Enhancing Steganography in 256×256 Colored Images with U-Net: A Study on PSNR and SSIM Metrics with Variable-Sized Hidden Images
Bibliographic record
Abstract
In digital communications, the imperative for secure data transmission is increasingly addressed through steganography, wherein information is clandestinely embedded within various digital media.This study is concerned with the enhancement of steganographic techniques through a modified U-Net architecture, designed to embed 256×256 colored message images into identically sized cover images, thereby augmenting capacity for data concealment.The classical U-Net architecture has been adapted by the incorporation of batch normalization and residual blocks, aiming to refine the embedding and extraction processes's efficiency.The novel model, trained on the expansive ImageNet database, introduces the one cycle learning rate scheduler and the AdamW optimizer into the U-Net framework, achieving enhanced training efficiency, hastened convergence, and superior generalization.Validation was conducted through two distinct analyses: the first evaluating the impact of secret image size variations on the cover image within the steganographic process, and the second assessing model performance on three datasets-Linnaeus 5, ImageNet, and Labeled Faces in the Wild (LFW).Empirical assessments indicate that the proposed model outperforms existing deep learning-based steganographic methods, as evidenced by the attained metrics, particularly Peak Signal-to-Noise Ratio (PSNR) and Structural Similarity Index Measure (SSIM).On the Linnaeus 5 dataset, embedding yielded a PSNR of 44.4656 dB and an SSIM of 0.9897, while extraction recorded a PSNR of 43.5393 dB and an SSIM of 0.9875.The ImageNet dataset saw an embedding PSNR of 45.3966 dB and an SSIM of 0.9906, with extraction values of 44.8206 dB PSNR and 0.9903 SSIM.Notably, the LFW dataset embedding resulted in a PSNR of 48.1407 dB and an SSIM of 0.9930, and extraction achieved a PSNR of 47.5296 dB and an SSIM of 0.9907.The qualitative and quantitative outcomes affirm the efficacy of the proposed method for the secure transmission of confidential imagery, with potential applications ranging from the safeguarding of medical records to the protection of sensitive data across various digital platforms.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.002 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.001 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".