Accessibility & Tag Check

Achieving digital document accessibility requires adhering to two primary technical frameworks: the Web Content Accessibility Guidelines (WCAG) developed by the W3C, and ISO 14289-1 (PDF/UA, or Universal Accessibility).

While WCAG was originally conceived for HTML web pages, WCAG 2.1 Level AA applies equally to non-web electronic documents. It establishes four foundational principles: content must be Perceivable, Operable, Understandable, and Robust (POUR). This mandates logical reading orders, high color contrast, text alternatives for non-text content, and full keyboard operability.

ISO 14289-1 (PDF/UA) builds upon WCAG, specifying the exact PDF object dictionaries and tag structures required for conforming assistive technology processing. Under PDF/UA, every piece of meaningful content must be categorized within a semantic structural tree, artifacts (decorative elements) must be explicitly flagged, and font encodings must map deterministically to Unicode.

The cornerstone of an accessible PDF is its logical structure tree, rooted in the document catalog's `/StructTreeRoot` dictionary. Without a structural tree, a PDF is merely a collection of graphical coordinates on a canvas; screen readers have no way of knowing whether a block of large text is a chapter heading, a table cell, or an incidental pull quote.

Our checker inspects the `/MarkInfo` dictionary to verify that the `/Marked` flag is set to `true`, indicating that the document contains tagged content. It then traverses the structural tree hierarchy, evaluating standard structure type tags: `<Document>`, `<Part>`, `<Art>`, `<Sect>`, `<H1>` through `<H6>`, `<P>`, `<L>` (lists), and `<Table>`. The audit checks for heading nesting errors (such as skipping directly from `<H1>` to `<H4>`), ensuring that assistive tech users can navigate documents logically.

Images, technical diagrams, charts, infographics, and corporate logos present major barriers to visually impaired users if alternative descriptions are missing. Under WCAG Guideline 1.1.1, all non-text content must have a text alternative that serves the equivalent purpose.

In PDF architecture, images are defined as `/XObject` with `/Subtype /Image` or as inline image operators (`BI ... EI`). When tagged, an image structure element (`<Figure>`) should contain an `/Alt` string dictionary entry providing a concise, descriptive explanation of the visual asset. Our accessibility engine scans every figure tag across the document, detecting missing `/Alt` attributes, flagging placeholder alt text (like 'image.png' or 'photo'), and checking that purely decorative graphics (such as background flourishes and border lines) are properly categorized as `/Artifact` to prevent screen reader clutter.

Beyond tagging and images, our audit checks several critical infrastructural requirements that dictate real-world assistive reading success:

Document Language: Under WCAG 3.1.1, the natural language of the document must be defined via the `/Lang` key in the document catalog (e.g., `en-US` or `fr-FR`). Without a valid language code, screen readers cannot select the correct pronunciation dictionary, resulting in garbled audio output.

Bookmarks and Outlines: For documents exceeding 20 pages, ISO 14289-1 mandates an outline hierarchy (`/Outlines` dictionary) that mirrors the document's heading structure, enabling rapid jumping between sections.

Table Headers: Data tables must utilize `<Table>`, `<TR>`, `<TH>` (table headers), and `<TD>` (table data) tags with appropriate `/Scope` attributes so screen readers can announce column and row context as users navigate cells.

Accessibility audits are frequently conducted on pre-publication corporate annual reports, pending legislation, proprietary training manuals, and sensitive student educational records. Uploading these unreleased materials to cloud-based audit tools risks premature leakage and intellectual property compromise.

Our PDF Accessibility Checker executes 100% locally within your web browser. Utilizing modern WebAssembly and Web Worker threads, the engine parses the raw PDF object graph directly in browser RAM. Your files never leave your device, ensuring complete data security while providing enterprise-grade compliance auditing.

Under WCAG 2.1 Success Criterion 1.4.3 (Contrast Minimum), text and images of text must have a contrast ratio of at least 4.5:1 against their background (or 3:1 for large-scale text exceeding 18pt regular or 14pt bold). Furthermore, Criterion 1.4.11 mandates 3:1 contrast for user interface components and graphical objects.

Our accessibility engine evaluates relative luminance using the standard W3C formula: L = 0.2126 * R + 0.7152 * G + 0.0722 * B where color channels are linearized from sRGB space. The engine scans text operators against intersecting background vector fills and raster graphics, highlighting low-contrast paragraphs that pose reading difficulties for users with low vision or color blindness.