ed1d0172025-01-24T13:22:33ZengMDPI AGBig Data and Cognitive Computing2504-22892025-01-01911410.3390/bdcc9010014A Secure Learned Image Codec for Authenticity Verification via Self-Destructive CompressionChen-Hsiu Huang0Ja-Ling Wu1Department of Computer Science and Information Engineering, National Taiwan University, Taipei 106, TaiwanDepartment of Computer Science and Information Engineering, National Taiwan University, Taipei 106, TaiwanIn the era of deepfakes and AI-generated content, digital image manipulation poses significant challenges to image authenticity, creating doubts about the credibility of images. Traditional image forensics techniques often struggle to detect sophisticated tampering, and passive detection approaches are reactive, verifying authenticity only after counterfeiting occurs. In this paper, we propose a novel full-resolution secure learned image codec (SLIC) designed to proactively prevent image manipulation by creating self-destructive artifacts upon re-compression. Once a sensitive image is encoded using SLIC, any subsequent re-compression or editing attempts will result in visually severe distortions, making the image’s tampering immediately evident. Because the content of an SLIC image is either original or visually damaged after tampering, images encoded with this secure codec hold greater credibility. SLIC leverages adversarial training to fine-tune a learned image codec that introduces out-of-distribution perturbations, ensuring that the first compressed image retains high quality while subsequent re-compressions degrade drastically. We analyze and compare the adversarial effects of various perceptual quality metrics combined with different learned codecs. Our experiments demonstrate that SLIC holds significant promise as a proactive defense strategy against image manipulation, offering a new approach to enhancing image credibility and authenticity in a media landscape increasingly dominated by AI-driven forgeries.https://www.mdpi.com/2504-2289/9/1/14secure learned image codecnon-idempotent codecimage authenticationimage manipulation defense |
| spellingShingle | Chen-Hsiu Huang Ja-Ling Wu A Secure Learned Image Codec for Authenticity Verification via Self-Destructive Compression Big Data and Cognitive Computing secure learned image codec non-idempotent codec image authentication image manipulation defense |
| title | A Secure Learned Image Codec for Authenticity Verification via Self-Destructive Compression |
| title_full | A Secure Learned Image Codec for Authenticity Verification via Self-Destructive Compression |
| title_fullStr | A Secure Learned Image Codec for Authenticity Verification via Self-Destructive Compression |
| title_full_unstemmed | A Secure Learned Image Codec for Authenticity Verification via Self-Destructive Compression |
| title_short | A Secure Learned Image Codec for Authenticity Verification via Self-Destructive Compression |
| title_sort | secure learned image codec for authenticity verification via self destructive compression |
| topic | secure learned image codec non-idempotent codec image authentication image manipulation defense |
| url | https://www.mdpi.com/2504-2289/9/1/14 |
| work_keys_str_mv | AT chenhsiuhuang asecurelearnedimagecodecforauthenticityverificationviaselfdestructivecompression AT jalingwu asecurelearnedimagecodecforauthenticityverificationviaselfdestructivecompression AT chenhsiuhuang securelearnedimagecodecforauthenticityverificationviaselfdestructivecompression AT jalingwu securelearnedimagecodecforauthenticityverificationviaselfdestructivecompression |