Boosting up Source Scanner Identification Using Wavelets and Convolutional Neural Networks
Bibliographic record
Abstract
In this paper, we present a conceptually innovative method for source scanner identification (SSI), that is to say, identifying the scanner at the origin of a scanned document. Solutions from literature can distinguish between scanners of different brands and models but fail to differentiate between scanners of the same models. To overcome this issue, the approach we propose takes advantage of a convolutional neural network (CNN) to automatically extract intrinsic scanner features from the distribution of the coefficients of the diagonal high-frequency (HH) sub-band of the discrete stationary wavelet transform (SWT) of scanned images. Such information serves as a reliable characteristic to classify scanners of different/same brands and models. Experiments conducted on a set of 8 scanners yielded a model with an accuracy of 99.31% at the block level and 100% at the full image level, showcasing the potential of using deep learning for SSI and outperforming existing schemes from literature. The influence of the model’s parameters such as the input size, the training data size, the number of layers, and the number of nodes in the fully connected layer as well as the effect of the pre-processing step were investigated.
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) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.001 |
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".