pdf1.8 KB

OCR Domain — Utility Bill (Searchable PDF)

A searchable PDF utility bill with selectable text — the OCR ground-truth twin of the scanned JPEG. Fictional SAMPLE content for measuring OCR accuracy.

OCR testing PDF editor testing

Rendered preview of OCR Domain — Utility Bill (Searchable PDF)

Rendered preview of the pdf file (1.8 KB). Download above for the original.

Specifications

Role: searchable text PDF
Domain: utility
Has Text: true
Page Size: US Letter

Testing contract

Expected to pass

Scenario: Exercise OCR Domain — Utility Bill (Searchable PDF) in its ocr domain workflow. A searchable PDF utility bill with selectable text — the OCR ground-truth twin of the scanned JPEG.
Expected result: PDF contains 1 pages; first-page text begins `Brightside Energy — Utility Bill (SAMPLE) Account: ACC-SAMPLE-7788 · Service address: 18 Market S…`. Declared feature checks: role=searchable text PDF; domain=utility; hasText=True; pageSize=US Letter. The declared comparison counterpart is doc-ocr-utility-jpg; preserve the stated difference instead of expecting the container bytes to match.

What is a .pdf file?

PDF (Portable Document Format) is a page-oriented format that fixes layout, fonts, and vector and raster graphics so a page renders identically anywhere. A `%PDF-` header is followed by numbered objects, a cross-reference table mapping each to a byte offset, and a trailer; edits append incremental updates rather than rewrite the file. Page content is a stream of drawing operators, so a PDF holds no words or paragraphs, only positioned glyph runs. Adobe released it in 1993 and gave it to ISO as ISO 32000-1 in 2008.

How to use this file

Use an example PDF to test text extraction, rendering, metadata parsing, AcroForm handling, and OCR pipelines: checking that extraction reconstructs reading order from glyph positions, that a scanned page yields no text, and that an incremental update leaves earlier revisions in the file.

How to use this file for testing

“OCR Domain — Utility Bill (Searchable PDF)” is a deterministic Novus Examples fixture for OCR testing, PDF editor testing. Image-only 'scanned' documents paired with their text source, for measuring OCR accuracy against a known ground truth.

Documented properties for this file: searchable text PDF. Compare results against paired or grouped companions on this page when present (clean↔damaged, searchable↔scanned, or format twins) so scores stay reproducible across runs.

Download the file once, keep the path stable in CI or local scripts, and treat the spec table as the contract: dimensions, seeds, field lists, and roles are intentional. Corrupt or invalid samples are labelled as such, expect parsers to fail loudly rather than silently accept them.

Document fixtures list their internal structure (pages, fields, tracked changes, embedded objects) in the spec table. Test extractors, converters, and OCR against that known structure, and compare searchable↔scanned or format-twin companions when present.

Run OCR on the scanned or image twin and score the output against the searchable or text ground truth on this page; the documented rotation, grain, and text content are the reference.

Code examples

import pdfplumber  # pip install pdfplumber

with pdfplumber.open("utility-searchable.pdf") as pdf:
    print(len(pdf.pages), "pages")
    print(pdf.pages[0].extract_text())

Generated by generation/docs_ocr_wave_b.py. Free for any use, no attribution required, license.