Error correction with PDF improvement
Which errors can PDF improvement in Eye-Able Report fix, and which not yet?
PDF improvement in Eye-Able Report helps you automatically detect and optimise common accessibility issues in PDF documents. However, automatic correction is not yet possible for certain errors.
What can already be improved
PDF improvement automatically detects and improves many common accessibility errors. This allows you to fix numerous issues efficiently, without having to intervene manually.
We can fix the following accessibility errors:
| Test step | Explanation | Improvement |
|---|---|---|
| 5 | The document does not contain a PDF/UA identifier | Setting the PDF/UA identifier in the metadata |
| 7.1.3 | The document is not tagged correctly or not tagged at all | Based on defined rules, we tag the PDF |
| 7.1.8 | Missing metadata in the documen | Setting the metadata (e.g. PDF/UA) |
| 7.1.9 | Missing title in the metadata | Setting the title, you can also define the title yourself |
| 7.1.11 | The document has no structure | Creating a StructTreeRoot entry followed by tagging |
| 7.2.3 | Missing language in the document | Setting the language, you can define the language yourself |
| 7.18.6 | No defined MIME type for embedded media | Sets the MIME type of the media |
| 7.3.3 | Missing alternative text for figures (e.g. graphics) | Automatically adding suitable alternative texts |
| 7.4 | Incorrect order of headings | Adjusting the heading structure |
Why can’t all errors be fixed automatically?
Some accessibility errors require manual rework or a human understanding of the content. There are several reasons for this:
- Missing logic: There is not yet a suitable logic for some error types. We are continuously developing PDF improvement so that more errors can be fixed automatically in future.
- Content or visual context: Certain errors depend heavily on context. In these cases, your manual assessment is required.
- No changes to the visible structure: PDF improvement does not change the visible structure, as automatic adjustments could distort content or layout.
Example: An empty heading is not automatically filled with text.
Manual review recommended
We generally recommend that you also carry out a manual review to determine whether a PDF file is accessible. Even PDF files with a score of 100% may not be accessible under certain circumstances.
Example: Scanned PDF files consist purely of image content. They may meet all test rules but are still not accessible. The content is contained in the image and is therefore not accessible to everyone.
For a more in-depth accessibility review of PDF documents, you can also request manual testing by the Eye-Able Accessibility Testing Team. If you are interested in this kind of testing, feel free to get in touch via our Contact details for customers and interested parties.