Layout Analysis for Scanned PDF and Transformation to the Structured PDF Suitable for Vocalization and Navigation
Bibliographic record
Abstract
Information can include text, pictures and signatures that can be scanned into a document format, such as the Portable Document Format (PDF), and easily emailed to recipients around the world. Upon the document’s arrival, the receiver can open and view it using a vast array of different PDF viewing applications such as Adobe Reader and Apple Preview. Hence, today the use of the PDF has become pervasive. Since the scanned PDF is an image format, it is inaccessible to assistive technologies such as a screen reader. Therefore, the retrieval of the information needs Optical Character Recognition (OCR). The OCR software scans the scanned PDF file and through text extraction generates an editable text formatted document. This text document can then be edited, formatted, searched and indexed as well as translated or converted to speech. A problem that the OCR software does not solve is the accurate regeneration of the full text layout. This paper presents a technology that addresses this issue by closely preserving the original textual layout of the scanned PDF using the open source document analysis and OCR system (OCRopus) based on geometric layout and positioning information. The main issues considered in this research are the preservation of the correct reading order, and the representation of common logical structured elements such as section headings, line breaks, paragraphs, captions, and sidebars, foot-bars, running headers, embedded images, graphics, tables and mathematical expressions.
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 | 0.006 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.001 |
| Bibliometrics | 0.004 | 0.004 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.003 | 0.002 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.057 | 0.023 |
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".